Home

Big Energy Upgrade Computer Science Final Report

1. In order to provide easy access to the data both within WP3 and the wider consortium a data access server was designed and implemented This is a web based server providing search browse and download facilities to the underlying consumption data Work focussed on providing access to daily consumption totals as well as access to each day s fine grained consumption traces The nomenclature for data available from the server has been formalised into three core types 1 feed 2 aggregates 3 now Fine grained data data points approximately every 6 seconds for electricity and approximately every 27 seconds for gas is referred to as feed data Aggregated data accumulated electricity or gas consumption per day is referred to as aggregates data The current consumption with the current day s electricity statistics is referred to as now data For both feed and aggregates data it is possible to specify time date ranges in order to obtain consumption data for specific time periods as opposed to obtaining either whole days or the entire dataset For both feed and aggregates data it is also possible to specify electricity only gas only or combined results In addition statistics min max mean can be obtained for those time periods In addition it is possible to obtain a week based analysis of energy consumption This covers the analysis of the immediately preceding seven days total electricity and gas consumption per day and the overall
2. 50 Viglen MPC Viglen approx 75 3G mobile broadband dongle O2 approx 25 10 pm Pulse blocks range of prices dependent on supplier and gas meter type o Meter Provida http meterprovida com metering accessories pulse blocks and brackets o Utility Meters Warehouse http www utilitymeterswarehouse com gas meters gas accessories pulse blocks html o The Code Store http www thecodestore co uk index php option2com virtuemart amp page shop product details amp flypag e flypage tpl amp ltemid2204 amp category id 481 amp product id 1299 amp vmcchk 1 amp ltemid 204 96 The University Of Sheffield sf X Q que PLU Doncaster LS Kirklees Metropolitan Borough Council HOMES Metropolitan Borough Council BARNSLEY west g Kirklees tial Neighbourhood D ec N noru cov Y LINCOLNSHIRE Housing meee CITY COUNCIL wes COUNCIL COUNCIL homes leeds www nelincs gov uk www northlincs gov uk The EUROPEAN UNION Ge University 24 Yorkshire Investing in Your Future North Lincolnshire cm LU a 7 P HOMES p Ny Of TASS Energy Services European Regional on Sheffield Development Fund 2007 13 Ag NY Y NYY
3. www google com powermeter about BBC News South Yorkshire s Digital Region broadband scheme to close http www bbc co uk news uk england south yorkshire 23 713493 residents This work is described in Chapter 3 Consumption information is provided at a range of granularities and allows comparison between different points in time Although it was originally envisaged to provide participants with the opportunity of comparing their consumption with other families living in similar households data privacy and ethical concerns meant that individual households were restricted to viewing only their own consumption patterns The households participating in the project are likely to have low incomes and therefore use less energy and produce less CO than households with higher income Hirsch et al 2011 Reinders et al 2003 Wier et al 2001 However such households tend to devote a greater share of their spending to fuel and water than higher income households In 2009 the poorest 10 per cent of households i e the lowest income decile spent on average over twice as much of their budget on fuel compared with the highest income decile 7 796 and 3 496 respectively Baker 2011 Therefore these households are at a greater risk of falling into fuel poverty although the exact definition of which households are in fuel poverty is under current debate the general concept is defined as a function of the amounts by which the assessed energy
4. Systems to BIG Energy Upgrade Requirements Diog NB RID www greenenergyoptions co uk British Gas 2 1 1 Electricity Monitoring onzo com www britishgas co uk Ewgeco O EUN AM the publicly available electricity consumption smart meters use the same basic technology A sensor clips on to the mains power cable connected to the electricity meter this monitors the instantaneous electrical current amperes passing through the cable via the magnetic field generated around the power cable The sensor is plugged into a transmitter which sends the readings wirelessly to the display unit see Figure 2 Figure 2 Electricity monitoring clamp and transmitter fitted in a domestic electricity meter box 2 1 2 Gas Monitoring The monitoring of gas consumption is more problematic due to this domestic smart metering systems are almost universally aimed at solely monitoring electricity The possible approaches to monitoring gas consumption depend upon the gas meter type used e Pulse Enabled Meters these devices produce an electric pulse therefore simply require the connection of a data logger or wireless transmitter e Pulse Ready Meters these meters generate a magnetic pulse which requires converting into an electric pulse via the use of a Hall Effect Sensor or Ree
5. average consumption per week day since the initial monitoring equipment installation This functionality allows the data visualisation web app to provide comparative consumption plots to each household see below Data can be accessed via a REST application programming interface API allowing clients to request either live data feeds aggregated consumption totals or statistics over arbitrary time spans https www energyuse org uk In addition weather information for each household s location can be requested All data can be accessed for all households with live broadband connections The format of the data can be specified to ensure it most closely matches its subsequent use For instance data be requested in JSON format which is a text based open standard designed for human readable data interchange The JSON format is often used for serializing and transmitting structured data over a network connection It is used primarily to transmit data between a server and web application serving as an alternative http en wikipedia org wiki Web_ service http en wikipedia org wiki JSON 29 to XML Alternatively data can be requested in CSV format to allow easy importing into Microsoft Excel or other analysis software Documentation for the API has been created Considerable effort was put into ensuring the security and privacy of the data being captured stored and made available Although individuals cannot be identifie
6. into the data access server In a similar fashion to the energy consumption data weather data is available to two levels of granularity feeds and aggregates Weather feeds provide temperature and relative humidity data approximately every 30 minutes whereas aggregates data provide daily average weather data This data will also allow residents to compare how their energy consumption changed in relation to the external temperature Historical weather data for all existing households as far back as the first energy data point for that household has been included in the database 3 2 2 Costings Following feedback from the partners at a joint WPI WP8 workshop the ability to store energy unit costings was included For electricity this involved the inclusion of per household recording of energy tariff including the ability to record differing tariffs for daytime and night time charging For gas this involved recording the cost per kWh as provided by the supplier together with the calorific value of the gas this can vary by geographical area for example this varies between 37 5 and 43 0 for the UK with the 25 exception of Stornoway which uses LPG and has a much higher calorific value of around 93 and the correction factor to take into account the expansion contraction of the gas supplied due to the ambient temperature Furthermore since homeowners can change tariff or change supplier at anytime all costings have to be time stamped
7. needs of fuel poor households exceed the threshold for reasonable costs Hills 2012 While income is the primary determinant of energy consumption and carbon emissions other factors such as the type of dwelling tenure household composition and rural urban location are also extremely important Druckman and Jackson 2008 some of which as addressed in other areas of BIG Energy Upgrade project s work The remainder of this document focuses on the monitoring technology deployed the feedback approaches adopted and our experiences of participant recruitment Section 2 address monitoring equipment by investigating the state of the art for both electricity and gas monitoring devices Section 2 1 the technology used in phase 1 Section 2 2 and phase 2 Section 2 3 and finally discusses equipment for standalone deployment of monitoring equipment Section 2 4 and the support for partner based monitoring Section 2 5 Section 3 describes the resident feedback technologies by providing an overview of the state of the art Section 3 1 the supporting technology for storing and analysing the consumption data Section 3 2 a standards compliant approach for accessing the data Section 3 3 and finally a user friendly web based application for exploring a resident s energy consumption behaviour Section 3 4 Section 4 details our experiences related to the participant recruitment process and Section 5 concludes with a number of key findings and
8. on the right vertical axes Figure 21 Historical Energy Consumption and External Temperature Zoom 1m 3m 6m YTD ly All From Jan 22 2013 To Jul 22 2013 S e a I A E uM uondunsuo A313032813 sayaw 231qn2 uorjduinsuo Highcharts com Figure 21 Full historical consumption plot showing electricity consumption gas consumption and external temperature Just as for the whole day consumption plots there are a number of preset zooms that allow the resident to quickly zoom into either 1 month 3 month 6 month year to date YTD full year or all views The 6m zoom returns to the usual zoom level showing the previous 6 months consumption information At any level of zoom the resident can move 39 around simply by dragging the zoom area scroll bar in the same way as described in Section 3 4 3 above To zoom into a specific date range the resident can also use the From and To boxes at the top right of the plot area to specify your range Figure 22 2013 03 07 To Jul 22 2013 Figure 22 Date range zoom In addition to viewing the consumption information in kilowatt hours kWh for electricity and cubic metres for gas they can also view it in Pounds to show how much their energy consumption is costing The resident can change the displayed units using the selection menu at the top of the page Figure 23 Energy units Raw units Cost Figure 23 Energy consumption unit
9. on when they need to use the Internet Therefore it is not possible to rely on an Internet connection with constant availability Some monitoring devices such as the ISA iMeter have no on board storage facility and thus when the Internet connection fails all data is lost until the connection is restored In the case of the Current Cost EnviR device a limited degree of logging on the RTD is available but of insufficient granularity for the purposes of the BIG Energy Project requirements the EnviR s native historic readings are aggregated which means there will be some information loss if the data is not transmitted before aggregation To address this issue ways of storing the consumption data locally i e in the participant s house until an Internet connection was available were investigated The Current Cost EnviR provides data access via a USB cable which can be used to download consumption data to a Windows or Linux PC In light of this it was decided to use a small low power mini PC a Viglen MPC L running Linux This device provides 80Gb of storage sufficient to capture many years of energy consumption data 4 USB ports capable of connecting to the EnviR and Ethernet connectivity to connect via the participants router to the Internet The Viglen mini PC is very small about the size of 3 stacked CD cases and consumes very little energy costing about 1 per year to run when switched on continuously A suite of custom made sof
10. or the presence of PV solar panels A minor issue identified with the Viglen PC used for data logging and transmission is its power management The power button on the front is susceptible to being accidentally knocked thus initiating a shutdown Furthermore if power is lost home owner accidentally unplugs the device or there is a power cut the device will not automatically restart when power is restored Although homeowners are told about this they occasionally forget to restart the device in these circumstances Recommendation 8 Use a logging and transmission device which switches on as soon as power is available A device such as the Raspberry Pi see Section 2 4 exhibits this behaviour and has been shown to be compatible with all the software developed for the BIG Energy Upgrade project and has very low power consumption 5 2 Feedback Development of the resident feedback technology see Section 3 and associated data access proceeded well and involved feedback from end users partners and other members of the Universitys BIG Energy Upgrade team in particular findings from the Psychology research team Effort was made to ensure the underlying data storage and access technology was secure and conformed to current standards e g the REST web architectural style The data access server as been designed to allow easy access to the data by members of the BlG Energy Upgrade team within the University but also to wider consortium members Ind
11. timer based circuit based around a 555 timer integrated circuit IC This IC is used in a variety of timer pulse generation and oscillator applications and can be used for example to provide time delays Although introduced forty years ago the 555 is still in widespread use thanks to its ease of use low price and good stability 17 A circuit was designed in which the 555 timer operated in monostable mode providing the required one shot pulse generator behaviour The pulse begins when the 555 timer receives a signal at the trigger input the closing of the reed switch the duration of the output pulse is determined by the combination of resistor and capacitor The final circuit design and test implementation are shown in Figure 3 555 pin out 1 Ground 2 Trigger 3 Output 4 Reset 5 Control not used 6 Threshold 7 Discharge 8 Vcc Figure 3 Version 2 of the gas monitoring circuit left and the prototype implementation right In our circuit design the LED light source has been configured to illuminate for 0 3 seconds determined by the 2 7k resistor and the 100uF capacitor The inclusion of the left most 1Ok resistor and 1 capacitor ensures that the circuit only illuminates the LED light source for 0 3 seconds regardless of how long or how briefly the reed switch is activated for The pulse block reed switch is denoted by the push switch symbol in the lower left of the circuit diagram
12. to navigate the web application and display any of the visualisations There is a wider movement toward citizen engagement and this could be achieved within the energy sector by providing easy access to energy consumption data The approach adopted in this report specifically in Sections 3 3 and 3 4 should be considered the starting point for providing standardised easy to use access to consumption data regardless of energy supplier Recommendation 11 majority of consumer grade off the shelf monitoring kits require the use of proprietary software or web applications Effort should be made to always provide the consumer with access to their raw underlying consumption data 5 3 Participant Recruitment and Interaction Recruitment of residents into the energy monitoring initiative was a significant challenge at all stages of the project Four key factors were identified 1 lack of interest 2 desire not to have equipment in their home 3 perceived lack of direct benefit to the resident despite explanations of the feedback information which would be made available and the potential to make savings in energy consumption and hence bills 4 unsuitability of meters or infrastructure The second issue equipment left in the property could be addressed by Recommendation 6 and issue four by Recommendation 2 However the remaining issues are related to the motivation of the potential participants 49 Recommendation 12 To maximise
13. uptake of monitoring in similar schemes and to ensure pre and post intervention data exists the installation of energy consumption monitoring equipment should be made a condition of the energy saving intervention offering Recommendation 13 As an alternative to Recommendation 12 suitable incentives ought to be offered Such incentives could include prize draws vouchers tax rebates etc Recommendation 14 Where monitoring equipment installation isnt a pre requisite of the intervention a combination of recruitment approaches ought to be pursued phone house to house word of mouth and presence at community events organised by the local authority or housing association The latter would improve the conversion rate since residents attending energy related community events are more likely to be motivated to participate Raw and Ross 2011 note that all studies highlighted a difficulty in initially contacting customers For example EDF undertook its recruitment by telephone and it appears that 2596 of its calls were to unobtainable numbers A further 1096 of calls were to numbers that did not accept number withheld calls customers who had moved address or customers who had insufficient English Similar experiences were noted in the BIG Energy Upgrade project Recommendation 15 Updates to resident contact information should be shared across local authorities housing associations and or centralised Recommendation 16 Project publicity
14. was households which are not connected to the Internet This has been called the problem of the digital divide referring to the proportion of people who are not connected to the Internet and therefore cannot access the digital resources The use of the Viglen MPC L offers the possibility of utilising mobile broadband to connect these household although obviously reliant on the quality of the mobile signal Although technologically possible see Section 2 3 3 wide scale deployment of the monitoring equipment using mobile 3G based broadband connectivity was deemed prohibitively expensive and was only used for two trial installations 45 Another issue concerned the availability of Ethernet sockets on the homeowners broadband router In a number of cases it was found that all sockets were in use thus preventing the connection of the Viglen logging and data transmission PC to the homeowners network and thus access to the Internet The provision of a low cost Ethernet switch Ethernet hub would solve this issue This unfortunately increases the cost and complexity of the installation i e increasing the potential points of failure and adds another piece of electrical hardware to the installation increasing the running cost However this is still cheaper than the alternative of using mobile connectivity 46 5 Key Findings and Recommendations 5 1 Monitoring Technology There are a large variety of gas meter models in use in domestic prop
15. will go a paler shade of red 36 Daily Electricity Consumption W Zoom 1hr 3hr hr All From Jul 22 2013 To Jul 22 2013 Figure 16 Energy consumption display showing zoomed region top and context bottom Figure 16 shows how the zoom area has been used to zoom into the time period from 1 34am to 8 19am to focus on the first electricity consumption spike Note how the upper area consumption area only shows the electricity consumption for this time period while the bottom area zoom area shows the whole day s consumption with the periods outside the current zoom being paler This zoom window can be moved around simply by dragging the zoom area scroll bar Figure 17 Zoom area scroll bar Figure 17 Zoom movement control The zoom can be changed again by either moving the zoom blocks or using the zoom preset buttons At any level of zoom the resident can explore the consumption area in more detail by moving the mouse over the plot area see Figure 18 This will cause a balloon popup to appear which updates continuously as the resident moves the mouse It shows for the current position of the mouse cursor the exact time and the precise electricity consumption at that point Monday Jul 22 07 53 Consumption W 311 07 50 Figure 18 Information for any particular datapoint can be shown by moving the mouse to the specific location of interest The gas consumption plot beh
16. However it does include houses which were monitored during phase 1 but found to be incompatible for monitoring in phase 2 41 Phase 1 D Sept 2012 Electricity Sept 2012 0 Dec 2012 O _ Dec 2012 O _ Oct 2013 O _ Oct 2013 O _ dan 2013 O EUN 0 0 0 0D Table 6 Overview of collected data Phase 1 and phase 2 indicate which phase the property was monitored under and the date of installation Initial letter s of the ID refer to the geographical area D Doncaster L Leeds NL North Lincolnshire B Barnsley K Kirklees missing numbers indicate households which chose to withdraw during the project Where properties are listed as live this data is transmitted to the University s data servers using broadband In some circumstances the connection may fail in which case the data is logged locally on the residents monitoring equipment and is sent once a connection is re established or can be manually retrieved by visiting the site Taking D13 as an example monitorin
17. TORING PHASE 1 INSTALLATIONS PHASE 2 INSTALLATIONS 2 3 1 LOGGING AND COMMUNICATIONS 2 3 2 GAS MONITORING 2 3 3 INDEPENDENT CONNECTIVITY 2 3 4 SUMMARY OF EQUIPMENT STAND ALONE EQUIPMENT PARTNER BASED MONITORING RESIDENT FEEDBACK TECHNOLOGY 3 1 3 2 3 3 3 4 3 5 STATE OF THE ART 3 1 1 REAL TIME DISPLAY DATA ANALYSIS AND STORAGE SERVER 3 2 1 WEATHER DATA 3 2 2 COSTINGS 3 2 3 IMPLEMENTATION API WEB BASED CONSUMPTION BROWSER 3 4 1 HOMEPAGE 3 4 2 NAVIGATION 3 4 3 TODAY S CONSUMPTION 3 4 4 THIS WEEK S CONSUMPTION 3 4 5 FULL HISTORY 3 4 6 USER MANUAL DATA SNAPSHOT MONITORING RECRUITMENT EXPERIENCES 4 1 RECRUITMENT A 10 11 12 13 13 14 15 21 22 22 23 24 24 25 25 25 25 26 29 30 32 34 34 38 39 41 41 44 44 4 2 LOGISTICAL ISSUES 45 5 KEY FINDINGS AND RECOMMENDATIONS 47 5 1 MONITORING TECHNOLOGY 47 5 2 FEEDBACK 48 5 3 PARTICIPANT RECRUITMENT AND INTERACTION 49 6 REFERENCES 52 7 APPENDIX HOUSE SELECTION DOCUMENTATION FOR PARTNER BASED ENERGY CONSUMPTION MONITORING 54 1 Introduction The increasing pressure to make rapid and substantial reductions in greenhouse gas emissions in order to mitigate dangerous climate change see IPCC 2007 UNFCCC 2009 requires considerable changes to the way that energy is generated and used These changes will depend in part on shifts in the way that the general public think about and use energy i
18. The circuitry was tested successfully for instantaneous behaviour on a Sensus Cubix U6 with a METUROOS pulse block and a Schlumberger R5 with integrated RJT1 output The test implementation was also installed at the test house Regular monitoring of the test house installation data feed showed gas monitoring stopped after 4 days It was determined that the circuitry had drained the battery much more quickly than anticipated In order to resolve this issue a number of options were investigated use of a mains power adapter multiple batteries in parallel use of a photocell However these solutions were decided to be infeasible or not value for money The use of a low power version of the 555 timer IC was also investigated These ICs are based on CMOS technology as opposed to transistor transistor logic TTL and draw less current A number of CMOS based 555 chips were acquired for testing and a new version of the gas monitoring circuit was produced An additional benefit of using the CMOS based chips is their ability to withstand sub zero temperatures a situation which is likely to occur given most gas meters are externally situated The CMOS based circuit has been installed at the test house and analysis of the data feeds was conducted It was found that despite version 2 of the circuit design above using a low power CMOS version of the 555 timer the battery power supply still only lasted 9 5 days much shorter than anticipated The circuit wa
19. Work Package 3 Energy Consumption Monitoring and Feedback 30 25 ar A 20 uorduinsuo set 0 EUROPEAN UNION Investing in Your Future European Regional Development Fund 2007 13 Authors Dr Stuart N Wrigley Professor Fabio Ciravegna Neil Ireson Dr Simon Tucker Publication Date February 2014 Publisher University of Sheffield Copyright O 2013 2014 The University of Sheffield All rights reserved No part of this report may be reproduced adapted stored in a retrieval system or transmitted by any means including photocopying recording or other electronic or mechanical methods without the prior written permission of the publisher For permission request please contact Professor Fabio Ciravegna f ciravegna gsheffield ac uk Professor Fabio Ciravegna Department of Computer Science University of Sheffield 211 Portobello Sheffield S1 4DP United Kingdom The EUROPEAN UNION University Investing in Your Future 0 European Regional Sheffield Development Fund 2007 13 BIG Energy Upgrade Energy Consumption Monitoring and Feedback Dr Stuart N Wrigley Professor Fabio Ciravegna Neil Ireson Dr Simon Tucker Acknowledgements This work has been undertaken by the University of Sheffield as a part of the Work Package 3 of the BIG Energy Upgrade programme aka Energy Innovation for Deprived Communities The BIG Energy Upgrade is a flag
20. and recruitment material ought to be available in a range of languages During the equipment replacement period in which phase 1 equipment ISA was replaced with phase 2 equipment EnviR there were a number of issues related to the gaining of access to the installed equipment Issues encountered included e difficulties in arranging appointments to visit the properties e no one being present in the home at the arranged time e person present in the home not being the home owner tenant e hostile behaviour In general these issues are expected to be common to any project of this nature in which equipment must be installed and left in a property for some time especially where motivation is low and or incentives are not used and interaction with the research team is infrequent Recommendation 17 Close collaboration with the local authority housing association when contacting and visiting residents will allow their existing relationship with the community to be leveraged On one occasion during the equipment replacement period a property was visited and after being invited into the house the resident presented all of the installed equipment energy monitoring equipment as well as air quality monitoring devices sealed inside a 90 plastic carrier bag It was unclear how long the devices had been sealed in this way The resident explained that they were unhappy with a number of aspects of the scheme under which their energy saving inte
21. appeared to incorporate leading edge pulse detection since a single pulse was generated for sustained LED activation Since there are many different manufacturers and models of gas meter the designs of the gas meter dials and magnetic pulse configuration vary significantly It had initially been envisaged to use standard reed switches and these were used in our initial testing However since detailed design documentation for gas meters is not always available it was difficult to assess the required sensitivity of the reed switch to ensure reliable activation by the meter s magnet Therefore the decision was Sit taken to use commercially available pulse blocks examples of which are shown to the right These are professionally produced units in which ACTARIS each unit is designed to be fitted to a specific model of gas meter and incorporate one or more reed switches which are positioned for optimal reed switch activation In addition the reed switches themselves are properly matched with respect to sensitivity to the design of the meters dials and strength of the meters magnet Pulse blocks are available for commonly used gas meters e g http meterprovida com metering accessories pulse blocks and brackets Pulse block connectivity is via a RJf1 socket The magnetic optical conversion circuit produced by the University of Sheffield connects to the pulse block via an lead In addition some meters such as ones made
22. aves in exactly the same way as the electricity consumption plot 3 4 4 This week s consumption This section of the website shows the current week s daily consumption totals relative to the residents home s average consumption In this section there are three plots the weekly consumption plot Figure 19 followed by two daily consumption plots one for electricity and the other for gas The two daily consumption plots behave in exactly the same way as for those described in Section 3 4 3 above In contrast to the daily plots the weekly plot shows the full day consumption total for each day of the week up to and including today This is measured in kilowatt hours kWh for electricity and cubic metres for gas the same units used in supplier bills Weekly Energy Consumption e saaw 31qn3 uorydunsuo sec S a E a 2 10 ao 0 i 0 e yesterday today tuesday wednesday thursday friday saturday Average electricity kWh ll This week s electricity kWh Average gas cubic metres Ill This week s gas cubic metres Figure 19 Week consumption comparison plot In addition to viewing the consumption information in kilowatt hours kWh for electricity and cubic metres for gas the resident can also view it in Pounds GBP to show you how much their energy consumption is costing The resident can change the displayed units using the selectio
23. b T L Benn Y and Chang B P submitted Antecedents and Consequences of 21 Monitoring Domestic Energy Consumption Journal of Environmental Psychology Wier M Lenzen M Munksgaard J and Smed S 2001 Effects of household consumption patterns on co requirements Economic Systems Research 13 3 259 274 2001 22 Yohanis Y G 2012 Domestic energy use and householders energy behaviour Energy Policy 41 654 665 February 2012 93 7 Appendix House selection documentation for partner based energy consumption monitoring 94 ERDF BIG Energy Upgrade EIDC Project House selection for partner based energy consumption monitoring Development and testing of the monitoring solution is still on going We have a stable investing in Your Future configuration for monitoring electricity consumption However due to the wide variability in gas meters we are still in the testing phase for gas consumption monitoring European Regional Development Fund 2007 13 In light of the current status we advise partners to select electricity only houses where possible since this will simplify the installation process and increase its reliability Therefore households should conform to the following criteria e Electricity only e Broadband available for immediate use e No renewable energy generation such as photovoltaic cells or wind generation since this is not compatible with the energy monitoring equipmen
24. bsolute temperature in P http www wunderground com weather api 33 other words the in the shade temperature to the feels like temperature which incorporates other factors such as wind chill Figure 10 Live external temperature 9 Absolute Feels like Figure 10 Temperature unit choice interface The temperature is shown in Celsius The website supports a number of different roles each of which provides differing degrees of access to the underlying data For instance an administrator has access to all data from all households with data stored in the database Similarly a researcher i e a member of the BIG Energy Upgrade team has access to all data In these cases the user is presented with an additional control to allow the selection of the household of interest Household Figure 11 Household selection control only visible to administrators and researchers In contrast a householder can only access their own property s energy consumption data and so the household selection menu is not shown 3 4 2 Navigation At the top of the page underneath the title banner is the navigation bar which allows the resident to show different views of their energy consumption Figure 12 BEU Energy Consumption Browser Show Now Today s consumption This week s consumption Explore full history Figure 12 Website navigation bar The default page is the current Now consumption The navi
25. by Schlumberger have integrated pulse blocks these meters provide connectivity directly and are thus compatible with our magnetic optical conversion circuit 16 Suitable pulse ready or pulse enabled meters can be identified via the face plate surrounding the dials There should be text similar to Pulse 0 01m Instead of Pulse the meter may use Impulse Imp or similar A pulse ready meter will have a RJ11 socket small telephone type socket on the front or underneath near the dials A pulse enabled meter will require a pulse block Compatible gas meters include but are not limited to Actaris G4 or U6 gas meters Pulse block required Actaris SC6 TVI Pulse block required Actaris R5 RJTI socket integrated no pulse block required Elster BK series Pulse block required Krom Schroder BK G4 Pulse block required Itron Pulse block required Sensus U6 Pulse block required Schlumberger G4 or G6 Pulse block required Schlumberger R5 RJTI socket integrated no pulse block required UGI Limited R5 socket integrated no pulse block required Wilson G4 meters RJ11 socket integrated no pulse block required Manufacturers do not necessarily differentiate within models for those which are pulse enabled and those which are not For example some Sensus Cubix U6 meters are pulse compatible with a pulse block whereas other older Sensus Cubix U6 meters cannot provide a pulse Each individua
26. ceipt without the requirement for human intervention The process is as follows 1 Raw consumption data from a remote monitor arrives and is stored in an incoming data directory a heremote monitor is responsible for the transmission of data b Data from each remote monitor is transferred to the data server every 5 minutes 2 raw consumption data for each monitor is extracted and cleansed a It is common for the Current Cost meter to occasionally transmit anomalous data This can be detected and removed b The Current Cost meter also transmits periodical history data which is not required this also removed Consumption data for electricity is distinguished from gas consumption data d The raw consumption data is processed every 10 minutes 26 3 The raw consumption data files are moved to long term storage archive backup 4 The consumption data for electricity and gas is stored in the database a Each datapoint consists of a time stamp and a consumption value b Each monitor generates an electricity consumption datapoint approximately every 6 seconds approx 14 400 datapoints per day per monitor for electricity alone and a gas consumption datapoint approximately eveny 29 seconds approx 2 979 datapoints per day per monitor for gas alone 5 The overall daily consumption total for each monitor is updated in the database 21 Figure 5 Database schema for the energy consumption data 28 9 9 API
27. choice control At any level of zoom the resident can explore the consumption information in more detail by moving the mouse over the plot area This will cause a balloon popup to appear which updates continuously as they move the mouse It shows for the current position of the mouse cursor the date external temperature and the precise electricity and gas consumption at that point Figure 24 40 Monday May 27 2013 External Temperature C 12 0 Electricity Consumption kWh 9 35 Gas Consumption cubic metres 2 02 Figure 24 Using the mouse over to show detailed information for a particular datapoint By clicking on either the temperature electricity or gas consumption lines the resident can show that day s full consumption view in the plots underneath The daily plots of historical data can be zoomed and explored in exactly the same was as described in Section 3 4 3 above 3 4 6 User Manual A full manual based on Sections 3 4 1 3 4 5 describing how to use the website was created and distributed to participants 3 5 Data snapshot Over the course of the BIG Energy Upgrade monitoring equipment has been installed in a number of properties The duration type and amount of data collected for each property is indicated by Table 6 Note that Table 6 does not include properties for which the resident requested to cease participation earlier than expected 4 in Doncaster 1 in Leeds 1 in NE Lincs and 1 in Barnsley
28. choices pdf IPCC 2007 International Panel on Climate Change Climate Change 2007 Synthesis Report Contribution of Working Groups I Il and lll to the Fourth Assessment Geneva Switzerland IPCC 02 14 15 16 1 18 19 UNFCCC 2009 United Nations Framework Convention on Climate Change Report of the conference of the parties on its fifteenth session held in Copenhagen from 7 to 19 December 2009 Addendum Part Two Action taken by the conference of the parties at its fifteenth session Petersen D 2009 WattBot A Residential Electricity Monitoring and Feedback System Proceedings of CHI Boston MA 2847 2852 Pierce J Odom W amp Blevis E 2008 Energy Aware Dwelling A Critical Survey of Interaction Design for Eco Visualizations Proceeedings of OzCHI Cairns QLD Australia Raw G and Ross D 2011 Energy demand research project Final analysis Technical report Ofgem June 2011 http www ofgem gov uk sustainability edrp Documentsli Energy9620Demand9620Res earch9620Project9620Final9620Analysis pdf Reinders A H M E Vringer K and Blok K 2003 The direct and indirect energy requirement of households in the European Union Energy Policy 31 2 139 153 2003 Ueno T 2005 Effectiveness of displaying energy consumption data in residential houses Analysis on how the residents respond Proceedings of ECEEE Summer Study What works amp who delivers 1289 1299 20 Web
29. d Switch the latter encapsulated into a pulse block e Meters with Dials for meters with spinning dials it is possible to count the rotation of the lowest unit dial This can be done using a photo reflector which transmits an LED light and senses the reflection Some meters have a small reflective disc on the dial There are a number of issues with this approach it will require covering the dial with a fairly robust framework so that the sensor does not move limiting the ability to take a visual reading and It is sensitive to light noise e Meters with Numeric Index Finally if no pulse can be acquired from the meter it is possible to use Optical Character Recognition OCR technology to process the image of the display to calculate the meter reading In general the commercial 1 versions of such technology is expensive e g around 300 for the Xemtec system10 however it does provide a fairly universal way of reading any meter A further issue encountered in the project was the use of Prepayment or Credit Meters Such meters utilise the magnetic signal to calculate consumption and so the signal is unavailable for the smart meter One somewhat radical solution to this issue was to install a secondary meter to provide another pulse However compared to the cost and technical complexity of other solutions i e using OCR secondary meters are more economical and straightforward As of Spring Summer 2012 there was no reliable so
30. d directly from any data stored or accessible from the database information about participant behaviour can be inferred from their energy consumption patterns As such access to the data has to be carefully secured Such data access security has three strands encryption authentication and authorisation In order to ensure eavesdroppers cannot intercept any data being transferred to or from the data server all traffic is encrypted When accessing data via a web browser this means using the HTTPS protocol as opposed to the more commonly used HTTP protocol Users know their connection is secure by the browser displaying a symbol such as a padlock Figure 6 eoo 4 i 5 https www energyuse org uk Figure 6 Padlock symbol used by the Safari web browser to indicate all communication with the web server is secure The second strand involves ensuring people attempting to access the service are allowed to do so and that they are who they say they are authentication This involves issuing people who need to access the service with a username and password The final strand involves making sure the authenticated users only have access to the data they are entitled to see authorisation For example a householder should only have access to energy consumption from their own household whereas a member of the University s BEU team should have access to all household energy consumption data These three security measures are fu
31. der to analyse the monitored consumption accuracy The recorded data is shown in 47 UF Table 5 Table 5 Version 4 circuit testing data Initial meter reading m 6032 219 Final meter reading m 6068 525 Time interval days ree The test house end of terrace town house uses gas for water heating central heating and hob cooking The error is deemed to be acceptable The longevity of the power supply is now expected to be approximately 1 year Version 4 was thus deemed to be accurate and have suitable longevity to match the requirements of the project Furthermore initial quotes for the professional manufacture of version 3 circuits indicated an approximate cost of 4 78 per unit excluding battery costs The cost of version 4 circuits were considerably lower 47 20 req RJTI cable Subtotal ENDS D battery 18Ah Toa M MOI The version 4 circuit superseded version 3 as the conversion circuit used in all subsequent in the field installations An alternative for non pulse ready or pulse enabled meters is to use optical sensors to detect the dial rotation Many meters have a silver reflector on one number position of the fastest moving dial It is possible to detect this although it is much less reliable than magnetic pulse based solutions However the circuitry is non trivial and development time could be longer than desired It was decided that this option would only be considered if it was found to b
32. e independent of our data servers that would provide all of the data analysis storage and visualisation functionality for a single household was also investigated The motivation for this was to investigate low cost both in terms of purchase price and running costs options for entirely in the home devices for residents who either did not have Internet access or preferred not to have their energy consumption behaviour transmitted over the Internet The standalone device allows all analysis and visualisation to occur in the resident s home A prototype stand alone device using a Raspberry Pi was developed and tested with the Current Cost EnviR device This used all of the previously described technology currently in use for the main energy consumption browser see Section 3 but all running on an ultra low power platform In addition to testing the suitability of the ultra low power platform and the constituent technologies an important aspect was to investigate what impact this would have on the deployment of the existing codebase both for the data server and the energy browser application and the overhead to adapt this to a standalone environment thus not needing support for multiple user access not requiring secure upload of data etc Our findings showed only relatively minor changes were necessary and that the energy consumption browser worked effectively These devices were not installed as part of the BlG Energy Upgrade since this would
33. e strictly necessary 2 3 3 Independent Connectivity An issue identified during participant recruitment was the potential lack of broadband or in some cases weren t prepared to allow BIG Energy Upgrade equipment to be connected to it Since the new monitoring equipment uses a micro PC with data logging no data is lost during periods of permanent or temporary loss of Internet access Data could be retrieved in person in a similar fashion to that used by WP1 However it is always preferable to have live upload of consumption data Therefore the use of mobile broadband dongles connected to the micro PC for households without broadband connection was investigated Specifically the possibility of using 3G mobile broadband dongles to provide the connectivity for data upload was investigated During testing it was found that the dongles were not compatible with the existing monitoring PC operating system Xubuntu 10 04 or the bespoke monitoring software installed Therefore the use of a new operating system deployment and associated configuration was investigated as well as updating the monitoring software Installing the latest version of Xubuntu 12 04 allowed the use of O2 mobile broadband 3G dongles These were tested and were shown to provide reliable upload connectivity to our data servers However the use of this additional USB based serial device caused incompatibility problems with the monitoring software since the Current Cost Env
34. edback information which would be made available 44 A large number of potential participants were lost at stage 5 site inspection Due to the financial circumstances of the target demographic of the BlG Energy Upgrade many properties were found to be using pre payment meters for their gas Such pre payment devices use the same physical meter connectivity as the monitoring devices to determine gas consumption This has the unfortunate side effect of preventing our equipment from being connected There is no way to share this connection to the meter therefore gas consumption cannot be monitored Although it would have been possible to monitor only the household s electricity consumption it provides no indication of overall house consumption ignores major energy usages such as water heating central heating gas cooking etc thus significantly reducing its value The current economic situation has also forced many households in our target demographic to look at additional ways to cut living costs One outgoing which is commonly considered non essential is broadband The Universitys BEU team has observed a number of households who have recently cancelled broadband subscriptions thus removing the option for wired connectivity to the Internet for the monitoring equipment As mentioned below although the use of a mobile broadband 3G dongle is technologically possible see Section 2 3 3 wide scale deployment of this approach is prohibitively ex
35. eed this mechanism can also be used by participating residents 48 to download part or all of their own consumption data for their own analysis The latter requires no additional software to be installed they can use their normal login credentials used for the web based consumption browser and as with the consumption browser they can only access their own data Following initial prototyping it was found that the Highstock Highcharts technology provided a much better supported and documented approach to displaying time series data The web based energy consumption browser application provides a rich source of information regarding each residents energy consumption but requires the resident to actively visit the website Benefits of a real time display in the home have been observed previously but such displays provide only very basic information see Section 3 1 1 In order to increase the benefit further the visualisation of the householder s consumption from the web application ought to be always available in the home Recommendation 9 Provide a small screen tablet sized that can be placed in the home and provides a live connection to the web based energy browser allowing the resident to instantly and passively observe current consumption It must remain switched on and visible to be effective It should be restricted to only run the energy browser Recommendation 10 The screen should be touch sensitive in order to allow the user
36. erties Unfortunately it is very difficult to establish their suitability for smart monitoring by the manufacturer s model code alone many traditional and smart enabled meters share the same model identifier see Section 2 3 2 Furthermore those that are compatible each require specific hardware to be used to connect monitoring equipment Recommendation 1 Site surveys are essential to establish the compatibility of the gas meter for smart monitoring prior to installation of the monitoring equipment Recommendation 2 Gas meters that are not compatible ought to be replaced or augmented by smart ready meters where possible In one case Recommendation 2 above was followed and the traditional gas meter was augmented with a smart ready gas meter However following discussions with a number of gas meter equipment suppliers it appeared the installed meter is rarely used in the UK and that pulse blocks were prohibitively expensive Fortunately a suitable pulse block had been supplied and installed not usually the case but required bespoke rewiring of the gas monitoring equipment on site time consuming and disconcerting for the homeowner Recommendation When a gas meter is replaced or augmented care must be taken to use a commonly used meter for which pulse blocks are readily and cheaply available see Section 2 3 2 Recommendation 4 system needs be relatively straightforward to install This will minimise cost disruption
37. es are likely to suit different individuals Some may prefer a simple traffic light indicator while other a more comprehensive feedback Within the BIG Energy Upgrade project there was no intension of evaluating these different technologies rather the RTD used was determined by the equipment provider 3 2 Data Analysis and Storage Server The processing and analysis of live data feeds with the ultimate goal of providing web based access to the data was investigated It is important to consider the volume of data produced when designing the processing and storage aspects For example for 25 houses approximately 121 8 million electricity data points and 25 8 million gas data points would be produced annually Care must be taken to balance accessibility of the data with speed of processing Therefore a combination of file system based storage will be used together with a database The latter would provide the core of the data storage for recent 1 2 months fine grained data together with all historical aggregated data average daily instantaneous consumption average total daily consumption etc 3 2 1 Weather data An important factor in the amount of energy consumed by a household is domestic heating In turn the amount of energy required for heating is influenced by the prevailing weather conditions most notably external temperature At the request of other members of the University s BEU team the detailed weather information was integrated
38. es of day at which the web based consumption browser is accessed 43 4 Monitoring Recruitment Experiences The choice of the types of houses to monitor and the communities to address were undertaken by mutual agreement with the Project Partners e g Local Authorities and members of other work packages within the University team In order to maximise the amount of data available to the BlG Energy Upgrade team for each house it was decided to install energy monitoring equipment in the same properties targeted by work package 1 who were monitoring air quality factors Common to both work packages was the desire to cover both pre and post installation time periods where installation refers to the installation of the energy efficient measures In addition it was desirable to monitor a range of properties with arrange of occupancy single occupancy couple occupancy family occupancy 4 1 Recruitment The general methodology for recruitment and installation was 1 Addresses of potential participants gathered by local authority 2 Potential participants contacted by the local authority to gauge interest Contact details of interested residents passed on to the University s BIG Energy Upgrade team 4 Each resident contacted by telephone to explain the purpose of the work the equipment to be installed and the timescales involved 5 Site inspections were conducted for each interested resident to ensure the gas meter was compatible wit
39. g to detect the LED flashes and the EnviR base unit converts this into a measure of electricity consumption Importantly the OptiSmart simply counts pulses The same device can in theory be used to monitor gas meters There are two alternatives for employing the OptiSmart for gas monitoring direct connection between the OptiSmart and the gas meter to measure the magnetic pulse and magnetic pulse to light pulse conversion following by use of the OptiSmart light sensor Since the meter faceplate indicates the volume of gas consumed per pulse this can then be used to calculate gas consumption both in terms of volume and kW Both of these solutions rely on the use of a reed switch to detect the magnetic pulse A reed switch is an electrical switch operated by an applied magnetic field and consists of a pair of contacts on ferrous metal reeds in a hermetically sealed glass envelope The contacts may be normally open closing when a magnetic field is present or normally closed and opening when a magnetic field is applied Once the magnet is pulled away from http www currentcost com product optismart htm 15 the switch the reed switch will go back to its original position For magnetic pulse enabled meters a reed switch can be used in an electrical circuit to detect the passing of the magnet and hence a revolution of the meter dial Initially the first option of direct connection between the OptiSmart and the gas meter was investiga
40. g equipment was installed at the property on 21 October 2013 The residents logged on to the web based consumption browser on the same afternoon and viewed the homepage for 8 minutes Following this the residents have logged on a further 18 times at regular intervals see Table 7 logged on 2013 11 12 2013 11 15 Table 7 Web based consumption browser usage statistics for D13 42 The most commonly viewed part of the web based consumption browser was unexpectedly the homepage showing the current electricity and gas consumption see Table 8 After this the detailed breakdown of the daily consumption Section 3 4 3 received 23 views with occasional viewing of the week s consumption relative to their historical consumption Section 3 4 4 Resource Times accessed 78 3 Weekyvew 6 Table 8 Usage of individual areas of the consumption browser The residents of D13 display a commonly observed pattern of usage in which the energy monitoring website is visited frequently in the first few days following installation and then rapidly reduces Webb et al submitted Encouragingly the residents have maintained regular access since checking their detailed daily consumption approximately eveny 2 or 3 days The residents also tend to check their consumption in the early evening with the most popular time being between 6pm and 7pm Table 9 Resource 8pm 9pm pm 8pm 6pm pm 4pm 5pm Table 9 Most common tim
41. gation bar can be used to change to a detailed view of todays energy consumption a view showing this week s energy consumption or explore all the energy consumption in detail since the equipment was installed 3 4 3 Today s consumption This section shows the detailed view of today s in other words since midnight energy consumption for the residents home There are two graphs the top shows electricity consumption red and the bottom shows gas consumption blue Each plot shows the instantaneous consumption at any point in time watts for electricity and cubic metres for gas 34 Daily Electricity Consumption W Zoom 3hr All From Jul 22 2013 To Jul 22 2013 2 5k Ij 22 Jul Daily Gas Consumption cubic metres From Jul 22 2013 To Jul 22 2013 Figure 13 Today energy consumption plots Each of these graphs is divided into a number of areas which will help the resident get the most out of the information displayed Preset Title and zooms energy units Daily Electricity Consumption W Consumption Zoom 1hr 3hr 6hr All From Jul 22 2013 To Jul 22 2013 area Zk Ok 22 Jul 02 00 04 00 06 00 08 00 10 00 12 00 Zoom area Figure 14 Electricity consumption plot Figure 14 shows an electricity consumption plot The consumption area shows the fine detail consumption information for the residents home Electricity consumption is recorded approximately every 6 seconds whi
42. gging in to the site This show s the resident s current electricity and gas consumption together with the external temperature for their area http shop highsoft com faq what is non commercial 32 BEU Energy Consumption Browser 4 fay https www energyuse org uk Qj BEU Energy Consumption Browser Show S9Now consumption QThis week s consumption Explore full history Live external temperature Absolute Feels like Live Energy Consumption 5k 5 ge e Re s n 5 k 9 2 S 5 ectricity T Watts 2s ubic metres Figure 8 Energy consumption browser homepage Both of the energy consumption dials show the current consumption along with today s peak and today s average In some instance such as shown for the Gas dial one or more of the 3 dials will be superimposed Live Energy Consumption i P i Electricity Gas Watts Cubic metres 097 Highcharts com Today s average Figure 9 Live energy consumption display In addition to the live energy information the live temperature for the residents home location is also shown No external temperature monitors have been installed at properties this data is gathered from wunderground com using their Weather API for Developers The temperature shown is based on weather observations from the resident s locality The temperature can be changed from the a
43. h our monitoring equipment and that broadband power sockets etc were available 6 Property specific equipment purchased e g the correct pulse block for the property s gas meter see Section 2 3 2 f Installation of equipment There are a number of points within this recruitment process where potential participants can be 105 Indeed it has been unexpectedly difficult to recruit participants in to the monitoring process Each local authority has worked extremely hard during steps 1 and 2 of the methodology but with a very low conversion rate In an attempt to increase the conversion rate in steps 2 to 4 it was decided to try recruiting in person in Doncaster The local Neighbourhood Energy Officer altered step 2 from being telephone based to visiting properties in person In an additional attempt to improve the conversion rate it was decided that a member of the University s BIG Energy Upgrade team would accompany him on these face to face visits in order to be able to provide more accurate detailed responses to any queries residents had about the equipment to be installed and the study itself Despite visiting a large number of properties the conversion rate has remained low Aside from the issues described below such as pre payment meters etc the key factors in not participating were a lack of interest desire not to have equipment in their home and a perceived lack of direct benefit to the resident even after explanations of the fe
44. he four different battery form factors an analysis of typical current draw was conducted From this the battery supply lifespan was estimated The results are shown in Table 4 Table 4 Comparison of circuit voltage versus battery type to indicate circuit longevity PP3 AA C D Resistor 565mAh 2700mAh 8000mAh 12000mAh Q 19 3 days 92 2 days 273 2 days 409 8 days 9 8 days 46 9 days 138 9 days 208 3 days 7 0 days 33 4 days 98 9 days 148 4 days Here it can be seen that converting the circuit to operate at 3V and using two 1 5V D cell batteries increases the lifespan to over a year Version 3 of the circuit design reflected the change in voltage see Figure 4 19 555 pin out 1 Ground 2 Trigger 3 Output 4 Reset 5 Control not used 6 Threshold 7 Discharge 8 Vcc Figure 4 Version 3 of the gas monitoring circuit An additional benefit of this change is the lower cost of batteries per circuit The version 3 circuit was installed at the test house A further alternative circuit version 4 shown right relies N on a much simplified circuit Here the number of 0 3V components is drastically reduced and the need for a 555 IC is eliminated with an associated positive impact on cost The accuracy and lifespan of the version 4 circuit was tested at the test house and gas meter readings were OG OV taken at the point of installation After 8 days a second reading was taken in or
45. hich information on bills is capable of inspiring behavioural change Informing Choices 2011 In order to reduce the energy consumption in BIG Energy Upgrade households the technology must inform the resident of their energy usage this is more than simply providing the raw figures or statistics on their usage rather it requires providing means to allow residents to focus attention on their behaviour which is most likely to result in energy saving In other words where their energy consumption is higher than expected Fischer presented an excellent review of the past 20 years research into identifying the key criteria in effective feedback Fischer 2008 The work consolidates the results of 26 projects identifying 5 main areas to consider when developing an effective feedback system these largely mirror the findings from the recent UK Energy Demand Research Project into Smart Meters Raw and Ross 2011 1 Frequency and duration feedback is more effective the more directly after an action it is given 2 Content e g feedback may be given on electricity consumption alone e g KWh on cost or on environmental impacts of consumption 9 Breakdown providing a awareness of the specifics of consumption e g for specific rooms appliances or times of the day is almost the only way of providing a direct link between action and result and thus establishing consciousness of the relevance of an individual s behaviour 4 Medium and mode
46. iR is also connected as a USB based serial device A number of alterations were made to the monitoring and logging software to ensure compatibility 21 A secondary issue which had to be resolved was the standard behaviour of the mobile broadband dongle Since these are external convenience devices connection initiation is assumed to be via human intervention It is desirable that mobile connectivity be independent of human intervention as soon as the logging PC is switched on connectivity should be established and it should be sufficiently resilient to manage loss of connections by resuming the connection at the earliest possible time An auto connect program was developed and deployed on the logging PC which ensures both broadband connectivity can be used if available and where this is not possible the mobile broadband connection is made and managed automatically This occurs directly after power on without any intervention from the installer 2 3 4 Summary of Equipment After development of the gas monitoring circuit was finalised the final set equipment to be installed in a property consisted of Current Cost EnviR RTD incl power block and data cable Current Cost electricity monitor clamp Current Cost OptiSmart Gas meter specific pulse block Magneto optical conversion circuit incl battery pack Viglen MPC L mini PC incl power block and Ethernet cable 2 4 Stand alone Equipment The possibility of a stand alone devic
47. in order to ensure historical consumption costings are calculated correctly by using the energy charges at that point in time 3 2 3 Implementation A database was designed and implemented The MySQL database is hosted on a dedicated database server and consists of 3 types of tables user details weather details and energy consumption details The user tables capture all registered users of the system BIG Energy Upgrade team members participants etc their login credentials authentication and their access rights authorisation The weather data tables store the information about individual locations properties and associated weather observations and daily weather summaries The final set of tables store information about energy consumption These provide metadata about each individual property e g the pulse volume factor of the gas meter as well as the fine grained and aggregated consumption data for electricity and or gas In addition costing information for each energy type is stored Figure 5 shows the full database schema The database is used to store live energy consumption data as it arrives The database is also used as the data backend for the web visualisation tool Section 3 4 In conjunction with the database a suite of fully automated live data feed processing software was created This allows the live consumption data stored in the Current Cost XML format to be stored in the database within minutes of its re
48. in the meter cabinets the minimal kit installed in the home consists of the EnviR screen the Viglen PC and various cables In homes with a lack of power sockets or broadband sockets additional power extension cables Ethernet switch and cabling is required A number of residents have commented on this both from the perspective of the amount of space required and its unsightliness In essence the EnviR display is non essential as the web interface provides a much richer description of the household consumption pattern its only required to collate the sensor data and pass this to the Viglen PC Recommendation 6 A move away from commercial off the shelf equipment for the base technology should be considered Bespoke electricity and gas sensors could communicate with a single small device responsible for data collation logging and transmission On a number of occasions it was found that the equipment to be installed once on site suffered from radio interference from an unknown source On many of these occasions the common factor was the presence of PV solar panels however on another occasion there were no PV panels and no obvious source of interference Since this prevented communications between the display screen and the electricity and gas sensors installation had to be abandoned Recommendation 7 During on site survey to establish the suitability of the gas meter the EnviR equipment should be tested to check for radio interference and
49. l meter must be inspected to check for pulse compatibility by looking for the Pulse 0 01m text Pulse blocks are designed for a specific meter and are generally not interchangeable Following successful lab based testing a full electricity gas monitoring installation was made in a test house and its effectiveness analysed A significant deviation between the expected gas consumption calculated from meter readings 6 25m and that estimated by the logging circuitry 14 63m was encountered It was determined that the OptiSmart device was not operating as expected it was concluded that the OptiSmart device contains a form of ticking counter when activated continually the counter increments repeatedly like a ticking clock Under most conditions gas flow in domestic environments is relatively slow As such the fastest moving dial and hence the associated magnet rotate relatively slowly Even during sustained gas consumption the moving magnet can still remain sufficiently proximate to the reed switch to cause activation Therefore each rotation of the dial can result in multiple impulses being registered by the OptiSmart device and thus recorded by the EnviR and logging PC This accounts for the over estimate of gas consumption In order to avoid this issue it is necessary to only illuminate the LED light source momentarily regardless of the duration of the reed switch activation One approach for achieving this was to employ a
50. le gas consumption is recorded approximately every 27 seconds As standard the consumption area displays all of today s consumption information In this example data is shown from midnight until the point at which the snapshot was taken at approximately 2pm It can easily be seen that there are two spikes in electricity consumption the first just before 8am and the second just after 10 30am these are likely to be due to a kettle being used 39 To view these in more detail the resident can zoom in using either the preset zooms or the zoom area The preset zooms allow the resident to quickly zoom into either 1 hour 3 hour or 6 hour views The All zoom returns to the usual zoom level showing all of todays consumption information The alternative way of zooming uses the zoom area to allow the resident to zoom into anything the resident is interested in right down to just a few minutes To zoom using this control one or both of the blocks at the end of the plot can be clicked and dragged see Figure 15 Daily Electricity Consumption W Zoom 1hr 3hr 6hr All From Jul 22 2013 To Jul 22 2013 blocks Figure 15 Energy consumption exploration using the zoom functionality As the resident drags the zoom blocks the upper consumption area updates automatically The zoom area itself will remain unchanged so the resident can still see the context of the whole day s consumption information outside the zoom
51. lly integrated into the data access server 3 4 Web based Consumption Browser The goal of the web based consumption browser was to build upon information gained from the monitoring activities in order to provide personalised energy use feedback to residents in order to instigate appropriate energy saving practices and enhance broader energy literacy The core requirements included e Personalized information about all house energy consumption e Up to date information e Graphical view http en wikipedia org wiki Comma separated_ values 8 https www energyuse org uk api 30 e Expenditure as well as raw energy units e Comparisons between current consumption and historical e Use the REST API for data access The principal concern of this work was the creation of optimal programs and interfaces to feedback the information to participants Building on our previous work on the current state of the art and best practices related to Human Computer Interfaces for displaying energy data in simple and effective ways the most suitable technology available for achieving good resident feedback was investigated At all stages of the design and development process discussions with the end users informed the type of information to be displayed and the way in which it was shown Feedback was sought on preliminary designs and implementations which was used to refine and guide future work In addition the Computer Science team collabora
52. lution which was both commercially available and met the BIG Energy Upgrade projects requirements for gas as well as electricity For instance e Ewgeco http www ewgeco com produces an award winning real time energy display device which is the only 3 utility including micro generation display accredited by the Code for Sustainable Homes where builders gain 2 credits under section ENE 3 However even this prosumer industrial level product only supports a small subset of gas meters those based on magnetic pulse outputs http www ewgeco com downloads document get Ewgeco Gas Meter Connect ion Guide pdf Furthermore the cost of 550 per unit was prohibitive given the number of installations required e UPL Meter Mimic device http www up Itd co uk services metering meter mimic is designed for industrial scale installations rather than domestic metering and again relies on magnetic pulse enabled equipment and in some cases uses optical sensor technology This is an expensive option e The SaveOmeter has recently offered gas monitoring http www saveometer com gas monitor aspx but there is no facility for data upload or logging e Current Cost was working on a gas monitoring solution compatible with the EnviR units However Current Cost were contacted twice during August and both times the response was that these were still under development and are not yet available It ought to be noted that the vast majority
53. n menu at the top of the page Figure 20 Energy units Raw units Cost Figure 20 Energy unit choice control 38 In the weekly consumption plot electricity is shown in red and gas in blue Daily consumption totals for the immediately preceding week are shown in dark colours The average consumption for each day of the week are shown in a lighter shade By clicking on any of the columns that day s full consumption view is shown in the plots underneath For example if the resident clicks on one of the Wednesday columns the detailed electricity and gas consumption for Wednesday will be shown in the plots below Using the week s consumption plot it is easy to see if the resident has been using more or less of each type of energy than usual Note that the site will only provides costing information if the resident has informed us of how much they pay for their electricity and gas 3 45 Full history This section of the website provides full access to all the historical energy consumption information for the residents home since the equipment was installed Just like the week view the consumption information is shown as a total for each day for each energy type Electricity is shown in red and gas in blue The average daily temperature for their home s location is shown in green to provide an indication of the weather conditions at each date Temperature is shown on the left hand vertical axis while energy consumption is shown
54. n the home and their willingness to adopt or facilitate interventions and policies aimed at reducing energy demand In the UK as in other countries e g Australia Italy United States France Ireland it is anticipated that the rollout of in home smart metering technology should help to reduce household energy consumption see DCLG 2010 DECC 2011 In short it is hoped that the real time energy use feedback that can be provided to householders via in home display units or smart monitors associated with smart meters should help to improve public awareness of the pattern and cost of their energy use behaviour resulting in enhanced energy literacy and decreased energy consumption see Darby 2006 That clear feedback is a necessary step towards educating consumers with regard to their energy consumption has been identified by several pieces of research e g Darby 2006 Fischer 2008 Gans et al 2011 Pierce et al 2008 The nature of successful feedback and the method by which success is identified is however less clear The majority of home energy monitors provide real time consumption data research demonstrating that this level of feedback is effective e g Ueno 2005 However more complex feedback technologies have either not received the same level of analysis e g Petersen 2009 or have been largely unsuccessful e g Google PowerMeter The fundamental aim of the energy monitoring activity within the BlG Energy U
55. nergyuse org uk aderi o Energy Consumption Browser beta Household 00140b2309f0 Dato 27 D 2012 Go Electricity Consumption 500 000 500 3 aoa I7 a NR RET E iL c m JL 7 1 1 r 1 p Ln 1 1 I T 04 AM 05 AM 08 AM 07 AM 08 AM 11 1 i T Tue 27 01 AM 02 AM 03 AM 09 AM 10 AM AM 12 PM Time GMT Gas Consumption sae fH J a 1 0 020 09 10 11 AM 12 PM EUROPEAN UNION Investing in Your Future European Regional Development Fund 2007 13 ee Figure 7 D3 based energy consumption browser prototype homepage The main alternative to D3 were Highcharts Highcharts JS is a JavaScript charting engine which is capable of running in all modern web browsers Although launched as a charting product targeting at financial visualization Highstock allows the creation of stock or general timeline charts in pure JavaScript including sophisticated navigation options like a small navigator series preset date ranges date picker scrolling and panning Highcharts is a mature technology and respected by the industry For instance 51 of the Fortune 100 Global list of companies are Highcharts Highstock licensees Since the BIG Energy Upgrade project is non profit Highchcarts is licensed free under the Creative Commons Attribution NonCommercial 3 0 License 3 4 1 Homepage The homepage is the first page seen after lo
56. of gas monitoring solutions even the premium ones are only compatible with pulse ready and pulse enabled meters 2 13 Remote monitoring For remote monitoring it is necessary to connect the in home display to some transmitter over the Internet Most systems provide some proprietary hardware which is hardcoded to send information to the manufacturer s data servers These approaches vary in their ability to deal with communication outages for example the ISA iMeter does not store any data so any outage will lose data The AlertMe system has a ten day data buffer alleviating to some extent the network issues encountered The Current Cost gas monitoring solution was finally released mid to late Summer 2013 too late for deployment by the BIG Energy Upgrade project 12 2 2 Phase 1 Installations The initial decision was to use the ISA iMeter Kit as this was the only equipment available which specifically claimed to monitor both gas and electricity In retrospect this decision was mistaken While the kit did in theory provide the ability to meet the project requirements in the field trials identified a number of drawbacks which severely limited its practicality in the actual real world households The version of the RTD used is the Efergy Classic 1 0 While this does have the benefit of being small and battery powered and is therefore both convenient to be situated in the home and is not reliant on proximity to mains power it did n
57. of presentation the information needs to capture attention and be understood before it can become effective 5 Comparisons There are two basic types of comparisons e Historic comparison relates actual to prior consumption often normalised e g temperature corrected with the same period in the previous week or year e Normative comparison compares consumption to that of other households e g with a national or regional average households in the neighbourhood or households that are in some way similar e g in size type of house application stock Therefore feedback systems must be e based on actual consumption and given frequently ideally daily or more e provide interaction choice for households to present information in understandable and appealing way e involve specific breakdown 24 e given over a continuous long historic period e may involve normative comparisons although research has indicated that such comparisons are appreciated by households the effects are less clear However there is a limited amount of research indicating the beneficial effects of normative comparisons Ayres et al 2009 3 1 1 Real time display There is a variety of Real time Displays provided with the Smart Meter systems and a recent report surveyed the 13 available systems in the UK Consumer Focus 2011 The general finding was that all the systems provide both pros and cons in their feedback and that different interfac
58. on to remote servers for monitoring and feedback 2 1 State of the Art At the start of the project a review of the state of the art energy consumption monitoring equipment was made In addition to the technological requirements mentioned above it was important that any system was cost effective in order for the project to afford to install full monitoring kits in a range households Table 1 details the eight leading devices and providers correct at time of compiling Summer 2011 of current energy consumption monitoring systems available to the general public and their applicability to the BlG Energy Upgrade requirements Table 1 Applicability of Currently Available Energy Consumption Monitoring Systems to BIG Net Log Website Cost Energy Upgrade Requirements ISAiMeter O0 0 Current Cost O O Oow __ 0 Eco eye 7 SaveOmeter O O po a ofo po a 29 17 2 www diykyoto com uk Alert Me Table 2 shows a number of other monitoring systems which not yet available or available via specific utility providers Table 2 Applicability of Other Energy Consumption Monitoring
59. ot provide any data logging This means that if the unit loses remote connection either within the home i e between the RTD and the electricity sensor or from the home to the data servers i e between the RTD and the server then all the data is irretrievably lost Indeed despite significant effort the gas monitoring sensor could not be made to communicate with the RTD Furthermore it was observed that the network connection hardware software was not robust the monitoring devices regularly failed to either initialise or in cases when a connection could be made maintain a connection to the data server Together these issues resulted in a system that was ineffective in providing in home real time feedback of electricity consumption and ineffective in transmitting data back to the central server Both of these requirements are majors factor in the effectiveness of smart meters 2 3 Phase 2 Installations In order to address the problems associated with the ISA equipment alternative monitoring kits were re evaluated Despite the lack of native support for monitoring gas consumption i e no commercially available Current Cost device for monitoring gas consumption the Current Cost EnviR device was selected There were a number of reasons for this choice 1 More detailed realtime feedback via the RTD 2 Consumption data easily available via USB from RTD 3 Potential to create bespoke gas monitoring device compatible with the EnviR Alth
60. ough the EnviRs RTD requires mains power the company claim this is more environmentally friendly than using batteries it does provide more complex feedback The possibility to download consumption data directly from the RTD via USB can be utilised to address the prevention of data loss These aspects are described in more detail below The Current Cost EnviR package consists of EnviR receiver and RTD Electricity monitoring clamp Electricity monitoring transmitter Data cable When the monitoring clamp is fitted around the mains cable in the electricity meter cabinet and connected to the transmitter instantaneous electricity consumption is displayed on the EnviR RTD The instantaneous consumption reading for electricity is updated every 6 seconds The RTD provides additional basic information such as a cost estimate and coarse historical consumption 2 3 1 Logging and Communications Current Cost also provide internet connection hardware NetSmart for sending data to Current Cost servers however there had been concerns about it s robustness and given both the need for a reliable logging of data and remote connection to the University rather than Current Cost the decision was made to a device which would allow custom logging and custom data communications It is not uncommon for Internet connections to become unavailable connection lost router unplugged etc Indeed it has been observed that some users only switch their router
61. pensive At all stages of the methodology it has been very difficult to persuade potential participants of the benefit of the monitoring work Even when the web based energy consumption browser was fully available the potential that this offered for increasing energy literacy and identifying energy consumption savings had little effect on the conversion rate 4 2 Logistical issues During installations a number of logistical issues were encountered which either delayed installation or in extreme circumstances prevented installation completely The first was a lack of available electricity sockets The electricity and gas monitoring equipment requires at least two 240V sockets one for the EnviR monitor and the other for the Viglen logging and data transmission PC In many cases an additional socket was also required by WP1 for air quality monitoring equipment This situation is made worse if the household requires an addition switch hub see below which would require an additional socket In the majority of houses it was found that only one socket was available A four gang extension lead would solve this issue However this increased the apparent bulk of the equipment being installed and was of concern to residents In one property in Doncaster the resident had such a severe lack of sockets that it was not possible or safe to use additional extension leads and so energy monitoring was not possible A second issue which was encountered
62. pgrade project is the empowerment of residents and communities so that they may contribute to the energy usage reduction objectives and in turn reduce expenditure and CO emissions Specifically Task 1 focussed on the implementation of remote monitoring of energy consumption in houses electricity gas etc via broadband The work performed to achieve this goal is described in Chapter 2 Houses were monitored over secure encrypted connections which effectively formed a virtual private network connecting households monitoring equipment and the University of Sheffield s servers via the public Internet Although initially intended to leverage South Yorkshire s Digital Region superfast broadband the failure of this initiative meant that low level connectivity was via the homeowner s personal domestic broadband The data collected was analysed to provide personal profiles of energy usage which were fed back to users see Task 3 Due to the unexpectedly low level of up take see Chapter 4 insufficient data was available to analyse cross regional profiles according to the type of accommodation family etc In order to achieve the latter monitoring equipment would have to have been installed in a number of households for each type of accommodation and each size of family and each representative geographical location etc something the low level of recruitment precluded Task 3 focussed on the provision of personalised energy use feedback to http
63. prevent access to those data streams 22 2 5 Partner based Monitoring To enable a larger number of dwellings to be monitored cost effectively the University provided a number of workshops and tutorials in which representatives from Partners were upskilled in the monitoring installation and maintenance techniques Documentation was created for partners describing the criteria for selecting additional houses for monitoring see Appendix This provided a brief status update on the monitoring kit development and a detailed description of the types of gas meters which are compatible for gas monitoring Furthermore it provided details of the equipment required for the installations their costs and suggested suppliers This information was disseminated to partners Training for equipment installation was provided to all partners via workshops and on site training was provided to Farrendale Ltd working on behalf of North Lincolnshire Homes 29 3 Resident Feedback Technology 3 1 State of the Art A recent article found that 0 8096 of householders undertook some kind of day to day energy efficiency measures However the two barriers to energy savings were identified as the cost and the provision of appropriate information to enable effective decisions Yohanis 2012 This finding is reinforced by another study which found 3596 of consumers did not understand their energy bills and that there is a gap in understanding about the extent to w
64. recommendations 2 In Home Monitoring Technology This work focussed on the collection of data on energy usage patterns Cost effective monitoring was achieved by remote monitoring via broadband thus allowing the monitoring of dwellings from across the region subject to availability of broadband access Houses were monitored via the creation of encrypted virtual communication channels able to carry sensor data to a central server located within the University while providing security and privacy see Figure 1 Participant s home University of Sheffield CA NS Electricity Data analysis Ni Data access server amp web Server Data logging and communication Figure 1 Data flow between monitored houses and the University Note all communications are secure Participant selection and recruitment was subject to a number of caveats and was made in collaboration with the other work packages e g the team from architecture investigating changes in air quality More details of this can be found in Section 4 Once the sensors were installed the collected consumption data was analysed to provide personal profiles of energy consumption which were fed back to users see Section 3 For the BlG Energy Upgrade project three core technology requirements were identified 1 ability to monitor electricity and gas 2 ability provide a real time display RTD for in home monitoring and 9 ability to transmit household energy consumpti
65. rsions of the bespoke data upload software All communications between the mini PC and the University s servers are encrypted and secure to ensure the residents energy consumption behaviour remains private Furthermore a secure bi directional communication channel is maintained between each residents mini PC and the University in order for members of the Computer Science team to perform remote maintenance of the mini PC The entire mini PC configuration procedure was semi automated using web based storage to allow the Viglen PCs to be installed and configured from any location but minimal interaction Instructions were produced describing the installation and configuration process to allow partner based configuration and installation without additional assistance from Computer Science The automated PC configuration program has also been updated to reflect these changes 2 3 2 Gas Monitoring Ways of cheaply adapting the Current Cost equipment the OptiSmart device shown right to allow basic gas monitoring in dwellings without a pre payment meter were investigated The Current Cost OptiSmart can be used as an alternative to the clamp transmitter that originally shipped with the energy monitor s if and only if the electricity meter is compatible Many modern electricity meters have a flashing LED indicator usually 1000 flashes per kWh Instead of measuring the current induced electromagnetic field the OptiSmart uses a light sensor s
66. rventions had been installed and the quality of the work undertaken The act of uninstalling the equipment was a form of retaliation Recommendation 18 Any information presented to the resident regarding such monitoring activities should stress the separation between activities conducted by the University wider research team and on site activities by contractors and the relevant channels for each by which complaints or concerns can be raised 91 6 References 1 10 12 18 Ayres l Raseman S and Shih 2009 Evidence from two large field experiments that peer comparison feedback can reduce residential energy usage Working Paper 15386 National Bureau of Economic Research September 2009 Baker W 2011 Understanding fuel expenditure Technical report Consumer Focus October 2011 http www consumerfocus org uk files 2011 10 Understanding fuel expenditure summary pdf Darby S 2006 The Effectiveness of Feedback on Energy Consumption A Review for DEFRA on the literature on metering billing and direct displays Environmental Change Institute University of Oxford Consumer Focus 2011 Getting to grips with smart displays research review Technical report Consumer Focus November 2011 http www consumerfocus org uk files 2011 11 Getting to grips with smart displays research review pdf DCLG Department of Communities and Local Government 2010 Focus on behaviour change reducing energ
67. s however shown to accurately reflect the gas consumption reported by the gas meter Since this monitoring period was still short of the agreed 2 week minimum options were investigated regarding power supply alternatives 1 Use two 9V batteries in parallel to double the effective lifespan of the power supply 18 2 Usean external mains power supply 9 Usean alternative battery supply and possibly change voltage of circuit Although option 1 was the quickest and easiest to implement option 3 offered the greatest potential benefit A survey of commercially available batteries was conducted Each type of battery offers differing capacities depending on the internal chemical construction The capacity of each battery type indicated in mAh with higher values being better are shown in Table 3 Table 3 Comparison of battery capacities Battery Alkaline Carbon Zinc Lithium 15 8000 3800 15 12000 8000 amp As can be observed the capacity available from a 9V PPS can be significantly lower that other battery form factors Indeed the D cell form factor exhibits approximately 21 times higher capacity than the 9V PP3 The minimum voltage at which the CMOS 555 IC can operate is 2V therefore the lowest battery supply voltage that can be used in 3V two 1 5V battery in series In order to determine the potential lifespan of battery based supplies to 3V 6V and 9V versions of the circuit using t
68. s do not necessarily differentiate within models for those which are pulse enabled and those which are not For example some Sensus Cubix U6 meters are pulse compatible with a pulse block whereas other Sensus Cubix U6 meters cannot provide a pulse Each individual meter will have to be inspected to check for pulse compatibility by looking for the Pulse 0 01m text Note also that pulse blocks are designed for a specific meter and are generally not interchangeable Equipment for energy consumption monitoring Electricity only 1x Current Cost EnviR with clamp 1x Viglen MPC 1x Network cable Additional equipment for non broadband households 1x 3G USB dongle 99 Aaditional equipment for gas monitoring 1x Current Cost OptiSmart 1x Magneto optical conversion circuit 1x Pulse block if RJ11 socket not already present Notes 1 2 3 The Viglen MPC will require installation and configuration Materials will be provided on how to do this Face to face training and assistance can also be provided The OptiSmart operates in conjunction with the EnviR used to monitor electricity consumption For households monitoring both has and electricity only one EnviR will be required The magneto optical conversion circuit is currently a custom built circuit Depending on the outcome of the on going gas consumption testing this circuit may be supplied by the University or by an external supplier Suppliers EnviR Amazon approx
69. ship 15 5 million programme part financed by the European Union Regional Development Fund ERDF through the Yorkshire and Humber ERDF Programme 2007 13 addressing the priority needs of reduction in carbon emissions and creation of jobs To address the issues in an integrated approach the University of Sheffield has brought together a multidisciplinary team of academics who are working alongside Local Authorities ALMOs social housing providers and an energy services company in delivering this project The University of Sheffield would like to thank the Local Authorities that have taken part in this research namely Barnsley MBC Doncaster MBC Kirklees Council Leeds City Council North Lincolnshire Council and North East Lincolnshire Council The University of Sheffield would like to extend particular thanks to Anthony Wright Neighbourhood Energy Officer for Doncaster MBC for his time and effort during the participant recruitment phases Our appreciation also goes to their Housing Partners Berneslai Homes Kirklees Neighbourhood Housing North Lincolnshire Homes St Leger Homes West North West Homes Leeds and Management Partners NPS Barnsley Leeds ALMO Business Centre which have also contributed to the University s work Contents ACKNOWLEDGEMENTS CONTENTS INTRODUCTION IN HOME MONITORING TECHNOLOGY 1 2 2 1 2 2 2 3 2 4 2 5 STATE OF THE ART 2 1 1 ELECTRICITY MONITORING 2 1 2 GAS MONITORING 2 1 3 REMOTE MONI
70. t Note that it possible to monitor households without broadband by using 3G mobile broadband dongles O2 E173 However this will increase the associated costs initial purchase of the dongle plus on going monthly charges The strong preference is for households with broadband If it is not possible to select electricity only households then the gas meter must be pulse ready or pulse enabled These types of meter can be identified via the face plate surrounding the dials There should be text similar to Pulse 0 01m Instead of Pulse meter may use Impulse Imp or similar e A pulse ready meter will have a RJ11 socket small telephone type socket on the front near the dials e A pulse enabled meter will require a pulse block These are clip on or screw on units which provide RJ1 1 access to the pulse Compatible gas meters include but are not limited to Actaris G4 or U6 gas meters Pulse block required Actaris SC6 TVI Pulse block required Actaris R5 RJ11 socket integrated no pulse block required Elster BK series Pulse block required Krom Schroder BK G4 Pulse block required ltron Pulse block required U6 Pulse block required ochlumberger G4 or G6 Pulse block required Schlumberger R5 RJ11 socket integrated no pulse block required UGI Limited R5 RJ11 socket integrated no pulse block required Wilson G4 meters RJ11 socket integrated no pulse block required Note that manufacturer
71. ted The OptiSmart light sensor plugs into the OptiSmart using a plug utilising 4 wires Two of these wires can be used to connect a reed switch Based upon testing the OptiSmart expects the reed switch to operate in a normally closed fashion When the reed switch opens the OptiSmart starts to increment every few seconds The implication of this was that a normally closed reed switch was required and that additional circuitry was required to create a pulse at the moment of the reed switch activating rather than have a period of time in which switch is open and thus the OptiSmart incrementing The latter is an issue since gas consumption could be sufficiently low that the slow moving dial and magnet would allow the switch to be activated for long periods of time contiguously and this result in a sustained period of counter increment Additionally if the gas consumption happened to stop with the dial in the switch activation position counter increment would occur for a long period of time hours until gas consumption resumed Circuits can be developed for this purpose but the estimated expense and development time were deemed to be to high given the current time constraints The alternative option was to create a circuit which utilised the reed switch to convert the magnetic pulse into an activation of an LED light source This light source would then be coupled to the OptiSmart optical sensor Testing showed that the OptiSmart sensor
72. ted with the BIG Energy Upgrade research team from the Department of Psychology in which findings from their user studies and interactions with the community was used to guide the design of the user interface and the types of information to be presented A number of state of the art web technologies for graphical data visualisation were investigated e g Flash D3 Highcharts etc Our initial prototypes were created using D3 see Figure 7 D3 provided relatively easy access to our REST based data but the overhead in creating visually appealing and functionally rich interfaces was considered too high For example in addition to simply showing the consumption trace for each energy type it was considered important to allow the user to use their mouse to explore the details of each trace Such explorations should allow the user to see the precise consumption at any point zoom in to specific regions of interest click on a particular plot location to load a new visualisation Although many of these functionalities were in principle possible they required significant development Since such functionality was already available in other technologies there was little point in re inventing the wheel Furthermore D3 is relatively immature as a technology and lacks support http www adobe com uk products flash html http d3js org http www highcharts com 31 BEU Energy Consumption Browser e 9 www e
73. to the consumer and potential need to rectify problems subsequently Development of the core monitoring hardware was significantly more difficult than anticipated at the beginning of the project This was caused by two main factors unreliability of of the shelf technology and difficulty in monitoring gas meters When the project started low cost monitoring technology for domestic environments was still in its infancy many solutions only focussed on electricity consumption monitoring and few allowed direct access to the collected data Initially there were a number of technical quality issues related to the ISA meters and displays which were identified only after installation To address this issue new equipment had to be investigated and used to replace the existing installations and then be used for subsequent installations A longer than expected time was required for development and testing of a suitable gas consumption monitoring solution This was a technically complex task which was not originally anticipated in the time plan Recommendation 5 feasibility study phase should be included in future studies of this nature in which alternative technologies are deployed in parallel to determine the optimum configuration given the project requirements Af Although effort was directed toward creating a minimal installation kit the monitoring equipment is still reasonably large In addition to the electricity and gas sensors which are hidden
74. tware modules was created in order for the Viglen PC to handle all of the data logging and secure data transmission The Viglen mini PC constantly polls the EnviR for new consumption data When new data is available it is read converted from the Current Cost proprietary format and saved to a file on the PC s hard disk Each data point is time stamped using the PC s internal clock as opposed to the EnviR clock which can easily become incorrect Periodically this log file is closed and a new one started This ensures the log file does not become too big and reduces the potential data loss if corruption occurs If no Internet connection is available all historical consumption logs remain on the mini PC s hard disk If an Internet connection is available the consumption data is securely transferred to the University every 5 minutes with successfully transferred data being removed from the mini PC If the Internet connection is intermittent logging data is stored on the mini PC until the Internet connection becomes available at which point as much data as possible is transferred before the connection disappears l4 To ensure optimal use of the householders broadband or the project funded mobile broadband data allowance work was conducted to reduce the amount of data transmitted via the Internet connection A number of options were investigated and the final implementation resulted in a 9396 reduction in bandwidth usage compared with previous ve
75. y demand in homes London UK Crown Copyright DECC 2011 Department of Energy and Climate Change Smart metering implementation programme A consultation on the detailed policy design of the regulatory and commercial framework for DCC London UK Druckman A and Jackson T 2008 Household energy consumption in the UK a highly geographically and socio economically disaggregated model Energy Policy 36 8 3167 3182 Fischer C 2008 Feedback on household electricity consumption a tool for saving energy Energy Efficiency 11 79 104 Gans W Alberini amp Longo A 2011 Smart Meter Devices and The Effect of Feedback on Residential Electricity Consumption Evidence from a Natural Experiment in Northern Ireland CEPE Working Paper No 78 Hills J 2012 Getting the measure of fuel poverty final report of the fuel poverty review Technical Report CASE report 72 Department of Energy and Climate Change March 2012 http www decc gov uk assets decc 11 funding support fuel poverty 4662 getting measure fuel pov final hills rpt pdf Hirsch D Preston l and White V 2011 Fuel expenditure Fuel poverty and spending on fuel Technical report Centre for Sustainable Energy October 2011 http www cse org uk downloads file understanding fuel expenditure pdf Informing choices consumer views of energy bills Technical report Consumer Focus March 2011 http www consumerfocus org uk files 2011 03 Informing

Big Energy Upgrade Computer Science Final Report

Contents

Download Pdf Manuals

Related Search

Related Contents