User manual for central monitoring and control
Contents
1. 9 a S IEC 61850 5 Communication zinv DRCC MMXU cswl Internal Inverter Controller gt 0 Interface ttt ee 3 g 6 oO n e 5 E S IEC 61499 Application 5 o0 5 v oO 2 gt gt s 2 5 E K IEC a 61850 Inverter qd POCE b J o o m l L p Inverter Control fal E Xe Power Electronics FIGURE 8 UPGRADE OF A PV INVERTER WITH IEC 61850 FUNCTIONS USING A GATEWAY DEVICE BE Page 20 of 29 User manual for central monitoring and control functions 31 10 2015 Some of the IEC 61850 functions may be implemented using a simple forwarding of the data between the inverter and the DSO e g measurements However more complex IEC 61850 functions may not have a direct equivalent inverter function In such cases IEC 61499 FBs can be used to convert one complex IEC 61850 function into a combination of simpler inverter functions An example may be to implement a Volt VAr droop control with IEC 61499 FBs for an inverter which uses a reactive power reference independent from the current voltage Using IEC 61499 it is also possible to aggregate IEC 61850 functionality provided by a LN into a FB In the following Figure 9 a an IEC 61850 logical node called DRCC is represented which allows to ad2. tease 10 2 5 5 CURRENT AND VOLTAGE CHECK 11 2 5 6 LOSS ATTRIBUTION si ee renee 11 2 5 7 IDENTIFY FAULTS AND 5 5 13 3 OPERATIONAL PV HEALTH SCAN s cciscccroscreccrsieesesehaieeaeeiehages deseaeenieeaaesicess 14 3 1 PURPOS Ei zea dcanaaaadnscien sdasceadewashariscaieeaas E A R E AAS 14 3 2 SC OPE O ncn tence 15 3 3 REQUIREMENTS sirrinin ennenen E E a ii 15 3 4 USER INTERFACE aa a a a aa 15 3 5 e 15 3 6 SALES CHANNEL 16 4 SOLAR SENSOR CHEGK rariora EE A 17 4 1 PUPO 17 4 2 SCORE sswunace cde E 17 4 3 17 4 4 USER INTERFACE rrii 17 4 5 RESULTS iasa A 18 4 6 caren nanaramne na narcnamanarmeenanaians 18 5 REMOTE CONTROL INTERFACE 5
3. 20 5 1 PURPOSE ruiou enn 20 5 5 ARCHITECTURE AND 20 5 3 PROTOTYPICAL IMPLEMENTATION stxtinuisencticuarseiemamrimeacuninuaanmiienigaiaenineiaid 21 5 3 1 21 5 3 2 22 REFERENCES 23 ANNEX 1 GOOD PRACTICE FOR PV PLANT MONITORING 24 ANNEX 2 26 PERFORMANCE RATIO aa a a 26 Page 2 of 29 User manual for central monitoring and control functions 31 10 2015 TEMPERATURE CORRECTED PERFORMANCE RATIO ENERGY Page 3 of 29 User manual for central monitoring and control functions 31 10 2015 SUMMARY Th
4. 12 Gsrc There are different options when choosing the reference temperature Tep The first option is to simply take the STC temperature i e 25 C However with T equal to 25 C the annual temperature corrected PR is typically higher than the annual non corrected PR Therefore in 9 it is proposed to carefully determine T such that the annual temperature corrected PR is equal to the annual PR This is achieved with the irradiation weighted average cell temperature over one or more complete years Y re DalGpaa 13 The accuracy of T is not very crucial but it is important that T in 12 and 13 are obtained by one and the same approach Therefore T is typically not measured but modelled E g the module temperature models mentioned in 6 are well suited for this purpose ENERGY PERFORMANCE INDEX The energy performance index EPI is defined as the ratio between the yield and the expected yield Y as determined by a PV model using the actual weather data as input to the model over the assessment period 10 oe 14 EPI 5 exp The advantage of using the EPI is that its expected value is 100 at project start up and independent of climate or weather This indicator relies on the accuracy of the expected model Unfortunately there is not yet an established model for the expected yield of PV systems let alone a well accepted methodology to determine its parameters
5. 31 10 2015 4 SOLAR SENSOR CHECK 4 1 PURPOSE The Solar Sensor Check is a computational tool It is applied to check the integrity of solar irradiance sensors based on measurements and indicate if their measurements are wrong or imprecise Typical causes for faults and imprecisions that have been identified in the past are calibration errors sensor shading soiling time shifts and wrong tilt or azimuth The Solar Sensor Check detects the most important faults and imprecisions automatically based on measurements It then returns recommendations on the most probable root causes Users may want to perform this check because the calculation of availability and performance ratio PR as used for contractual guarantees fines and bonuses in PV relies directly on the measurements from the irradiance sensors Consequently an independent method and tool to verify the correct operation of these sensors is very valuable The tool may be offered as a tool to consultants or to plant owners and operators 4 2 SCOPE The Solar Sensor Check is focussed on e All kind of sensors for global solar irradiance e Different orientations e Sensors installed at PV plants or elsewhere 4 3 REQUIREMENTS Applying the Solar Sensor Check requires the following e Metadata to be entered by the user geographical position tilt and azimuth sensor type if available e 3E requires reliable satellite based irradiance data for the region e At least one mo
6. E g the industry leading PVSyst software provides a detailed and relatively well known model for the expected behaviour but the determination of its many parameters is an art in itself and default parameters have undergone important modifications with every new version of the software and are still being disputed EE Page 28 of 29 User manual for central monitoring and control functions 31 10 2015 Remark that the performance ratio refers to standard test conditions which are different from expected conditions in the field As a result the expected performance ratio is typically between 80 and 90 This is in contrast with the performance index of a quantity which typically refers to the expected value of that quantity in a well operating plant at start up Thus the expected performance index during commissioning is typically 100 However the performance index may decrease over time due to soiling and degradation BE Page 29 of 29
7. S Kurtz J L B Cruz E Riley and C Hansen Weather corrected performance ratio National Renewable Energy Laboratory Golden Colorado NREL TP 5200 57991 Apr 2013 10 Shelton Honda Alex Lechner Sharath Raju and Ivica Tolich Solar PV System Performance Assessment Guideline for SolarTech San Jose State University San Jose California Jan 2012 BE Page 23 of 29 User manual for central monitoring and control functions 31 10 2015 ANNEX 1 GOOD PRACTICE FOR PV PLANT MONITORING A robust monitoring system is a prerequisite for the realization of a reliable predictive performance qualification The main purposes of a monitoring system are to measure the energy yield to assess the PV system performance and to quickly detect and identify operational malfunctions Many large PV systems use analytical monitoring to prevent economic losses due to operational problems in order to reach the highest possible final energy production Figure 10 shows the full energy conversion chain of a generic PV system according to 1 and the different parameters to be measured in real time Not all components of Figure 10 are present in every case actually back up sources energy storage and load are missing in most large PV systems Energy storage Back up Me sources lts lks L Load ce Prs Prs PL ANU gt ai Ne Voc Vs Vic G PV array lbc lac Grid Poc AC gt CONVERTER gt Tamb T moda
8. Theoretic Datasheet System Laboratory Tests Parameters Theoretic Performance Analyse Analyse Modelled System Parameters Modelled Performance Irradiation Analyse Measured Performance FIGURE 1 PV HEALTH SCAN METHODOLOGY Page 9 of 29 User manual for central monitoring and control functions 31 10 2015 2 5 2 DATA INTEGRITY CHECK Review the input data as follows Specification of location altitude total DC power number and type of inverters number type tilts and azimuths of PV modules array configuration of inverters Specification of latitude longitude altitude type tilt azimuth of sensor Completeness of data are all inverters sensors meters monitored data availability of time with data correctness of data format value range This is done semi automatically The PV Health Scan analysis toolbox contains standard functions to plot the data and identify gaps or significant inconsistencies in the data by a human expert 2 5 3 ON SITE SENSOR DATA CHECK Check the sensor data as follows Data from irradiance sensors can be checked with the Solar Sensor Check as described in Section 3 Data from temperature sensors should be checked visually based on a plot of temperature difference module minus ambient temperature over solar irradiance 2 5 4 PLANT PERFORMANCE CHECK The overall plant performance is reviewed as a function of time This is done by
9. metadata of the site and the PV arrays PV array specifications number of inverters and parallel strings per inverter and device types inverters modules and measurement devices e The plant has been monitored ideally in line with IEC 61724 1 Good practice for PV plant monitoring in line with 1 3 has been summarized in Annex 1 For a PV Health Scan the following parameters should have been recorded o Solar irradiance Ambient temperature PV module temperature Wind speed Total plant electricity generation at energy meter AC output power of each inverter AC voltage at each inverter DC input voltage of each DC input DC input current or power of each DC input e Data have been recorded with 15 minute time resolution with one minute sampling or faster e At least seven days of measurement to be reliable 000 0 The required accuracies and data quality check procedures are detailed in 1 3 Moreover a maintenance log should be available in order to reconcile the monitoring data with potential events like component repairs and replacements maintenance activities observed issues e g glass breakage delamination snail trails BE Page 8 of 29 User manual for central monitoring and control functions 31 10 2015 2 5 HEALTH SCAN PROCESS 2 5 1 APPROACH The PV Health Scan process is illustrated in Figure 1 The PV Health Scan analysis toolbox contains the functions required for running through the fu
10. Sw Yr Ya Reference yield Array yield Is Ic System losses Array capture losses PR FIGURE 10 ENERGY FLOW A PV SYSTEM AND PARAMETERS TO MONITORED The required accuracies and data quality check procedures are detailed in 1 3 The sampling interval for parameters which vary directly with irradiance should be 1 minute Page 24 of 29 User manual for central monitoring and control functions 31 10 2015 or less In most monitoring systems this sampled data is averaged over a recording period of 5 or 15 minutes in order to reduce storage needs The shorter the monitoring period is the more advanced the monitoring intelligence could be Mean in plane irradiation in Wh m is calculated by integrating the recorded irradiance in over the reporting period with recording interval expressed in hours Hpoa Tr Gpoa 1 The accuracy of the irradiance is today the most important factor impacting the accurate determination of the system performance 6 According to current standards 1 2 the in plane irradiance should be measured with a crystalline silicon reference device which should be calibrated and maintained in accordance with IEC60904 2 or IEC60904 6 However the use of a thermopile pyranometer is generally preferred above silicon reference devices for PV performance evaluation due to its superior accuracy 7 In practice large PV plants generally use
11. plotting energy based availability and performance ratio PR per inverter as indicated in Figure 2 The consistency of performance ratio is evaluated per inverter This is done by means of a box plot for all inverters immediately identifying outliers see example in Figure 3 140 Performance Ratio 56 FIGURE 2 PERFORMANCE Performance Ratio P0792 30 inverters Overall Availability Overall Performance ratio 120 100 80 40 fe n Apr jul Oct jan 2014 2015 STRING FAILURE VISIBLE BE Page 10 of 29 RATIO REPRESENTED IN A LINE PLOT WINTER SHADING AND ONE User manual for central monitoring and control functions 31 10 2015 Performance Ratio Box Plot 122 inverters 100 60 40 Performance Ratio 56 20 tt 6 7 18 9 0 25 ara 4 07 03 4 07 04 4 07 05 4 07 06 2014 07 07 2014 07 08 4 08 06 4 06 26 4 07 02 4 07 4 08 07 4 08 08 4 08 09 4 08 10 tt wt tt 2014 06 25 20 2014 07 13 2014 07 01 2014 07 11 2014 07 12 2014 07 14 2014 07 22 2014 07 23 2014 07 24 2014 07 ttt wt ot PERRERA 20 20 20 20 20 2014 07 09 2014 07 10 2014 07 15 2014 07 26 2014 07 27 2014 07 28 2014 07 29 2014 07 30 2014 07 31 2014 08 01 2014 08 02 2014 08 03 2014 08 04 2014 08 05 20 20 20 20 20 20 FIGURE 3 PERFORMANCE RATIO REPRESENTED IN A BOXPLOT SEVERAL STRING FAILURES UNTIL 2014 07 04 SHADING BY VEGETA
12. Modelied loss inverter not available mmm Unidentified residuals FIGURE 5 PERFOMANCE RATIO AND LOSSES OVER TIME FROM COMPARING MEASURED PERFORMANCE WITH THEORETIC AND MODELLED PERFORMANCE EE Page 12 of 29 User manual for central monitoring and control functions 31 10 2015 Sankey Diagram of Unavailability Losses for Year 2013 irradiation 100 Missing Data 2 Modelled loss 14 inverter not available 0 Park available 9 nterarray shading 196 b DC input not available Unidentified residuals 0 o FIGURE 6 PERFOMANCE RATIO AND LOSSES INTEGRATED OVER ONE YEAR FROM COMPARING MEASURED PERFORMANCE WITH THEORETIC AND MODELLED PERFORMANCE 2 5 7 IDENTIFY FAULTS AND ROOT CAUSES With the PV Health Scan analysis toolbox the final analysis and interpretation of the computed indicators is done by a human expert This expert will produce a standardized report following the steps of the PV Health Scan process as described above He or she documents each step and illustrates the indicators with the figures produced by the toolbox Then based on her experience she deduces the most probable root causes from the visual indicators over the operations history reflected by the available data EE Page 13 of 29 User manual for central monitoring and control functions 31 10 2015 3 OPERATIONAL PV HEALTH SCAN 3 1 PURPOSE The operational PV Health Scan is a computational tool to be used along with an ope
13. PERFORMANCE PLUS WP4 Deliverable 4 6 User Manual for Central Monitoring and Control functions Achim Woyte Karel De Brabandere Babacar Sarr Filip Andr n Thomas Strasser Johannes St ckl 3E AIT 31 10 2015 Final version Checked by Johannes St ckl Karel De Brabandere Geert Palmers This project has received funding from the European Union s Seventh Framework Programme for research technological development and demonstration under grant agreement No 308991 User manual for central monitoring and control functions 31 10 2015 USER MANUAL FOR CENTRAL MONITORING AND CONTROL FUNCTIONS Tools for Monitoring and Control D4 6 CONTENTS SUMMAR mannose mana 4 1 INTRODUCTION siering 6 2 PV HEALTH SCAN ANALYSIS TOOLBOX 7 2 7 2 7 2 3 A S 8 2 4 daa 8 2 5 HEALTH SCANIPROGESS snccdancedoambdakeuciutamidanhesaiiadnideeenghiaitagsedaabienaeikaeckotdeent 9 2 5 1 APPROACH 9 2 5 2 DATA INTEGRITY CHECK wrscevictiaceraetiosetistacvvectiuetian blesvvastlareven Shestheretacevecwts 10 2 5 3 ON SITE SENSOR DATA CHECK es cea a kaa cage eee 10 2 5 4 PLANT PERFORMANCE CHECK
14. Page 26 of 29 User manual for central monitoring and control functions 31 10 2015 PR 4 6 Y AVAILABILITY INDICATORS An important performance indicator is the availability of the PV plant The time based availability Ar represents the percentage of time during which the PV plant was producing and is expressed as the ratio between the duration of production activity tp expressed in hours in the recording period t with sufficient irradiation for a minimal production usually one month or one year expressed in total hours of daylight and the recording period as given by 54 7 While relatively easy to calculate the drawback of this time based indicator is that it does not allow for the calculation of the impact of unavailabilities on the overall system yield Hereto an alternative availability indicator is needed namely the energy based availability which takes into account the reference yield and therefore indicates the energy lost during times of unavailability The energy based availability is calculated as the ratio between the reference yield that has been converted to electricity Y cony and the total reference yield Y as given by Ay 8 In order to determine the impact of individual components on the system availability the availability of individual components is defined The time based availability of a component or block of components Acomp is calculated using the generic formu
15. TION VSIBLE UNTIL 2014 7 14 2 5 5 CURRENT AND VOLTAGE CHECK DC voltage and DC current are normalized to the expected value from simulation with theoretic system parameters and their relative deviations anomalies are plotted versus each other At the MPP their behaviour should follow the IV curve of the PV array If this is not the case this indicates a fault e Negative current anomaly with positive voltage anomaly or vice versa points towards bad MPP tracking for example in Figure 4 showing repeated voltage anomalies of up to 20 combined with current anomalies of down to 50 e Negative current anomaly with normal voltage points at string faults e Negative voltage anomaly with normal current points at faults causing reduced module voltage e g bypass diode failures e Negative voltage and current anomaly may point towards a reduction in fill factor hence degradation effects of the PV array resulting e g in increased series resistance 2 5 6 LOSS ATTRIBUTION Losses are attributed to the different energy conversion steps happening within the PV plant This is done by comparing the measured performance with a the theoretic performance and b the modelled performance based on modelled system parameters Applied to the time series of monitoring data as illustrated in Figure 5 this allows interpreting deviations from the expected PR and allocating them to the different energy conversion steps Integration of the loss and performan
16. but are not yet ready to be offered to the market This is planned for the course of 2016 Consequently the present user manual should be read as a draft specification document for the different tools It allows for having the product specifications reviewed by potential customers from different target groups and iterate on them for the final product development Page 6 of 29 User manual for central monitoring and control functions 31 10 2015 2 PV HEALTH SCAN ANALYSIS TOOLBOX 2 1 PURPOSE The PV Health Scan analysis toolbox is a computational library written in the programming language Python It mainly contains functions for data analysis and visualisation of PV monitoring data The toolbox is applied to check the integrity of PV plants in the past i e from the start of their operating history until today In case of faults or reductions in performance the toolbox allows to drill down to inverter and string level in order to understand their root causes Typical phenomena that can be identified are e g degradation of PV string currents over time sub optimal maximum power point MPP tracking and outages of smaller subsections of PV plants All these phenomena can be evaluated over time Users may want to perform this check for several reasons e Provisional or final acceptance testing as part of PV plant commissioning when the ownership and warranties for a new PV plant are transferred from the installer to the new owner the
17. but its value is influenced heavily by local weather as well as system characteristics such as orientation of the PV modules Nevertheless it can be used to monitor several PV sub systems operating under the same conditions by monitoring the deviation between them BE Page 25 of 29 User manual for central monitoring and control functions 31 10 2015 ANNEX 2 PERFORMANCE INDICATORS Performance indicators describe the performance of the system by comparison with a theoretical reference value or other systems They are evaluated during design during project commissioning as part of every due diligence investigation and on a continuous basis in order to detect major design operating or maintenance issues These indicators often form a key factor in contracts such as operation and maintenance contracts product and system performance warranties and liabilities PERFORMANCE RATIO A well known and often used indicator is to compare the system yield with the reference yield Y through the system performance ratioPR The system PR captures differences due to local weather different tilts and or orientations prow 4 The system PR represents the overall effect of losses on the output due to temperature incomplete utilisation of the irradiation and system component inefficiencies or failures and thus the health of the system Continuous monitoring of the system PR allows detecting when the system is performing significantly bel
18. ce measures over longer time spans e g weeks or months will reduce noise and thus improve the specificity of the Health Scan while maintaining its sensitivity In other words the longer the integration time for analysis EE Page 11 of 29 User manual for central monitoring and control functions 31 10 2015 the higher is the reliability of the results For monthly integration a Sankey diagram illustrating the energy and losses of the PV plant is the adequate visualisation as it allows for comparison with the energy losses as identified during the project development phase Figure 6 35 NG 120 30f 22722272 S25 90 5 20 15 60 u 10 Os 30 250 300 350 400 450 500 0 Z s 2856 T py a azi 11 34 2 o f 60 5 0 0 A 3 20 20 tempCu 0 00 90 TEk tempCi 0 00 8 40 ae Si 40 i 120 250 300 350 400 450 500 0 40 30 20 10 0 10 20 30 DC Voltage V DC Voltage anomaly 56 EP FIGURE 4 DC CURRENT VERSUS DC VOLTAGE AS ABSOLUTE VALUES AND NORMALIZED ANOMALIES 140 Performance loss cause analysis 120 100 Performance ratio 3 10 07 10 21 11 18 12 02 2013 05 20 2013 06 03 2013 06 17 2013 07 01 2013 08 26 2013 09 09 2013 09 23 2014 01 27 RR Modelled Performance Interarray shading DC Input not available Measured Performance Park not available Missing Data me
19. e present document serves as preliminary user manual for the central monitoring and control features developed and implemented within Performance Plus It covers the following four specific tools for monitoring and control of photovoltaic PV plants that go beyond common practice in PV plant operations today e The PV Health Scan analysis toolbox Monitoring intelligence for analysis of historical PV plant monitoring data e Operational PV Health Scan Monitoring intelligence for analysis close to real time based on PV data streams from operational PV plants e Solar Sensor Check A tool for verifying the measurements of solar irradiance sensors e The IEC 61850 gateway for the control and remote management of PV plants The different tools currently exist as prototypes but are not yet ready to be offered to the market This is planned for the course of 2016 Consequently the present user manual should be read as a draft specification document for the different tools It allows for having the product specifications reviewed by potential customers from different target groups and iterate on them for the final product development The PV Health Scan analysis toolbox is a computational library It mainly contains functions for data analysis and visualisation of PV monitoring data The toolbox is applied to check the integrity of PV plants in the past until today In case of faults or reductions in performance the toolbox allows to drill down to inv
20. erter and string level in order to understand their root causes Due to its complexity and the multitude of use cases the PV Health Scan analysis toolbox is currently not offered as a software Instead it is applied by 3E s consultants to offer a semi automated and standardized service at adequate costs The document further describes user stories and data requirements It then details the process steps for a semi automated analysis with the toolbox The operational PV Health Scan is a computational tool to be used along with an operational monitoring platform It automatically runs through the process steps for the PV Health Scan analysis toolbox The results are then interpreted by the computer The operational PV Health Scan detects deviations from the expected performance and availability by means of parametric models of the PV plant s components and the variation of their model parameters over time It then returns recommendations on the most probable root causes for the deviations The operational PV Health Scan is offered to PV asset managers and PV operation and maintenance O amp M contractors It is made available as a scan on request and as subscription for periodic and automatic execution within the monitoring tool of the customer The Solar Sensor Check is a computational tool It is applied to check the integrity of solar irradiance sensors based on measurements and indicate if their measurements are wrong or imprecise As such i
21. just the maximum generation level of an inverter based system Its implementation as an IEC 61499 FB in the proposed IEC 61850 gateway device is depicted in Figure 9 b a DRCC Data object Common Explanation name data class Common LN information Settings MaxWLimPct ASG Percent of reference active power watts as maximum allowed watts output OutPFSet ASG Setpoint for maintaining fixed power factor b EVENT EVENT DRCC REAL WaxWLimPct STRING REAL OutPFSet FIGURE 9 IEC 61850 PV INVERTER FUNCTION IMPLEMENTED AS IEC 61499 FUNCTION BLOCKS 5 3 PROTOTYPICAL IMPLEMENTATION In the following a low cost solution for such a gateway controller using mainly open source software is presented 5 5 3 1 HARDWARE PLATFORM The implementation is carried out using a Raspberry Pi Model B controller 700 MHz 32 bit microprocessor 512 MB with the embedded open source operating system Raspbian based on Debian GNU Linux This low cost embedded controller provides several communication ports i e 1 built in Ethernet port 4 built in USB ports as well as general purpose digital I Os Moreover several extension boards exist providing additional interfaces to the controller e g serial communication over RS 232 RS 485 The USB ports can be also used to connect communication adapters as well e g USB to Ethernet Therefore this device is an ideal platform for a low cost automation platform solutio
22. la below bony WEE k Where e N is the total number of subsystems composing the component or block of components e the cumulative number of hours during which the component or block of components should have been operating e Ty the cumulative number of hours of unavailability of the sub system composing the component or block of components The accuracies of the indicators presented in this document depend of course on the performance of the data acquisition system itself The availability of monitored data allows assessing the quality of the data acquisition system It is expressed as the ratio of the duration of monitoring activity over the reporting period t as given by Ay 10 Page 27 of 29 User manual for central monitoring and control functions 31 10 2015 Sensors should be calibrated at least every two years except specific manufacturer recommendations In addition it is recommended that the quality of the irradiation sensors is assessed regularly by monitoring the difference in irradiation assessed by different on site sensors or by an on site sensor and the satellite estimation TEMPERATURE CORRECTED PERFORMANCE RATIO In 9 10 temperature corrected performance ratios are proposed in order to remove the seasonal dependency of PR due to temperature Y PRre 7 11 Where the temperature corrected reference yield is defined as Tet De Groa 1 02 Teen
23. monitored by SynaptiQ If meteorological data are missing they will be imported from a satellite weather service 3 5 RESULTS The operational PV Health Scan returns results in the form of a table as illustrated in Table 1 BE Page 15 of 29 User manual for central monitoring and control functions 31 10 2015 TABLE 1 FICTIVE SAMPLE OUTPUT OF THE OPERATIONAL PV HEALTH SCAN FOR A PORTFOLIO OF PV PLANTS 1D0004 Easington 190004 Easington 1D0243 San Martino 1D0243 San Martino 1D1245 Varese 1D1245 Varese 1D2347 Vidauban 1D2347 Vidauban 3 6 SALES CHANNEL Jul 15 Aug 15 Jul 15 Aug 15 Jul 15 Aug 15 Jul 15 Aug 15 Degradation String disconnected PID Soiling grid overvoltage Root Cause unclear Everything fine Unexpected shadow full generator String 12 3 Stringit 2 12 full generator all inverters i entire plant String 1 3 18 1 9 7 49 10 0 00 77496 Error Error Error Error Error Error Notification Error The operational PV Health Scan will be offered to PV plant operators acting as asset managers or O amp M contractors Customers of SynaptiQ can use the operational PV Health Scan on request or as a periodic check Customers who use another monitoring solution than SynaptiQ can be served on short term by importing their plant configuration and monitoring data into the database of SynaptiQ BE Page 16 of 29 User manual for central monitoring and control functions
24. n which can be acquired for a very low price In principle also other comparable low cost embedded controllers can be used EE Page 21 of 29 User manual for central monitoring and control functions 31 10 2015 5 3 2 SOFTWARE PLATFORM In order to implement the IEC 61850 compliant functions as well as higher level control functions of the DER device the IEC 61499 compliant open source software package 4DIAC is used It provides an engineering programming tool called 4DIAC IDE and an execution platform for embedded controllers called FORTE Furthermore the automation platform integrates the open source stacks libIEC61850 and libmodbus and represents their capabilities as IEC 61499 compliant FBs With this software setup the above presented architecture and concept can be realized Since commercial off the shelf PV inverters often provide a proprietary Modbus based remote control interface the implemented gateway controller prototype was equipped with a corresponding Modbus library as stated above BE Page 22 of 29 User manual for central monitoring and control functions 31 10 2015 REFERENCES 1 IEC 61724 edi 0 Photovoltaic System Performance Monitoring Guidelines for Measurement Data Exchange and Analysis International Electrical Commission 1998 2 G Blaesser and D Munro Guidelines for the Assessment of Photovoltaic Plants Document A Photovoltaic System Monitoring Commission of the European Communities Joint Resea
25. nce and availability by means of parametric models of the PV plant s components in the same way as it is illustrated in Figure 5 above It then returns recommendations on the most probable root causes for the deviations Users may want to perform the operational PV Health Scan monthly weekly or even daily for several reasons e Benchmark the performance of plants within their portfolio for controlling and reporting purposes e Identify performance reductions and reductions in availability fast understand the root cause and foresee a maintenance intervention with optimal dispatch of maintenance technicians and spare parts Document the root causes for performance reductions and reductions in availability in order to decide on remediation actions which may be of technical e g recabling or legal nature e g warranty claims The operational PV Health Scan is offered to PV asset managers and PV O amp M contractors It is made available as a scan on request and as subscription for periodic and automatic execution within the monitoring tool of the customer The first release will read PV Page 14 of 29 User manual for central monitoring and control functions 31 10 2015 monitoring data from 3E s monitoring service SynaptiQ Subsequent releases will have open interfaces in order to be able to process monitoring data from different monitoring systems as well 3 2 SCOPE The operational PV Health Scan is focussed on PV plan
26. nctional blocks as indicated While the theoretic system parameters originate from data sheets of the applied PV components the modelled system parameters originate from regression models of DC MPP voltage and current Consequently by applying environment data measured at the site to a model of the PV plant the theoretic and modelled performance indicators are computed respectively The main performance indicators as used by PV operators and asset managers are availability performance ratio with and without temperature correction and energy performance index For their definitions we refer to Annex 2 The PV Health Scan process consists of the following steps 1 Check data integrity configuration data completeness and availability Check data from on site sensor e g with a plausibility check or with the Solar Sensor Check as described in Section 3 3 Check plant performance first plant level go down to inverter level DC input and finally if available string level e Chose regression models for MPP voltage and current as a function of the environment irradiance temperature wind speed e Train coefficients of the regression models with operational data e Analyse differences between theoretic and trained model parameters e Analyse differences between measured modelled and theoretic performance 4 Attribute losses to components and physical characteristics 5 Identify faults and root causes Configuration
27. nth of measurements to be reliable e At least hourly granularity recorded with one minute sampling or faster 4 4 USER INTERFACE The Solar Sensor Check is operated through an interactive user interface The interface allows the user to specify its input data and launch a check for a set of sensors at one or several sites The user specifies the following input data e Desired start and end dates e Metadata geographical co ordinates orientation amp azimuth sensor type EE Page 17 of 29 User manual for central monitoring and control functions 31 10 2015 A time series of measured irradiance data from the sensor under study The tool uses satellite data from 3E s satellite service as reference for the coordinates of the site under study 4 5 RESULTS The Solar Sensor Check returns results in the form of a table as illustrated in Table 1 On top of the table with the estimated error per plant the user receives a figure that illustrates the error The user can use this figure for documenting the error in case a more detailed report is requested The example in Figure 7 shows the measured and satellite irradiance for a pyranometer at the plant Easington UKt The Solar Sensor Check identified a time shift of the data logger s clock by 3 hours 4 6 SALES CHANNEL Currently 3E does not ship the Solar Sensor Check as software but offers it through its technical consultants and support staff Direct access to external partie
28. one or more thermopile pyranometers for performance evaluation whereas medium size PV plants use a crystalline silicon reference device and small PV plants mostly rely on satellite estimations It is important that the choice of the reference device has an important impact on the calculated performance ratio see below Typically performance ratios are 2 4 higher when using a crystalline silicon device as a reference compared to a pyranometer 6 8 The in plane irradiation is used to determine the reference yield which is key in determining the performance ratio of the PV plant The reference yield Y is calculated by dividing the mean in plane irradiation Hpo in Wh m by the module s reference in plane irradiance Gsrc 1000 W m Y Hpoa 2 Gsrc The reference yield represents the number of hours over the recording period during which the solar radiation would need to be at the reference in plane irradiance level in order to contribute the same incident energy as was monitored The system yield Y is the ratio between the energy provided to the utility grid E4c in kWh and the nominal DC peak power Perc installed in kW y Ec 3 The system yield Y represents the number of hours over the recording period during which the system would need to operate at its nominal peak power to contribute the same energy to the utility grid as was monitored The system yield could be used to compare PV plants
29. ow its reference The system performance ratio allows comparing PV systems of different configurations and at different locations During the design phase overall system losses are determined and an expected average yearly system performance ratio is calculated The real life system PR could be compared to this design system PR and trigger an alert when a certain deviation occurs The design system PR should take into account the yearly degradation of the system However deviations between real life and design system PR will be present as real life system PR varies with irradiation temperature and solar incident angle whereas the design system PR is an average value over the year The same performance indicators can be used at component level to assess the performance of the different PV components most notably the array Equivalent to the system yield the array yield Y is defined as the array energy output Epc in kWh per kW of installed PV array Porc Epc 5 y 4 Porc Thus the array yield Y represents the number of hours over the recording period during which the array would need to operate at its rated output power Ppc to contribute the same energy to the system as was monitored The comparison of array yields with identical orientation technology and configuration in a PV system presents a relatively easy and effective way for the detection of single dysfunctional arrays The array performance ratio PR is defined as BE
30. rational monitoring platform It automatically runs through the process steps described in Section 2 5 for the PV Health Scan analysis toolbox The results are then interpreted by the computer 3E is planning to offer the operational PV Health Scan as a plugin to its SynaptiQ monitoring platform It is applied to check PV plants in operation for faults and degradation We define the operational PV Health Scan as follows e The operational PV Health Scan is the generation and delivery of technical and business intelligence for a PV plant or portfolio of PV plants from measured data e This intelligence should go above current practice particularly by linking the computed KPIs to underlying root causes thus generating actionable insights towards increased profitability This should work in an initially semi automated and finally fully automated way Typical root causes for reduced availability and performance are soiling of the PV array string faults bypass diode faults optical degradation of PV modules potential induced degradation PID sub optimal MPP tracking or irregular grid disconnection The most obvious failures e g inverter shutdown due to insulation faults or inverter failure are usually detected by the inverter and do not need further explanation Therefore the operational PV Health Scan focusses on those faults that are more complex to detect The operational PV Health Scan detects deviations from the expected performa
31. rch Centre Ispra Italy EUR 16338 EN Issue 4 2 June 1993 1995 3 G Blaesser and D Munro Guidelines for the Assessment of Photovoltaic Plants Document B Analysis and Presentation of Monitoring Data Commission of the European Communities Joint Research Centre Ispra Italy EUR 16339 EN Issue 4 1 June 1993 1995 4 F Andr n R Brundlinger and T Strasser IEC 61850 61499 Control of Distributed Energy Resources Concept Guidelines and Implementation Trans Energy Convers vol 29 no 4 pp 1008 1017 Dec 2014 5 C Zanabria F Andr n F Leimgruber R Bruindlinger and T Strasser A Low Cost Open Source based IEC 61850 61499 Automation Platform for DER Components presented at the IEEE PowerTech International Eindhoven The Netherlands 2015 p 3 6 M Richter K De Brabandere and J Kalisch Best Practices on Uncertainty in PV Modelling Performance Plus WP2 Deliverable D2 4 Dec 2014 7 A Woyte M Richter D Moser N Reich M Green S Mau and H G Beyer Analytical Monitoring of Grid connected Photovoltaic Systems Good Practice for Monitoring and Performance Analysis IEA PVPS Report IEA PVPS 113 03 2014 Mar 2014 8 N Reich B Mueller A Armbruster W G J H M van Sark K Kiefer and C Reise Performance ratio revisited is PR gt 90 realistic Prog Photovolt Res Appl vol 20 no 6 pp 717 726 2012 9 T Dierauf A Growitz
32. s via a web interface has been considered This will be implemented when interest from customers in this offer can be confirmed TABLE 2 SAMPLE OUTPUT OF THE SOLAR SENSOR CHECK FOR A PORTFOLIO OF DIFFERENT PLANTS vt x 1D0004 Easington Pyr 1 Jul 15 Clock setting 3h Error 1D0004 Easington Pyr1 Aug 15 everything fine 0 Notification 1D0243 San Martino Pyr1 Jul 15 Orientation 6 3 deg Error 1D0243 San Martino Pyr1 Aug 15 Orientation 5 8 deg Error 1D1245 Varese Si 2 Jul 15 Calibration Slope 17 Error 1245 Varese Si 2 Aug 15 Root Cause unclear Wy Error 2347 Vidauban Pyr 2 Jul 15 Tilt Error 3 deg Error 2347 Vidauban Pyr 2 Aug 15 Calibration Offset 8 3 Error Fictive site name while the data originate from an existing plant the plant name has to be changed for reasons of confidentiality BE Page 18 of 29 User manual for central monitoring and control functions 31 10 2015 1000 Month 2015 07 800 irradiance 400 200 00 00 00 06 00 00 12 00 00 18 00 00 00 00 00 06 00 00 12 00 00 18 00 00 datetime FIGURE 7 WRONG CLOCK SETTING IDENTIFIED AND ILLUSTRATED BY COMPARISON BETWEEN SENSOR RED AND SATELLITE IRRADIANCE GREEN Page 19 of 29 User manual for central monitoring and control functions 31 10 2015 5 REMOTE CONTROL INTERFACE FOR PV PLANTS 5 1 PURPOSE To provide a flexible and cost effective solution for upgrading existing PV inverters with state of the art communica
33. se occasions require a thorough review of the operating history of the PV plant The PV Health Scan analysis toolbox serves to confirm the correct installation and operation and identify non conformities through a standardized and efficient process e Due diligence on older plants as part of mergers and acquisitions whenever an existing PV plant is acquired on the secondary market the investor will perform a due diligence check The use case is comparable to the case of commissioning above but it may cover a longer operating history In this case the technical due diligence check will tell the investor whether a plant is worthwhile investing which actions to improve availability or performance may be recommended and which yield improvements these may yield e Operators of PV plants regularly report business and yield figures as well as operating history and maintenance to their management and shareholders usually including a comparison with expected yield figures from the initial business plan for the plant The PV Health Scan analysis toolbox allows for a regular automated check with a standardized and efficient approach In case of yield reductions it allows for identifying the root causes and taking adequate mitigation measures where possible Due to its complexity and the multitude of use cases the PV Health Scan analysis toolbox is currently not offered as a software Instead it is applied by 3E s consultants to offer a semi automated and
34. standardized service at adequate costs 2 2 APPLICATION With the PV Health Scan analysis toolbox the user checks the configuration availability of data data integrity sensor accuracy plant availability plant performance and component performance Moreover it enables the skilled user to draw conclusions on possible root causes BE Page 7 of 29 User manual for central monitoring and control functions 31 10 2015 Ideally the user runs a full health check seven days after every significant event This could be at the occasion of commissioning plant hardware changes component replacement and or repair inverter and logging software upgrades configuration changes change of owner or operation and maintenance O amp M contractor etc The scan should afterwards be repeated monthly to identify evolutions For a one time check of a plant with a longer history of operation the user will run the health check once and compute performance indicators per month Section 2 5 below describes the different steps to be executed with the PV Health Scan analysis toolbox to be executed by a skilled user in order to perform a semi automatic health scan 2 3 SCOPE The PV Health Scan analysis toolbox is focussed on small commercial to large size PV plants 2 4 REQUIREMENTS The PV Health Scan analysis toolbox can be applied if the following requirements are fulfilled e PV plant configuration data are available based on a template requesting
35. t is a precondition to the operational PV Health Scan that also has business value on its own BE Page 4 of 29 User manual for central monitoring and control functions 31 10 2015 The Solar Sensor Check detects the most important faults and imprecisions based on measurements It then returns recommendations on the most probable root causes The Remote Control Interface for PV plants has been developed to introduce a standard compliant and open source based environment together with a low cost controller platform for upgrading off the shelf PV inverts with Smart Grid functions Page 5 of 29 User manual for central monitoring and control functions 31 10 2015 1 INTRODUCTION This user manual serves as public documentation for the monitoring and control functions developed and implemented in Performance Plus It covers the following four specific tools for monitoring and control of photovoltaic PV plants that go beyond common practice in PV plant operations today e The PV Health Scan analysis toolbox Monitoring intelligence for analysis of historical PV plant monitoring data e Operational PV Health Scan Monitoring intelligence for analysis close to real time based on PV data streams from operational PV plants e Solar Sensor Check A tool for verifying the measurements of solar irradiance sensors e The IEC 61850 gateway for the control and remote management of PV plants The different tools currently exist as prototypes
36. tion interfaces like IEC 61850 a gateway device has been developed using an already existing proprietary remote control interface of the inverter The concept and the implementation of the gateway controller are described below 5 2 ARCHITECTURE AND CONCEPT A gateway device can be used as an interface between an inverter system and the Distribution Management System DMS Supervisory Control and Data Acquisition System SCADA of the Utility Operator UO or Distribution System Operator DSO Figure 9 provides an overview of the proposed IEC 61850 gateway concept 4 It connects the PV inverter and the DSO control centre Thus the gateway is interacting with the UO DSO according to the IEC 61850 interoperability approach The gateway device therefore acts as a kind of translator of the IEC 61850 commands coming from the DSO control centre into the proprietary data format of the PV inverter To achieve this translation a Function Block FB oriented IEC 61499 application designed for distributed automation systems is used This application always consists of at least two FBs One FB handles the IEC 61850 communication with the DSO and the other FB handles the proprietary communication with the inverter see Figure 8 The rest of the FB network decides how the IEC 61850 functions are translated into commands understood by the PV inverter i e the actual gateway functionality DSO SCADA
37. ts with analytical monitoring as specified in Annex 1 and the relevant standards and guidelines 3 3 REQUIREMENTS Data requirements are in principle the same as for the PV Health Scan analysis toolbox see Section 2 4 If no reliable meteorological data are available irradiance in the plane of the PV array and ambient temperature from a satellite weather model may be sufficiently precise On the DC side voltage and power or current measurements are a must ideally on string level but at least on inverter level If currents are available only as aggregated values per inverter the root cause analysis will accordingly be less specific and sensitive to faults occurring in individual strings 3 4 USER INTERFACE The operational PV Health Scan is operated through an interactive user interface The interface allows the user to specify a plant with monitoring data from SynaptiQ for a complete check The user specifies the following input data e Periodicity one time weekly or monthly Desired start date in case of a one time check Plant identifier The tool reads the monitoring data from 3E s monitoring service SynaptiQ This way the operational PV Health Scan can also be offered as a one time check to customers who don t use SynaptiQ for monitoring This is done by first importing the configuration and monitoring data into the database of SynaptiQ Afterwards the operational PV Health Scan can be applied as if the plant would be
Download Pdf Manuals
Related Search