B-2 RFQ attachment - Statement of Work-final.pdf

STATEMENT OF WORK

Title: Photovoltaic (PV) Modeled Versus Measured Performance Evaluations

I. BACKGROUND

The mission of the National Institute of Standards and Technology (NIST) is to promote U.S.

innovation and industrial competitiveness by advancing measurement science, standards, and technology in ways that enhance economic security and improve our quality of life. The Energy and Environment Division of the Engineering Laboratory at NIST consists of four Groups, one being the Heat Transfer and Alternative Energy Systems Group. This group collects performance data, develops measurement methods, and validates computer models used to predict the energy performance of photovoltaics and other alternative energy production systems in building applications.

In addition, the Heat Transfer and Alternative Energy Systems Group provides measurement services and standard reference materials to assess the thermal performance of insulation;

develops novel methods and tools to make in-situ measurements of the energy consumption in buildings and the energy efficiency of building components; develops and assesses test methods for water heaters, pool heaters, and other heating products; and conducts research to improve knowledge and standards for water use and delivery in buildings.

II. PURPOSE

NIST has collected a sizeable dataset consisting of short-term characterization measurements on three solar modules and long-term field measurements on several other solar modules. Although none of the modules used for the short-term characterizations are among the modules used for the field measurements, all of these modules are the same make and model. NIST seeks to contract with a company that can use these datasets to address specific research questions tied to algorithms for predicting field performance of solar modules and arrays, and the associated measurement and modeling uncertainties in doing so.

III. PROJECT OBJECTIVE

As an overall goal, the NIST Engineering Laboratory seeks to maximize the use of Five Datasets described below for the purpose of evaluating algorithms that predict the installed performance of solar photovoltaic (PV) modules and arrays. This procurement, however, is limited to evaluating two specific algorithms in a narrowed-scope, first step manner. One specific algorithm that must be evaluated is the one specified in Part 3 of IEC Standard 61853; the second algorithm must be one that meets the definition for an “IEC 61853-calibrated diode-based model” (see below section IV.A). Both algorithms must be evaluated from comparing their predictions for two or more field-installed module’s power output versus the measured values over a wide range of operating conditions. This two-algorithm evaluation task constitutes Task #3 of this procurement.

The utility of these NIST datasets for evaluating PV device performance algorithms depends on the measurement uncertainties associated with many of the parameters recorded in these datasets.

One objective of this current procurement is for the vendor to generate substantiated recommendations on a statistical-based approach for estimating the uncertainty in a PV module’s performance prediction when using each algorithm: the vendor-selected IEC 61853-calibrated diode-based model and Part 3 of IEC 61853. Although beyond the scope of this procurement, the eventual and significant goal will be to develop and demonstrate a detailed procedure for estimating the uncertainty when using each of the PV performance algorithms. The prerequisite work to this eventual goal, as applied to each of the modeling algorithms, is addressed separately in this procurement as Task # 1 and Task #2, respectively. These investigations are judged as significant due to the complication added from having to evaluate the two algorithms based on the field performance of surrogate modules. More discussion on this complication is provided in the next section.

Added Complication:

NIST Dataset #1 contains the results from testing three modules in accordance with the first two Parts of IEC Standard 61853. This IEC standard defines a set of tests used to characterize and “map” the performance of a solar module over a range of representative operating conditions.

One of the key hurdles to effectively using Dataset #1 is to estimate how much uncertainty is introduced from not having tested one or more of the field modules of Datasets #2 (of which, Dataset #5 is a subset) and Dataset #3 as per IEC 61853. Consequently, some additional sources of uncertainty specifically tied to using Dataset #1 for analyzing the performance of installed, surrogate modules include:

• Two of the three modules tested as per IEC 61853 were acquired from a different shipment.

These modules were shipped within several weeks of the much larger campus array shipment and sourced from the same manufacturing plant.

• These same two modules were aged outdoors as part of the campus rooftop array but while being open circuited. The modules were installed at the same time as the rooftop array (in 2012) and then removed in the Spring of 2016, prior to being sent for testing.

• Only one of the modules tested as per IEC 61853 was taken from the large module shipment used to construct the three campus arrays. This one module was stored indoors until it was tested.

• The campus arrays were constructed and commissioned in 2012, although the first year of complete data occured in 2015 (Dataset #2). The instrumentation needed to generate Dataset #5 was not installed and fully operational until the summer of 2016. And, finally, the four modules of Dataset #3 were installed several weeks before data collection commenced in late 2014.

IV. SCOPE OF WORK – GENERAL

The available data the contractor shall analyze and perform PV analysis of is from NIST datasets that are listed below. For the below listed datasets, the same model of PV module was used: a mono-Si, 60-cell module that is representative of early-to-mid 2010’s technology.

• Dataset #1: results from testing the one model of PV module as per IEC Standard 61853, Parts 1 and 2. A total of three modules, all the same model, were tested to provide a

(very) small test population. The testing was conducted by a private laboratory that is qualified for conducting the suite of IEC 61853 tests.

• Dataset #2: multiple years of minutely data from three grid-tied solar systems located on the NIST-Gaithersburg campus. These systems can be differentiated based on their arrays: a ground-mount array, a parking lot canoy array, and a commerical building rooftop array. Information on these sites and the collected data can be obtained at https://www.nist.gov/el/energy-and-environment-division-73200/heat-transfer-alternative-energy-systems/photovoltaic and https://nvlpubs.nist.gov/nistpubs/TechnicalNotes/NIST.TN.1896.pdf.

• Dataset #3: multiple years of minutely data from four individual solar modules that are all connected to a single multi-tracer device which performs max power tracking while recording that power and integrating it over time. In addition, the multi-tracer device sweeps an IV curve for each module every minute. The four modules are mounted in the same orientation as used in the three grid-tied arrays. Notably, the parking lot canopy array is evenly divided between modules in two different orientations (i.e., all modules are tilted down 5 degrees from horizontal but half face east and the other half face west).

• Dataset #4: multiple years of minutely data from a research-grade meteorology station that is installed on the upper roof of a NIST general purpose building. Information on this station and the instruments deployed is available at thttps://nvlpubs.nist.gov/nistpubs/TechnicalNotes/NIST.TN.1913.pdf.

• Dataset #5: over 18 months of minutely data from 96 IV tracers that are connected to one module in each of the 96-series strings that collectively create the ground-mount array.

These IV tracers measure the operating current for each string of modules every few seconds, with the average being calculated and saved every minute. These tracers also sweep an IV curve for the one, pre-selected module within each string every 5 minutes.

A. DEFINITIONS

• IEC 61853: means the four part standard developed and maintained by the International Electrotechnical Commission (IEC) that covers procedures for characterizing the electrical performance of solar photovoltaic modules and then estimating their energy generation using the same characterization data. The shared title of the four-part standard is “Photovoltaic (PV) module performance testing and energy rating”; the extended portion of the standard’s title for the four parts are:

o Part 1 – Irradiance and temperature performance measurements and power rating o Part 2 – Spectral responsivity, incidence angle and module operating temperature measurements o Part 3 – Energy rating of PV modules o Part 4 – Standard reference climatic profiles

• Surrogate module: means any module who performance is reflected in Datasets #2 to

#5. All surrogate modules were never tested as per IEC 61853; however, surrogate https://www.nist.gov/el/energy-and-environment-division-73200/heat-transfer-alternative-energy-systems/photovoltaic https://www.nist.gov/el/energy-and-environment-division-73200/heat-transfer-alternative-energy-systems/photovoltaic https://nvlpubs.nist.gov/nistpubs/TechnicalNotes/NIST.TN.1896.pdf https://nvlpubs.nist.gov/nistpubs/TechnicalNotes/NIST.TN.1913.pdf modules are the same model of module as the three Dataset #1 modules that were tested as per IEC 61853.

• IEC 61853-calibrated diode-based model: means an alogorithm that uses an equivalent circuit approach to model the electrical characteristics of a solar cell or module. The model typically comes in the form of multiple equations that collectively approximate the operations of a solar device as an electrical circuit containing one or two diodes, resistance elements, and a source. The model typically includes five or more charactistic parameters. For this procurement, the parameters must be determined exclusively from a general description of the module (e.g., number of cell in series) and data obtain from conducting the tests specified in Part 1, and optionally in Part 2, of IEC Standard 61853.

V. SPECIFIC TASKS

The Contractor shall provide a Firm Fixed Price quotation for the following work to be performed. The Contractor shall accomplish each of the following tasks/tasks. The contractor shall provide all labor, supervision, materials, equipment, and facilities (unless otherwise stated herein)

1. Task 1 - Title: Substantiated recommendation for estimating the expanded uncertainty associated with using Dataset #1 to predict the power generation of surrogate field modules using an “IEC 61853-calibrated diode-based model”.

Task 1 - Quantity: The contractor must consider and report the pros and cons of using at least two different statistical methods for the eventual uncertainty estimates while substantiating their specific recommendation. One of the statistical methods that must be considered is an approach conducted in accordance with the ISO/IEC Guide 98-3:2008, Uncertainty of Measurement – Part 3: Guide to the Expression of Uncertainty in Measurement.

Task 1 - Specifications:

1.1. Develop the recommendation based on the eventual goal of determining the expanded uncertainties for at least one surrogate module from Dataset #3 or Dataset #5, while using the data from at least one IEC 61853-tested module from Dataset #1.

The method will eventually be used to determine expanded uncertainties for predicted power values that correspond to a wide range of operating conditions.

1.2 The procedure must address the specific issues tied to the additional sources of measurement uncertainty created from predicting the power generation of a surrogate module(s).

Task 1 - Deliverables:

1.3 Provide a written summary of the two or more statistical approaches investigated, and then detail the pros and cons of each method for the specific future application of estimating the uncertainty of the performance predictions of the NIST surrogate modules if using the IEC 61853-calibrated diode-based model. Also, state any simplifying assumptions and any supportive rationale for making those assumptions when considering each statistical approach.

1.4

2. Task 2 - Title: Substantiated recommendation for estimating the expanded uncertainty associated with using Dataset #1 to predict the power generation of surrogate field modules using the algorithm in IEC 61853 – Part 3.

Task 2 - Quantity: The contractor must consider and report the pros and cons of using at least two different statistical methods for the eventual uncertainty estimates while substantiating their specific recommendation. One of the statistical methods that must be considered is an approach conducted in accordance with the ISO/IEC Guide 98-3:2008, Uncertainty of Measurement – Part 3: Guide to the Expression of Uncertainty in Measurement

Task 2 - Specifications:

2.1. Develop the recommendation based on the eventual goal of determining the expanded uncertainties for at least one surrogate module from Dataset #3 or Dataset #5, while using the data from at least one IEC 61853-tested module from Dataset #1.

The method will eventually be used to determine expanded uncertainties for predicted power values that correspond to a wide range of operating conditions.

2.2 The procedure must address the specific issues tied to the additional sources of measurement uncertainty created from predicting the power generation of a surrogate module(s).

Task 2 - Deliverables:

2.3 Provide a written summary of the two or more statistical approaches investigated, and then detail the pros and cons of each method for the specific future application of estimating the uncertainty of the performance predictions of the NIST surrogate modules if using the IEC 61853 – Part 3 algorithm. Also, state any simplifying assumptions and any supportive rationale for making those assumptions when considering each statistical approach.

3. Task 3 - Title: Comparison of measured versus predicted power generation as determined using two algorithms: IEC 61853-3 and an IEC 61853-calibrated diode-based model.

Task 3 - Quantity: For each predictive algorithm, document the results of the measured versus predicted power generation comparisons for two or more field-deployed surrogate modules from Dataset #3 or Dataset #5 over a wide range of days.

Task 3 - Specifications:

3.1. The comparisons of predicted versus measured power generation for each predictive algorithm shall be conducted using the same sets of daily data.

3.2 Comparisons shall be made using minutely data from each selected day.

3.3 Generate plots and calculated statistical parameters that are helpful for evaluating each algorithms predictive performance on an instantaneous/minutely basis (e.g, time-series plots, histographs) and on an overall/daily basis (e.g., root mean square error, mean bias error).

3.4 Conduct the comparison using data from days falling into one of the below categories. If multiple days meet any of the below conditions, conduct up to 3 replicates per category. Some days may meet the criteria for two or more of the below categories, which is fine. The overall goal is to identify between 10 and 30 days out of the year, and then using those days for evalating the performance of the two algorithms.

1. mostly to completely clear skies, outdoor temperatures remained below 60°F during the 4 hour prior to and after solar noon

2. mostly to completely clear skies, outdoor temperatures remained above 60°F during the 4 hour prior to and after solar noon

3. mostly to partially cloudy skies, outdoor temperature remained below 60°F during the 4 hour prior to and after solar noon

4. mostly to partially cloudy skies, outdoor temperatures remained above 60°F during the 4 hour prior to and after solar noon

5. mostly to completely clear skies, outdoor temperatures remained below 60°F during the 4 hour prior to and after solar noon

6. mostly to completely clear skies, outdoor temperatures remained above 60°F during the 4 hour prior to and after solar noon

7. mostly to completely clear skies, a day in December of January

8. mostly to completely clear skies, a day in June or July

9. mostly to completely clear skes, a day in September or October

10. mostly to completely clear skies, comparatively higher sustained wind speeds during the 4 hours prior to and after solar noon.

11. mostly to completely clear skies, comparatively low wind speeds sustained during the 4 hours prior to and after solar noon.

Task 3 - Deliverables:

3.5 Provide a written report summarizing the results from comparing the measured versus predicted power from using both algorithms, IEC 61853-3 and an IEC 61853-calibrated diode-based model, and the NIST datasets. Include and discuss representative plots and informative statistical parameters that were generated as part of the Section 3.3 analysis. Share any trends observed regarding the performance of a predictive algorithm with respect to one or more prevailing weather conditions.

3.6 Provide an electronic copy of all plots, spreadsheets, and analysis and/or statistical code generated as part of the Section 3.3 analysis.

VI. PERIOD OF PERFORMANCE

The Period of Performance shall be 12 months from award of contract, anticipated to be on or around October 1, 2020 through September 30, 2021.

VII. GOVERNMENT FURNISHED PROPERTY

All property, data and information provided by the Government in the performance of this task remains the property of the Government and shall be surrendered to the government upon completion or termination of this requirement. Likewise, all deliverables generated under this requirement remain the property of the Government. All written material produced under this contract become the property of the U.S. Government and as such will be in the public domain.

The NIST CO and COR will provide access to the NIST datasets specified in this SOW as well as NIST personnel with relevant subject matter expertise, along with relevant background information such as technical papers, technical reports, and/or presentations on the relevant subject in digital format to complete defined tasks.

VIII. PLACE OF PERFORMANCE

The Place of Performance shall be the Contractor’s facility.

IX. RIGHTS

All data first produced under this contract become the property of the U.S. government and as such will be in the public domain.

X. KEY PERSONNEL

It is estimated that this work will require at least one contractor Key Personnel employee. The Key Personnel shall have at least three (3) years of experience in solar photovoltaic modeling and measurement uncertainty analysis.