Estimating performance of photovoltaic systems

Patent metadata

Field	Value
Publication number	US-11615487-B2
Application number	US-202016813285-A
Country	US
Kind code	B2
Filing date	Mar 9, 2020
Priority date	Mar 14, 2019
Publication date	Mar 28, 2023
Grant date	Mar 28, 2023

Abstract

An energy evaluation system includes an energy system modeling framework and an Application Program Interface (API) as part of the energy system evaluation framework. The energy evaluation system can be configured to query a simulation application program interface (API) with PV system configuration parameters as arguments, generate a shade loss time series based on the system configuration parameters, simulate energy output for the PV system based on the shade loss time series, and output energy aggregations based on the simulation.

First claim

The invention claimed is:

1. A method for energy output evaluation of a photovoltaic (PV) system, comprising:
querying a simulation application program interface (API) with PV system configuration parameters as arguments;
generating a shade loss time series based on the system configuration parameters and a trained machine learning opto-electrical shade model, wherein generating the shade loss time series based on the trained machine learning opto-electrical shade model includes;
receiving, at the trained machine learning opto-electrical shade model, an irradiance ratio for the PV system, wherein the irradiance ratio corresponds to a ratio of diffuse irradiance to plane of array irradiance, wherein the irradiance ratio is received from a weather model;
receiving, at the trained machine learning opto-electrical shade model, a shade ratio for each sub-circuit of a solar module in the PV system; and
generating, by the trained machine learning opto-electrical shade model, a module power based on the irradiance ratio and the shade ratio for each sub-circuit of the solar module;

simulating energy output for the PV system based on the shade loss time series; and
outputting energy aggregations based on the simulation.

2. The method of claim 1 , further comprising:
calculating bill savings based on the energy aggregations, a customer load profile, and applicable tariffs; and
display the bill savings.

3. The method of claim 1 , further comprising:
calculating a payback period based on the energy aggregations, a customer load profile, and applicable tariffs; and
display the payback period.

4. The method of claim 1 , wherein generating the shade loss time series further comprises:
requesting weather data from a weather API corresponding to a location of the PV system.

5. The method of claim 1 , wherein training the machine learning opto-electrical shade model further comprises:
receiving an irradiance ratio, where in the irradiance ratio corresponds to a diffuse irradiance to plane of array irradiance ratio;
receiving a shade ratio for each sub-circuit in a solar module; and
generating a training dataset based on the irradiance ratio and each shade ratio.

6. The method of claim 1 , wherein simulating energy output for the PV system further comprises:
requesting an array layout of the PV system from a database; and
triggering the simulation in response to requesting the array layout.

7. The method of claim 1 , further comprising:
receiving the energy aggregations via a web-based business logic interface.

8. The method of claim 1 , wherein the PV system configuration parameters include a PV system location, a solar panel array layout, and solar cell string level shade ratios.

9. The method of claim 8 , wherein the shade loss time series is an array layout specific weather timeseries based on the PV system configuration parameters.

10. One or more computer readable medium including computer program instructions, which when executed by an information processing system, cause the system to:
querying a simulation application program interface) with PV system configuration parameters as arguments;
generating a shade loss time series based on the system configuration parameters and a trained machine learning opto-electrical shade model, wherein generating the shade loss time series based on the trained machine learning opto-electrical shade model includes;
receiving, at the trained machine learning opto-electrical shade model, an irradiance ratio for the PV system, wherein the irradiance ratio corresponds to a ratio of diffuse irradiance to plane of array irradiance, wherein the irradiance ratio is received from a weather model;
receiving, at the trained machine learning opto-electrical shade model, a shade ratio for each sub-circuit of a solar module in the PV system; and
generating, by the trained machine learning opto-electrical shade model, a module power based on the irradiance ratio and the shade ratio for each sub-circuit of the solar module;

simulating energy output for the PV system based on the shade loss time series; and
outputting energy aggregations based on the simulation.

11. The one or more computer readable medium of claim 10 , wherein the information processing system is a server.

12. The one or more computer readable medium of claim 10 , wherein the information processing system is a cloud-based architecture.

13. A method for energy evaluation of an energy system, comprising:
receiving system configuration parameters from a third-party as a request to an Application Program Interface (API);
passing the system configuration parameters to an energy system evaluation framework, wherein passing the system configuration parameters to the energy system evaluation framework includes;
querying a simulation API with the system configuration parameters as arguments;
generating, by the framework, a shade loss time series based on the system configuration parameters and a trained machine learning opto-electrical shade model, wherein generating the shade loss time series based on the trained machine learning opto-electrical shade model includes;
receiving, at the trained machine learning opto-electrical shade model, an irradiance ratio for the PV system, wherein the irradiance ratio corresponds to a ratio of diffuse irradiance to plane of array irradiance, wherein the irradiance ratio is received from a weather model;
receiving, at the trained machine learning opto-electrical shade model, a shade ratio for each sub-circuit of a solar module in the PV system; and
generating, by the trained machine learning opto-electrical shade model, a module power based on the irradiance ratio and the shade ratio for each sub-circuit of the solar module;

simulating, by the framework, energy output for the PV system based on the shade loss time series; and
outputting, by the framework, energy aggregations based on the simulation;

receiving energy yield information for the energy system based on the system configuration parameters; and
outputting energy yield information to the third-party.

14. The method of claim 13 , further comprising:
receiving system configuration parameters for an energy system from an Application Program Interface (API);
generating a 3D site geometry based on the system configuration parameters;
calculating an effective irradiance (Ee) for photovoltaic (PV) modules;
calculating inverter-level Maximum Power Point (P mp );
adjusting the P mp to comply with grid integration requirements;
calculating energy yield information for the energy system; and
returning the energy yield information to the API.

15. The method of claim 14 , wherein the Ee for each cell in PV modules across the energy system is calculated using cell shade ratios and a weather data model, the weather data model including highly localized components of irradiance and cell temperatures based on a location of the energy system.

16. The method of claim 14 , wherein the inverter-level P mp ) is calculated using an electrical Maximum Power Point Tracking (MPPT) model.

17. The method of claim 13 , wherein the energy yield information corresponds to an energy yield of the energy system over a predetermined time interval.

18. The method of claim 13 , wherein the third party can be one or more of a website, a customer relations management tool, a third-party tool, and one or more individual users.

Assignees

• Sunpower Corp

Inventors

• Chaudhari Chetan
• Lance Tamir
• Shishavan Amir Asgharzadeh

Classifications

• H02J2103/30

  Simulating, planning, modelling, reliability check or computer assisted design [CAD] of electric power networks · CPC title

• H02J2101/25

  involving maximum power point tracking control for photovoltaic sources · CPC title

• G06F2119/06

  Power analysis or power optimisation · CPC title

• G06Q50/06 Primary

  Energy or water supply · CPC title

• G06F9/54

  Interprogram communication · CPC title