Estimation of energy losses due to partial equipment failure for photovoltaic systems from measured and modeled inputs

1. A method of monitoring a photovoltaic system and quantifying energy losses due to partial equipment failure, comprising:

inputting modeled power and energy time series data for the photovoltaic system into a computer processor, wherein the modeled power and energy time series data includes a set of a timestamp, a modeled power and a modeled energy;

receiving measured power and energy time series data for the photovoltaic system, wherein the measured power and energy time series data includes a set of actual data readings having a power and energy reading with an associated measured time series data timestamp;

inputting the measured power and energy time series data into a computer processor;

merging by the computer processor the modeled power and energy time series data with the measured power and energy time series data to determine and to provide a performance ratio time series;

determining in a computer processor a normalized daily effective system capacity time series by calculating a daily maximum of the performance ratio times series with outliers removed to provide the normalized daily effective system capacity time series;

determining in a computer processor a set of differentials between consecutive daily data points in the normalized daily effective system capacity time series;

applying an anomaly filter to the set of differentials between consecutive daily data points in the normalized daily effective system capacity time series to identify days on which a capacity of the photovoltaic system is likely to have changed;

partitioning the normalized daily effective system capacity time series and then calculating measures of central tendency for the resulting partitions to generate a piecewise constant time series that describes a normalized daily effective system size; and

modulating the modeled power and energy time series data by an inverse of the piecewise constant time series to provide an estimate of power and energy losses due to partial equipment failure.

2. A method as in claim 1 , wherein the outliers are days with excessive levels of snow cover.

3. A method as in claim 1 , wherein the modeled power and energy time series data is calculated from irradiance data, weather data and photovoltaic system configuration information.

4. A method as in claim 1 , wherein the measured power and energy time series data is measured with an on-site physical sensor installed in the photovoltaic system or in an inverter.

5. A method as in claim 1 , wherein the modeled power and energy time series data provides an estimate of what the power and energy production of the photovoltaic system would be in the absence of equipment failure.

6. A method as in claim 1 , wherein the measured power and energy time series data includes measured irradiance data.

7. A method as in claim 1 , further comprising:

remotely identifying PV systems that are currently experiencing equipment failure and are in need of repair.

8. A method as in claim 1 , wherein the modeled power and energy time series data is according to configuration data, wherein the configuration data includes at least one of a size of the photovoltaic system, an orientation of the photovoltaic system and a location of the photovoltaic system.

9. A method as in claim 1 , wherein the measured power and energy time series data is according to at least one of solar irradiance data at the location of the photovoltaic system, weather data including temperature, wind conditions and precipitation at the location of the photovoltaic system, module level measured temperature data, snow depth data, and precipitation data for the location of the photovoltaic system.

10. A method as in claim 1 , wherein determining in a computer processor a normalized daily effective system capacity time series further comprises:

removing outliers in the performance ratio time series to provide a performance ratio times series with outliers removed; and

calculating the daily maximum of the performance ratio times series with outliers removed to provide the normalized daily effective system capacity time series.

11. A method as in claim 1 , wherein the modeled power and energy time series data is according to photovoltaic system configuration data.

12. A method as in claim 11 , wherein the photovoltaic system configuration data is a size, an orientation, an age, and a location of the photovoltaic system.

13. A method as in claim 12 , wherein the orientation is a tilt angle and an azimuth angle for panels of the photovoltaic system.

14. A method as in claim 1 , further comprising assessing the impact of the energy losses due to partial equipment failure on a contract associated with the photovoltaic system.

15. A method as in claim 1 , further comprising validating design estimates of expected equipment failure losses for the photovoltaic system.

16. A method as in claim 1 , further comprising developing a statistical model of the failure rates of components of the photovoltaic system.

17. A method as in claim 1 , further comprising determining a cause for underperforming a contractual performance guarantee associated with the photovoltaic system.

18. A method as in claim 1 , further comprising assessing risk associated with a performance guarantee on energy production of the photovoltaic system.