Method and apparatus for detecting anomalies in solar power generation

Abstract

The present disclosure relates to a method for detecting an abnormality in solar power generation, performed by at least one processor, including receiving inverter input / output data of a solar power plant, receiving hourly solar radiation data for a region associated with the solar power plant, and determining an abnormal state of the solar power plant based on at least one of the inverter input / output data or the hourly solar radiation data.

Claims

1. In a method for detecting anomalies in solar power generation, which is performed by at least one processor, The steps include receiving inverter input / output data from a solar power plant, The steps include receiving hourly solar radiation data for the region associated with the aforementioned solar power plant, A step of determining an abnormal state of the solar power plant based on at least one of the inverter input / output data or the hourly solar radiation data. Includes, The step of determining the abnormal condition of the solar power plant is: The steps include: estimating the expected power generation of the solar power plant based on the hourly solar radiation data; The steps include determining the actual power generation amount of the solar power plant based on the inverter input / output data, The step of performing a primary verification to determine whether the difference between the predicted power generation amount and the actual power generation amount exceeds a predetermined first threshold. Includes, The inverter input / output data includes the DC power value input to the inverter at each time interval and the AC power value output from the inverter at each time interval. In response to the determination that the aforementioned solar power plant was in an abnormal state, If both the hourly DC power value input to the inverter and the hourly AC power value output from the inverter show a decreasing trend, the step of detecting a decrease in the performance of the solar panel connected to the inverter. A method for detecting anomalies in solar power generation, which further includes the above.

2. The method for detecting an anomaly in a solar power plant according to claim 1, wherein the anomaly in the solar power plant includes at least one of the following: an abnormal state of the inverter, an abnormal state of the solar panels connected to the inverter, or a decrease in the performance of the solar panels.

3. The step of receiving the hourly solar radiation data is: The steps include receiving multiple satellite images of ground-reaching solar radiation (where the RGB values ​​of each pixel on the hourly satellite map image indicate the amount of solar radiation reaching the ground in that region), Based on the multiple satellite images of ground-reaching solar radiation, the steps include: extracting hourly solar radiation data for the region associated with the solar power plant; The method for detecting anomalies in solar power generation according to claim 1, including the following:

4. The aforementioned predicted power generation is estimated using a machine learning model. The method for detecting anomalies in solar power generation according to claim 1, wherein the machine learning model is trained on past actual power generation amounts and satellite images of past solar radiation reaching the ground associated with the past actual power generation amounts.

5. The step of determining the abnormal condition of the solar power plant is: In response to the determination that the difference between the predicted power generation and the actual power generation exceeds a predetermined first threshold, a secondary verification is performed to determine whether the difference between the past average actual power generation of the solar power plant and the actual power generation exceeds a predetermined second threshold. The method for detecting anomalies in solar power generation according to claim 1, further comprising:

6. The inverter input / output data includes the DC power value input to the inverter at each time interval and the AC power value output from the inverter at each time interval. In response to the fact that the aforementioned solar power plant was determined to be in an abnormal state, A step of calculating the power conversion efficiency of the inverter based on the inverter input / output data, If the calculated power conversion efficiency of the inverter exceeds a predetermined normal operating power conversion efficiency range associated with the inverter of the solar power plant, the step of determining that the inverter is in an abnormal state is as follows: The method for detecting anomalies in solar power generation according to claim 1, further comprising:

7. If the calculated power conversion efficiency of the inverter is within the predetermined normal operating power conversion efficiency range, the calculated power conversion efficiency of the inverter is compared with the past power conversion efficiency of the inverter of the solar power plant to determine if there is an abnormality in the inverter's condition. The method for detecting anomalies in solar power generation according to claim 6, further comprising:

8. The inverter input / output data includes the DC power value input to the inverter at each time interval and the AC power value output from the inverter at each time interval. In response to the determination that the aforementioned solar power plant was in an abnormal state, The step of comparing the expected power generation of the solar power plant with the hourly DC power input to the inverter to detect abnormal conditions of the solar panels connected to the inverter. The method for detecting anomalies in solar power generation according to claim 1, further comprising:

9. The inverter input / output data includes the DC power value input to the inverter at each time interval and the AC power value output from the inverter at each time interval. In response to the determination that the aforementioned solar power plant was in an abnormal state, The step of detecting an abnormality in the inverter by comparing the expected power generation of the solar power plant with the hourly AC power output from the inverter. The method for detecting anomalies in solar power generation according to claim 1, further comprising:

10. A computer-readable non-temporary recording medium that stores instruction words for executing the method according to claim 1 on a computer.

11. An information processing system, Communication module and Memory and The memory is connected to at least one processor configured to execute at least one computer-readable program contained in the memory. Includes, The at least one program is Receives inverter input / output data from a solar power plant. The solar power plant receives hourly solar radiation data for the region associated with the aforementioned solar power plant. The command includes instructions for determining an abnormal state of the solar power plant based on at least one of the inverter input / output data or the hourly solar radiation data, The command words for determining the abnormal state of the aforementioned solar power plant are: The steps include: estimating the expected power generation of the solar power plant based on the hourly solar radiation data; The steps include determining the actual power generation amount of the solar power plant based on the inverter input / output data, The step of performing a primary verification to determine whether the difference between the predicted power generation amount and the actual power generation amount exceeds a predetermined first threshold. Includes, The inverter input / output data includes the DC power value input to the inverter at each time interval and the AC power value output from the inverter at each time interval. In response to the determination that the aforementioned solar power plant was in an abnormal state, If both the hourly DC power value input to the inverter and the hourly AC power value output from the inverter show a decreasing trend, the step of detecting a decrease in the performance of the solar panel connected to the inverter. An information processing system that further includes this.

Images