A power generation abnormality detection system and method based on a prediction algorithm

Abstract

The present application relates to the technical field of photovoltaic power generation, and more particularly to a power generation abnormality detection system and method based on a prediction algorithm, the steps of the method comprising: obtaining operating parameters and environmental parameters of a photovoltaic module, the operating parameters including current data, voltage data, temperature data and actual power generation, and the environmental parameters including weather type, irradiance data and air quality index; predicting the degree of dust accumulation of the photovoltaic module based on the operating parameters and the environmental parameters; constructing a power generation prediction model to obtain predicted power generation of the photovoltaic module; and marking the photovoltaic module whose difference between the predicted power generation and the actual power generation is higher than a preset threshold as a power generation abnormality component. The present application comprehensively considers the operating parameters and the environmental parameters of the photovoltaic module to predict the degree of dust accumulation of the photovoltaic module, thereby improving the accuracy of power generation abnormality detection of the photovoltaic module.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation technology, and in particular to a power generation anomaly detection system and method based on a prediction algorithm. Background Technology

[0002] Solar photovoltaic (PV) power generation is a clean, renewable, and pollution-free new energy source. With the increasing proportion of new energy power generation, PV power generation is gradually becoming a major energy form. The operating status and environment of a PV power generation system directly affect its power generation efficiency and capacity, thus impacting system safety and economic benefits. Considering the complexity of the environment and the inherent characteristics of PV power generation systems, abnormal situations such as hardware failures, component aging, and dust accumulation on photovoltaic panels occur frequently. These not only lead to energy waste but also increase maintenance costs and affect power generation efficiency and stability.

[0003] For example, Chinese patent CN111711417B discloses a method for diagnosing abnormal power generation in a photovoltaic system, relating to the field of photovoltaic power generation technology. This method involves collecting the first voltage value and first output power value of each photovoltaic module at a first moment; collecting the second voltage value and second output power value of each photovoltaic module at a second moment; comparing the power difference between the first and second output power values ​​of each photovoltaic module with the voltage difference between the first and second voltage values ​​to obtain a data source; combining the data sources of all photovoltaic modules into a dataset; selecting abnormal data sources from the dataset that do not conform to preset rules, thereby determining that photovoltaic modules with abnormal data sources are in a state of abnormal power generation.

[0004] Chinese patent CN116015207A discloses a photovoltaic power generation anomaly early warning device and system, including the following modules: a light intensity sensor, set on the same plane as the photovoltaic module, for real-time monitoring of ambient light intensity; a temperature sensor, for real-time monitoring of the ambient temperature of the photovoltaic module; an energy detection component, for real-time monitoring of the energy output of the photovoltaic module; a lifespan management module, for predicting the conversion rate based on the historical information and model information of the photovoltaic module; and an early warning module, for obtaining the predicted power generation based on the ambient light intensity, ambient temperature, and predicted conversion rate, comparing the predicted power generation with the energy information detected by the energy detection component, and issuing an early warning based on the comparison result.

[0005] All of the above patents suffer from the problem described in this background: existing methods for detecting abnormal power generation typically do not consider the environmental parameters of the area where the photovoltaic modules are located or the degree of dust accumulation on the photovoltaic modules, resulting in a decrease in the accuracy of abnormal power generation detection under special weather conditions. Summary of the Invention

[0006] To overcome the defects and shortcomings of existing technologies, this invention provides a power generation anomaly detection system and method based on a prediction algorithm. It comprehensively considers the operating parameters and environmental parameters of photovoltaic modules to predict the degree of dust accumulation in photovoltaic modules, thereby further improving the accuracy of photovoltaic module power generation anomaly detection.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] This invention provides a method for detecting power generation anomalies based on a prediction algorithm, comprising the following steps:

[0009] Obtain the operating parameters and environmental parameters of the photovoltaic modules. The operating parameters include current data, voltage data, temperature data, and actual power generation. The environmental parameters include weather type, irradiance data, and air quality index.

[0010] Predicting the degree of dust accumulation on photovoltaic modules using operating and environmental parameters;

[0011] Construct a power generation prediction model to obtain the predicted power generation of photovoltaic modules;

[0012] Photovoltaic modules whose predicted power generation differs from actual power generation by more than a preset threshold are marked as modules with abnormal power generation. The preset threshold is determined by those skilled in the art based on a large number of experiments.

[0013] As a preferred technical solution, the weather types include sunny, cloudy, overcast, rain, snow, sunny turning cloudy, cloudy turning overcast, overcast turning rain, and cloudy turning snow.

[0014] As a preferred technical solution, the degree of dust accumulation in the photovoltaic module is predicted based on the photovoltaic module's IV curve, and the specific steps include:

[0015] The actual operating IV curve of the photovoltaic module is obtained by using current and voltage data;

[0016] Input the temperature data, weather type, irradiance data and air quality index of the photovoltaic module into the photovoltaic module simulation model to calculate the theoretical reference IV curve of the photovoltaic module;

[0017] The difference between the actual IV curve and the theoretical reference IV curve is analyzed using the derivative dynamic time warping algorithm to obtain the difference value;

[0018] The degree of dust accumulation in photovoltaic modules is calculated by matching the difference values ​​using fuzzy membership functions.

[0019] As a preferred technical solution, the power generation prediction model predicts power generation based on days with similar meteorological conditions and similar photovoltaic modules, and the specific steps include:

[0020] Match days with similar meteorological conditions based on weather type and air quality index;

[0021] Match similar photovoltaic modules based on the degree of dust accumulation on the photovoltaic modules;

[0022] The power generation of similar photovoltaic modules on days with similar weather conditions is used as the predicted power generation.

[0023] As a preferred technical solution, the photovoltaic module simulation model is expressed by the following formula:

[0024]

[0025] In the formula, I represents the output current of the photovoltaic module, U represents the output voltage of the photovoltaic module, and I d I represents the diode output current. ph R represents the photocurrent. sh R represents parallel resistance. s This indicates a series resistance.

[0026] As a preferred technical solution, the derivative dynamic time warping algorithm reduces the noise influence in the original current-voltage sequence, as expressed by the following formula:

[0027]

[0028] In the formula, X represents the current-voltage sequence of the actual operating IV curve, Y represents the current-voltage sequence of the theoretical reference IV curve, C represents the set of dynamic time bending paths, s represents the number of elements in each path, and l r Let d(·) represent the coordinates of the r-th point in the path, and let d(·) represent the distance function. DDTW (X,Y) represents the difference value.

[0029] As a preferred technical solution, the fuzzy membership function is one of the following: triangular membership function, trapezoidal membership function, Gaussian membership function, generalized bell-shaped membership function, and S-shaped membership function.

[0030] As a preferred technical solution, the photogenerated current is expressed by the following formula:

[0031]

[0032] In the formula I STC The short-circuit current represents the short-circuit current under standard test conditions, α represents the temperature coefficient at the short-circuit current point, G represents the irradiance data, T represents the temperature data of the photovoltaic module, and A represents the air quality index. STC T represents the air quality index under standard test conditions. STC G represents the reference temperature under standard test conditions. STC Reference irradiance data under standard test conditions, Iph It represents photocurrent.

[0033] The present invention also provides a power generation anomaly detection system based on a prediction algorithm, comprising:

[0034] The parameter acquisition module is used to acquire the operating parameters and environmental parameters of the photovoltaic module. The operating parameters include current data, voltage data, temperature data and actual power generation, while the environmental parameters include weather type, irradiance data and air quality index.

[0035] The dust accumulation prediction module is used to predict the dust accumulation level of photovoltaic modules based on operating parameters and environmental parameters.

[0036] The power generation prediction module is used to build a power generation prediction model and obtain the predicted power generation of photovoltaic modules.

[0037] The abnormal component marking module is used to mark photovoltaic modules whose predicted power generation and actual power generation differ from a preset threshold as abnormal power generation modules.

[0038] The present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for detecting abnormal power generation based on a prediction algorithm.

[0039] The present invention provides a controller comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to implement a method for detecting abnormal power generation based on a prediction algorithm.

[0040] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0041] (1) This invention predicts the degree of dust accumulation in photovoltaic modules by collecting operating parameters and environmental parameters of photovoltaic modules, and incorporates the degree of dust accumulation in photovoltaic modules into the influencing factors of power generation prediction, thereby improving the accuracy of abnormal power generation detection of photovoltaic modules.

[0042] (2) The present invention matches similar days of meteorological conditions based on weather type and air quality index, matches similar photovoltaic modules based on the degree of dust accumulation on photovoltaic modules, and uses the power generation of similar photovoltaic modules on days of similar meteorological conditions as the predicted power generation, thereby improving the accuracy of photovoltaic module power generation prediction. Attached Figure Description

[0043] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0044] Figure 1 This is a schematic diagram of the overall process of a power generation anomaly detection method based on a prediction algorithm according to the present invention.

[0045] Figure 2 This is a schematic diagram of the structure of a power generation anomaly detection system based on a prediction algorithm according to the present invention. Detailed Implementation

[0046] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0047] Example 1

[0048] like Figure 1 As shown, this embodiment provides a method for detecting power generation anomalies based on a prediction algorithm, which specifically includes the following steps:

[0049] S1: Obtain the operating parameters and environmental parameters of the photovoltaic module. The operating parameters include current data, voltage data, temperature data, and actual power generation. The environmental parameters include weather type, irradiance data, and air quality index. The weather type includes sunny, cloudy, overcast, rainy, snowy, sunny turning cloudy, cloudy turning overcast, overcast turning rainy, and cloudy turning snowy.

[0050] S2: Predict the degree of dust accumulation on photovoltaic modules by using operating and environmental parameters;

[0051] S21: The degree of dust accumulation on photovoltaic modules is predicted based on the photovoltaic module IV curve. The specific steps include:

[0052] The actual operating IV curve of the photovoltaic module is obtained by using current and voltage data;

[0053] Input the temperature data, weather type, irradiance data and air quality index of the photovoltaic module into the photovoltaic module simulation model to calculate the theoretical reference IV curve of the photovoltaic module;

[0054] The difference between the actual IV curve and the theoretical reference IV curve is analyzed using the derivative dynamic time warping algorithm to obtain the difference value;

[0055] The degree of dust accumulation in photovoltaic modules is calculated by matching the difference values ​​using fuzzy membership functions. Fuzzy membership functions are functions that map state variables to fuzzy sets. The fuzzy membership function can be one of the following: triangular membership function, trapezoidal membership function, Gaussian membership function, generalized bell membership function, and S-shaped membership function.

[0056] S22: The photovoltaic module simulation model is represented by the following formula:

[0057]

[0058] In the formula, I represents the output current of the photovoltaic module, U represents the output voltage of the photovoltaic module, and I d I represents the diode output current. ph R represents the photocurrent. sh R represents parallel resistance. s Indicates series resistance;

[0059] S23: Photocurrent is expressed by the following formula:

[0060]

[0061] In the formula I STC The short-circuit current represents the short-circuit current under standard test conditions, α represents the temperature coefficient at the short-circuit current point, G represents the irradiance data, T represents the temperature data of the photovoltaic module, and A represents the air quality index. STC T represents the air quality index under standard test conditions. STC G represents the reference temperature under standard test conditions. STC Reference irradiance data under standard test conditions, I ph Represents photocurrent, where,

[0062] The reference irradiance data under standard test conditions is 1000W / m2, the spectral distribution under standard test conditions conforms to the AM1.5 standard, that is, the actual distance of light passing through the atmosphere is 1.5 times the vertical thickness of the atmosphere, and the reference temperature under standard test conditions is 25℃.

[0063] S24: The derivative dynamic time warping algorithm reduces the noise in the original current-voltage sequence, as expressed by the following equation:

[0064]

[0065] In the formula, X represents the current-voltage sequence of the actual operating IV curve, Y represents the current-voltage sequence of the theoretical reference IV curve, C represents the set of dynamic time bending paths, s represents the number of elements in each path, and l r Let d(·) represent the coordinates of the r-th point in the path, and let d(·) represent the distance function. DDTW (X,Y) represents the difference value.

[0066] S3: Construct a power generation prediction model to obtain the predicted power generation of photovoltaic modules;

[0067] The power generation prediction model predicts power generation based on days with similar meteorological conditions and similar photovoltaic modules. The specific steps include:

[0068] Match days with similar meteorological conditions based on weather type and air quality index;

[0069] Match similar photovoltaic modules based on the degree of dust accumulation on the photovoltaic modules;

[0070] The power generation of similar photovoltaic modules on days with similar weather conditions is used as the predicted power generation.

[0071] S4: Mark photovoltaic modules whose predicted power generation and actual power generation differ from a preset threshold as modules with abnormal power generation.

[0072] Example 2

[0073] like Figure 2 As shown, this embodiment provides a power generation anomaly detection system based on a prediction algorithm, including:

[0074] The parameter acquisition module 21 is used to acquire the operating parameters and environmental parameters of the photovoltaic module. The operating parameters include current data, voltage data, temperature data and actual power generation, and the environmental parameters include weather type, irradiance data and air quality index.

[0075] Dust accumulation prediction module 22 is used to predict the dust accumulation level of photovoltaic modules based on operating parameters and environmental parameters;

[0076] The power generation prediction module 23 is used to build a power generation prediction model and obtain the predicted power generation of the photovoltaic module.

[0077] The abnormal component marking module 24 is used to mark photovoltaic modules whose difference between predicted power generation and actual power generation is higher than a preset threshold as abnormal power generation modules.

[0078] In this embodiment, the parameter acquisition module 21 is used to acquire the operating parameters and environmental parameters of the photovoltaic module. The operating parameters include current data, voltage data, temperature data and actual power generation. The environmental parameters include weather type, irradiance data and air quality index. The weather type includes sunny, cloudy, overcast, rain, snow, sunny turning cloudy, cloudy turning overcast, overcast turning rain and cloudy turning snow.

[0079] In this embodiment, the dust accumulation prediction module 22 is used to predict the dust accumulation level of the photovoltaic module based on operating parameters and environmental parameters.

[0080] The degree of dust accumulation on photovoltaic modules is predicted based on the module's IV curve. The specific steps include:

[0081] The actual operating IV curve of the photovoltaic module is obtained by using current and voltage data;

[0082] Input the temperature data, weather type, irradiance data and air quality index of the photovoltaic module into the photovoltaic module simulation model to calculate the theoretical reference IV curve of the photovoltaic module;

[0083] The difference between the actual IV curve and the theoretical reference IV curve is analyzed using the derivative dynamic time warping algorithm to obtain the difference value;

[0084] The degree of dust accumulation in photovoltaic modules is calculated by matching the difference values ​​using fuzzy membership functions.

[0085] The photovoltaic module simulation model is represented by the following formula:

[0086]

[0087] In the formula, I represents the output current of the photovoltaic module, U represents the output voltage of the photovoltaic module, and I d I represents the diode output current. ph R represents the photocurrent. sh R represents parallel resistance. s Indicates series resistance;

[0088] Photocurrent is expressed by the following formula:

[0089]

[0090] In the formula I STC The short-circuit current represents the short-circuit current under standard test conditions, α represents the temperature coefficient at the short-circuit current point, G represents the irradiance data, T represents the temperature data of the photovoltaic module, and A represents the air quality index. STC T represents the air quality index under standard test conditions. STC G represents the reference temperature under standard test conditions. STC Reference irradiance data under standard test conditions, I ph Indicates photocurrent;

[0091] The derivative dynamic time warping algorithm is expressed by the following formula:

[0092]

[0093] In the formula, X represents the current-voltage sequence of the actual operating IV curve, Y represents the current-voltage sequence of the theoretical reference IV curve, C represents the set of dynamic time bending paths, s represents the number of elements in each path, and l r Let d(·) represent the coordinates of the r-th point in the path, and let d(·) represent the distance function. DDTW (X,Y) represents the difference value;

[0094] Fuzzy membership functions are one of the following: triangular membership function, trapezoidal membership function, Gaussian membership function, generalized bell-shaped membership function, and S-shaped membership function.

[0095] In this embodiment, the power generation prediction module 23 is used to construct a power generation prediction model and obtain the predicted power generation of the photovoltaic module, wherein...

[0096] The power generation prediction model predicts power generation based on days with similar meteorological conditions and similar photovoltaic modules. The specific steps include:

[0097] Match days with similar meteorological conditions based on weather type and air quality index;

[0098] Match similar photovoltaic modules based on the degree of dust accumulation on the photovoltaic modules;

[0099] The power generation of similar photovoltaic modules on days with similar weather conditions is used as the predicted power generation.

[0100] The parameters and steps of each unit module in the power generation anomaly detection system based on the prediction algorithm of the present invention described above can be referred to the parameters and steps in the embodiment of the power generation anomaly detection method based on the prediction algorithm above, and will not be repeated here.

[0101] Example 3

[0102] This invention discloses a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the aforementioned method for detecting anomalies in power generation based on a prediction algorithm. It should be noted that all computer programs for this method are implemented using Python. The dust accumulation prediction module, power generation prediction module, and anomaly component marking module are all controlled by a remote server. The remote server has an Intel Xeon Gold 5118 CPU, an NVIDIA GTX 2080Ti 11GB GPU, an Ubuntu 18.04.2 operating system, PyTorch 1.7.0 deep learning framework, CUDA version 10.2, and uses cuDNN 7.6.5 for accelerated inference. The Intel Xeon Gold 5118 includes a memory and a processor. The memory stores the computer program, and the processor executes the computer program, causing the Intel Xeon Gold 5118 to implement the method for detecting anomalies in power generation based on a prediction algorithm.

[0103] Those skilled in the art will know that this invention can be implemented as a system, method, or computer program product.

[0104] Therefore, this disclosure can be implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product in one or more computer-readable media, the computer-readable medium containing computer-readable program code.

[0105] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0106] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for detecting abnormal power generation based on a prediction algorithm, characterized in that, Includes the following steps: Obtain the operating parameters and environmental parameters of the photovoltaic modules. The operating parameters include current data, voltage data, temperature data, and actual power generation. The environmental parameters include weather type, irradiance data, and air quality index. The degree of dust accumulation in photovoltaic modules is predicted by using operating parameters and environmental parameters. The prediction of the degree of dust accumulation in photovoltaic modules is based on the IV curve of the photovoltaic modules. The specific steps include: obtaining the actual operating IV curve of the photovoltaic modules through current data and voltage data; inputting the temperature data, weather type, irradiance data and air quality index of the photovoltaic modules into the photovoltaic module simulation model; calculating the theoretical reference IV curve of the photovoltaic modules; using the derivative dynamic time warping algorithm to analyze the difference between the actual operating IV curve and the theoretical reference IV curve; obtaining the difference value; using the fuzzy membership function to match the difference value; and calculating the degree of dust accumulation in the photovoltaic modules. A power generation prediction model is constructed to obtain the predicted power generation of photovoltaic modules. The power generation prediction model predicts power generation based on days with similar meteorological conditions and similar photovoltaic modules. The specific steps include: matching days with similar meteorological conditions according to weather type and air quality index, matching similar photovoltaic modules according to the degree of dust accumulation on photovoltaic modules, and using the power generation of similar photovoltaic modules on days with similar meteorological conditions as the predicted power generation. Photovoltaic modules whose difference between predicted and actual power generation exceeds a preset threshold are marked as modules with abnormal power generation. The photocurrent in the photovoltaic module simulation model is expressed by the following formula: ； In the formula This represents the short-circuit current under standard test conditions. The temperature coefficient representing the point of short-circuit current. Represents irradiance data. This indicates the temperature data of the photovoltaic module. This indicates the air quality index. This represents the air quality index under standard testing conditions. This indicates the reference temperature under standard test conditions. Reference irradiance data under standard test conditions. It represents photocurrent.

2. The method for detecting abnormal power generation based on a prediction algorithm according to claim 1, characterized in that, The photovoltaic module simulation model is represented by the following formula: ； In the formula Indicates the output current of the photovoltaic module. Indicates the output voltage of the photovoltaic module. Indicates the diode output current. Indicates photocurrent, Indicates parallel resistance. This indicates a series resistance.

3. The method for detecting abnormal power generation based on a prediction algorithm according to claim 1, characterized in that, The derivative dynamic time warping algorithm is expressed by the following formula: ； In the formula This represents the current-voltage sequence of the actual operating IV curve. This represents the current-voltage sequence of the theoretical reference IV curve. Represents a set of dynamic time-bending paths. This indicates the number of elements in each path. Indicates the first in the path The coordinates of the points Represents the distance function. Indicates the difference value.

4. The method for detecting abnormal power generation based on a prediction algorithm according to claim 1, characterized in that, The fuzzy membership function is one of the following: triangular membership function, trapezoidal membership function, Gaussian membership function, generalized bell-shaped membership function, and S-shaped membership function.

5. A power generation anomaly detection system based on a prediction algorithm, implemented according to any one of claims 1-4, characterized in that, The system includes: The parameter acquisition module is used to acquire the operating parameters and environmental parameters of the photovoltaic module. The operating parameters include current data, voltage data, temperature data and actual power generation, while the environmental parameters include weather type, irradiance data and air quality index. The dust accumulation prediction module is used to predict the dust accumulation level of photovoltaic modules based on operating parameters and environmental parameters. The power generation prediction module is used to build a power generation prediction model and obtain the predicted power generation of photovoltaic modules. The abnormal component marking module is used to mark photovoltaic modules whose predicted power generation and actual power generation differ from a preset threshold as abnormal power generation modules.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the power generation anomaly detection method based on the prediction algorithm according to any one of claims 1-4.

7. A controller comprising a memory and a processor, the memory for storing a computer program, characterized in that, When the processor executes the computer program, it implements the power generation anomaly detection method based on the prediction algorithm according to any one of claims 1-4.

Images