Machine learning-based distributed photovoltaic power station temperature anomaly detection method and system

Abstract

This invention discloses a method and system for detecting temperature anomalies in distributed photovoltaic (PV) power plants based on machine learning. The method includes: collecting PV power plant data, which includes temperature data, environmental data, and electrical parameter data; preprocessing the PV power plant data to obtain preprocessed data; extracting features from the preprocessed data to obtain temperature change rate features, inter-module temperature difference features, and correlation features between temperature and other parameters; and inputting the temperature change rate features, inter-module temperature difference features, and correlation features between temperature and other parameters into a trained hybrid model to determine whether the current temperature state of the PV power plant is abnormal. This method and system can achieve real-time and accurate monitoring of the power plant's temperature state.

Description

Technical Field

[0001] This invention belongs to the field of operation and maintenance technology of distributed photovoltaic power stations, and relates to a method and system for detecting temperature anomalies in distributed photovoltaic power stations based on machine learning. Background Technology

[0002] In existing distributed photovoltaic power station operation and maintenance technologies, the following are the main methods for detecting temperature anomalies and their drawbacks: Traditional threshold detection methods rely on pre-setting fixed temperature thresholds. When a monitored temperature exceeds this threshold, it is considered an anomaly. However, this method is too simplistic and cannot adapt to the complex and variable operating environment of photovoltaic power plants. For example, the normal operating temperature range of photovoltaic modules varies depending on the season, sunlight intensity, geographical location, and the degree of module aging. Fixed thresholds are insufficient for accurately identifying anomalies, leading to false alarms or missed alarms. This prevents maintenance personnel from promptly and accurately detecting genuine temperature anomalies, impacting the stable operation and power generation efficiency of the power plant.

[0003] Physical model-based detection methods rely on mathematical models constructed based on the physical characteristics and heat conduction principles of photovoltaic power plants to predict and determine whether temperatures are abnormal. However, this method requires precise power plant parameters and complex calculations. Furthermore, the models are often based on ideal conditions. In actual operation, differences in photovoltaic module manufacturing processes and uncertainties in environmental factors (such as dust obstruction and uneven illumination caused by cloud cover) significantly reduce the accuracy of the models, making them difficult to apply effectively to actual temperature anomaly detection and unable to provide a reliable basis for operation and maintenance decisions.

[0004] Simple statistical analysis methods utilize historical temperature data for statistical analysis, such as calculating the average and standard deviation, to determine the normal temperature range and thus judge whether the current temperature is abnormal. However, this method is overly reliant on data and lacks the ability to deeply analyze data trends and potential anomaly patterns. When encountering sudden, atypical temperature anomalies (such as a sudden temperature rise caused by a local component failure), it cannot promptly capture such abnormal trends, resulting in detection lag and failure to provide timely warnings. This may allow the fault to escalate further, increasing power plant maintenance costs and downtime.

[0005] In summary, existing temperature anomaly detection technologies for distributed photovoltaic power plants have significant shortcomings in terms of accuracy, adaptability, and timeliness, making it difficult to meet the growing demand for efficient operation and maintenance of photovoltaic power plants. There is an urgent need for a more advanced, accurate, and reliable temperature anomaly detection algorithm. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning. This method and system can achieve real-time and accurate monitoring of the temperature status of the power plant.

[0007] To achieve the above objectives, this invention discloses a method for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning, comprising: Collect data from photovoltaic power plants, including temperature data, environmental data, and electrical parameter data. The photovoltaic power station data is preprocessed to obtain preprocessed data; Feature extraction is performed on the preprocessed data to obtain features such as temperature change rate, temperature difference between components, and correlation between temperature and other parameters. The temperature change rate characteristics, inter-module temperature difference characteristics, and temperature correlation characteristics with other parameters are input into the trained hybrid model to determine whether the current temperature state of the photovoltaic power station is abnormal.

[0008] Furthermore, the temperature data includes temperature information at various detection locations of the photovoltaic module, temperature information inside the inverter, and temperature information inside the combiner box; the environmental data includes the irradiance and wind speed of the photovoltaic power station; and the electrical parameter data includes the current and voltage information of the photovoltaic power station.

[0009] Furthermore, the preprocessing process for the photovoltaic power station data is as follows: outlier handling is performed on the temperature data in the photovoltaic power station data using the 3σ principle.

[0010] Furthermore, the correlation coefficient characterizes the correlation between temperature and other parameters, and the correlation coefficient is:

[0011] Where T is the temperature data sequence, X is the data sequence of other parameters, and n is the number of data points. and These are the average values ​​of temperature and other parameters, respectively.

[0012] Furthermore, the hybrid model is constructed based on long short-term memory networks and support vector machines.

[0013] Furthermore, the hybrid model is trained based on the stochastic gradient descent algorithm and the backpropagation algorithm.

[0014] Furthermore, the loss function of the hybrid model during training is:

[0015] Where N is the number of training samples. Let i be the true label of the i-th sample. This represents the predicted output probability of the model.

[0016] This invention discloses a machine learning-based distributed photovoltaic power plant temperature anomaly detection system, comprising: The data acquisition module is used to collect data from the photovoltaic power station, including temperature data, environmental data, and electrical parameter data. The preprocessing module is used to preprocess the photovoltaic power station data to obtain preprocessed data; The extraction module is used to extract features from the preprocessed data to obtain features such as temperature change rate, temperature difference between components, and correlation features between temperature and other parameters. The judgment module is used to input the temperature change rate characteristics, the temperature difference characteristics between components, and the correlation characteristics between temperature and other parameters into the trained hybrid model to determine whether the temperature state of the current photovoltaic power station is abnormal.

[0017] This invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the machine learning-based distributed photovoltaic power station temperature anomaly detection method.

[0018] This invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the machine learning-based distributed photovoltaic power station temperature anomaly detection method.

[0019] The present invention has the following beneficial effects: In specific operation, the distributed photovoltaic power station temperature anomaly detection method and system based on machine learning described in this invention extracts features from the preprocessed data to obtain temperature change rate features, inter-module temperature difference features, and temperature correlation features with other parameters. These features are then input into a trained hybrid model to determine whether the current temperature state of the photovoltaic power station is abnormal. The multi-feature fusion approach is used to improve the real-time performance and accuracy of photovoltaic power station temperature state monitoring. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the hybrid model in this invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0024] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0025] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0026] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0027] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0029] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0030] Example 1 refer to Figure 1 and Figure 2 The temperature anomaly detection method for distributed photovoltaic power plants based on machine learning described in this invention includes the following steps: 1) Collect data from photovoltaic power plants; In a typical distributed photovoltaic (PV) power plant, 120 high-precision temperature sensors are deployed, installed at different locations on the PV modules (such as the surface of the cells, backsheets, etc.), inside the inverters, and in the combiner boxes to obtain comprehensive and accurate temperature data. Simultaneously, 20 illuminance sensors, 20 wind speed sensors, and current and voltage sensors are installed to collect environmental and electrical parameter data. These sensors acquire data at a frequency of 10kHz and transmit the data in real time to the data processing center via wireless communication modules (such as LoRa or 4G networks).

[0031] 2) Data collection and organization; After a period of continuous data collection, a total of 120 temperature data points were collected, along with corresponding data on light intensity, wind speed, current, voltage, and other related parameters. These raw data were initially processed to remove obvious errors (such as data outside the sensor's measurement range) and records with excessive missing values, ultimately yielding a valid data sample set D.

[0032] 3) Data preprocessing; 31) Data cleaning; 31) Use the 3σ principle to handle outliers in temperature data; Assuming the temperature data follows a normal distribution, calculate the mean μ and standard deviation σ of the temperature data. For data points x that satisfy |x - μ|>3σ, they are considered outliers and are removed or corrected (linear interpolation can be used for correction). After data cleaning, a more reliable temperature dataset D_clean is obtained.

[0033] 4) Feature extraction; Calculate the temperature change rate characteristics: for each temperature sensor's data sequence Calculate its rate of temperature change within the time window Δt. The formula is:

[0034] Where i represents the i-th temperature sensor, and t represents time.

[0035] Extracting temperature difference characteristics between modules: Calculating the temperature difference between photovoltaic modules at different locations, such as... ,in, and These are the temperature values ​​of two different components. These temperature difference characteristics can reflect the temperature distribution between the components and help to detect local temperature anomalies.

[0036] Calculate the correlation characteristics between temperature and other parameters: The correlation coefficient ρ between temperature and parameters such as light intensity, wind speed, and power generation is calculated using the Pearson correlation coefficient. The formula is:

[0037] Where T represents the temperature data sequence, X represents other parameter data sequences (such as light intensity, wind speed, etc.), and n represents the number of data points. and These are the average values ​​of temperature and other parameters, respectively.

[0038] The extracted features are combined with the original temperature data to form the final feature dataset F, which is used for subsequent machine learning model training.

[0039] 5) Machine learning model building and training; 51) Model selection and architecture design; A hybrid model architecture employing a Long Short-Term Memory (LSTM) network and integrating a Support Vector Machine (SVM) for classification.

[0040] LSTM layer: The number of LSTM units is set to 20 to learn and memorize the time-series features of temperature data, capturing the trends and periodic patterns of temperature changes. The input data dimension is the number of features F after feature engineering. dim The output dimension is LSTM output_dim ].

[0041] SVM layer: The output of LSTM is used as the input feature of SVM, and the radial basis function (RBF) is used as the kernel function. The classification performance of SVM is optimized by adjusting the penalty parameter C and the kernel function parameter γ to achieve the final classification judgment of temperature anomalies.

[0042] 52) Model training; The feature dataset F is divided into a training set F_train, a validation set F_val, and a test set F_test, with each set comprising 70%, 15%, and 15% of the data. The constructed hybrid model is trained using the training set. During training, stochastic gradient descent (SGD) is employed with a learning rate of 0.06 and 150 iterations. In each training iteration, the cross-entropy loss function Loss between the model's predicted output and the true labels on the training set is calculated, using the following formula:

[0043] Where N is the number of training samples. Let be the true label of the i-th sample (0 indicates normal, 1 indicates abnormal). This represents the predicted output probability of the model. The model parameters are updated using backpropagation based on the gradient information of the loss function to minimize the loss function. During training, the model's performance is evaluated every 10 iterations using a validation set, calculating evaluation metrics such as accuracy, recall, and F1 score. The formulas are as follows:

[0044]

[0045]

[0046] Where TP represents the number of true positive samples (actually abnormal but predicted as abnormal by the model), TN represents the number of true negative samples (actually normal but predicted as normal by the model), FP represents the number of false positive samples (actually normal but predicted as abnormal by the model), and FN represents the number of false negative samples (actually abnormal but predicted as normal by the model). When the model's performance metrics on the validation set no longer improve, training is stopped, and the model's optimal parameters are saved.

[0047] 6) Model evaluation and optimization; 61) Model evaluation; The trained model was finally evaluated using the test set F_test. The model's accuracy, recall, F1 score, and other relevant evaluation metrics, such as mean squared error (MSE), were calculated on the test set to comprehensively measure the model's performance and generalization ability. After evaluation, the model achieved the [specific accuracy value] and [specific recall value] on the test set, indicating that the model can effectively detect temperature anomalies in distributed photovoltaic power stations.

[0048] 62) Model optimization; Hyperparameter tuning: The hyperparameters of the model (such as the number of LSTM units, convolutional kernel size, learning rate, penalty parameter C, and kernel function parameter γ) are further optimized through methods such as grid search or random search to find the optimal combination of hyperparameters and improve the performance of the model.

[0049] 63) Model fusion; Consider using multiple models with different structures or parameters for fusion, such as weighted average or voting methods, to combine the prediction results of multiple models in order to improve the accuracy and stability of temperature anomaly detection.

[0050] 7) Deployment of anomaly detection and alarm system; 71) Real-time anomaly detection; The trained model is deployed into the monitoring system of a distributed photovoltaic power station. It receives preprocessed temperature data and related feature data in real time, and the model makes real-time predictions based on the input data to determine whether the current temperature state is abnormal. When the abnormal probability output by the model exceeds a preset threshold (e.g., [threshold value]), it is determined that an abnormal temperature situation has occurred.

[0051] 72) Alarm system; When an abnormal temperature is detected, the alarm system is immediately triggered.

[0052] The alarm system can notify power plant maintenance personnel in multiple ways, such as sounding and flashing alarms, and sending SMS notifications to their mobile phones. The SMS messages include detailed information such as the time, location, temperature, and possible causes of the anomaly (generated from the anomaly diagnosis knowledge base). In addition, pop-up windows appear on the monitoring system interface displaying detailed anomaly information, along with corresponding handling suggestions and operation guidelines, to help maintenance personnel respond quickly and handle abnormal situations, ensuring the safe and stable operation of the power plant.

[0053] Example 2 Taking a real distributed photovoltaic power station as an example, before implementing the temperature anomaly detection algorithm of the present invention, the power station used the traditional fixed threshold detection method, which often resulted in false alarms and missed alarms. This caused maintenance personnel to frequently conduct unnecessary inspections and troubleshooting, and also failed to detect some potential temperature anomalies in a timely manner, affecting the power station's power generation efficiency and equipment lifespan.

[0054] This invention, through real-time monitoring and analysis of power plant operation data, successfully and accurately detected multiple temperature anomalies, including temperature rises caused by partial component shading and temperature anomalies caused by inverter heat dissipation failures. Timely alarms and effective fault handling prevented further escalation of the faults, improving the reliability and stability of the power plant while reducing operation and maintenance costs and downtime. After a period of operation, statistics showed that the power plant's power generation efficiency increased by 20%, and the equipment failure rate decreased by 18%, achieving significant economic and social benefits.

[0055] Example 3 The machine learning-based distributed photovoltaic power station temperature anomaly detection system of the present invention includes: The data acquisition module is used to collect data from the photovoltaic power station, including temperature data, environmental data, and electrical parameter data. The preprocessing module is used to preprocess the photovoltaic power station data to obtain preprocessed data; The extraction module is used to extract features from the preprocessed data to obtain features such as temperature change rate, temperature difference between components, and correlation features between temperature and other parameters. The judgment module is used to input the temperature change rate characteristics, the temperature difference characteristics between components, and the correlation characteristics between temperature and other parameters into the trained hybrid model to determine whether the temperature state of the current photovoltaic power station is abnormal.

[0056] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0057] Example 4 A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the machine learning-based distributed photovoltaic power station temperature anomaly detection method. For example, the method includes: collecting photovoltaic power station data, including temperature data, environmental data, and electrical parameter data; preprocessing the photovoltaic power station data to obtain preprocessed data; extracting features from the preprocessed data to obtain temperature change rate features, inter-module temperature difference features, and temperature correlation features with other parameters; and inputting the temperature change rate features, inter-module temperature difference features, and temperature correlation features with other parameters into a trained hybrid model to determine whether the current temperature state of the photovoltaic power station is abnormal. The memory may include main memory, such as high-speed random access memory, or it may also include non-volatile memory, such as at least one disk storage device. The processor, network interface, and memory are interconnected via an internal bus, which can be an industry standard architecture bus, a peripheral component interconnection standard bus, an extended industry standard architecture bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs; specifically, the program may include program code, which includes computer operation instructions. The memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0058] Example 5 A computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a machine learning-based distributed photovoltaic (PV) power plant temperature anomaly detection method. For example, the method includes: collecting PV power plant data, including temperature data, environmental data, and electrical parameter data; preprocessing the PV power plant data to obtain preprocessed data; extracting features from the preprocessed data to obtain temperature change rate features, inter-module temperature difference features, and temperature correlation features with other parameters; and inputting the temperature change rate features, inter-module temperature difference features, and temperature correlation features with other parameters into a trained hybrid model to determine whether the current temperature state of the PV power plant is abnormal. Specifically, the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0059] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0060] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0061] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0062] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0063] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and disclosure of the invention. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0064] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0065] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for detecting temperature anomalies in a distributed photovoltaic power station based on machine learning, characterized in that, include: Collect data from photovoltaic power plants, including temperature data, environmental data, and electrical parameter data. The photovoltaic power station data is preprocessed to obtain preprocessed data; Feature extraction is performed on the preprocessed data to obtain features such as temperature change rate, temperature difference between components, and correlation between temperature and other parameters. The temperature change rate characteristics, inter-module temperature difference characteristics, and temperature correlation characteristics with other parameters are input into the trained hybrid model to determine whether the current temperature state of the photovoltaic power station is abnormal.

2. The method for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning according to claim 1, characterized in that, The temperature data includes temperature information at various detection locations of the photovoltaic module, temperature information inside the inverter, and temperature information inside the combiner box; the environmental data includes the irradiance and wind speed of the photovoltaic power station; and the electrical parameter data includes the current and voltage information of the photovoltaic power station.

3. The method for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning according to claim 1, characterized in that, The process of preprocessing the photovoltaic power station data is as follows: outlier handling is performed on the temperature data in the photovoltaic power station data using the 3σ principle.

4. The method for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning according to claim 1, characterized in that, The correlation coefficient characterizes the correlation between temperature and other parameters. The correlation coefficient is: Where T is the temperature data sequence, X is the data sequence of other parameters, and n is the number of data points. and These are the average values ​​of temperature and other parameters, respectively.

5. The method for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning according to claim 1, characterized in that, The hybrid model is constructed based on long short-term memory networks and support vector machines.

6. The method for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning according to claim 1, characterized in that, The hybrid model is trained based on stochastic gradient descent and backpropagation algorithms.

7. The method for detecting temperature anomalies in distributed photovoltaic power plants based on machine learning according to claim 1, characterized in that, The loss function of the hybrid model during training is: Where N is the number of training samples. Let i be the true label of the i-th sample. This represents the predicted output probability of the model.

8. A distributed photovoltaic power station temperature anomaly detection system based on machine learning, characterized in that, include: The data acquisition module is used to collect data from the photovoltaic power station, including temperature data, environmental data, and electrical parameter data. The preprocessing module is used to preprocess the photovoltaic power station data to obtain preprocessed data; The extraction module is used to extract features from the preprocessed data to obtain features such as temperature change rate, temperature difference between components, and correlation features between temperature and other parameters. The judgment module is used to input the temperature change rate characteristics, the temperature difference characteristics between components, and the correlation characteristics between temperature and other parameters into the trained hybrid model to determine whether the temperature state of the current photovoltaic power station is abnormal.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the machine learning-based distributed photovoltaic power plant temperature anomaly detection method as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the machine learning-based distributed photovoltaic power station temperature anomaly detection method as described in any one of claims 1-7.

Images