Cold-wave weather photovoltaic power generation prediction method under semi-supervised learning

Abstract

The invention discloses a cold-wave weather photovoltaic power generation prediction method under semi-supervised learning, and the method comprises the following steps: (1), obtaining the historical power generation data and the same-period meteorological data of a photovoltaic power station, and defining a cold-wave discrimination standard based on the sudden temperature drop characteristics; (2) detecting and removing abnormal values by adopting Grubbs inspection, complementing missing data through a segmented cubic Hermite interpolation method, and constructing a cold wave small sample data set; (3) performing data expansion on the cold wave small samples by using an SMOTE algorithm, screening high-confidence samples in combination with K-medoids clustering, and training and initializing a CNN-LSTM model by using labeled data; (4) optimizing CNN-LSTM model parameters by using a whale optimization algorithm WOA, and finally outputting a photovoltaic output prediction result and performing error verification; the method is of great significance to technical research in the fields of new energy power prediction, power grid dispatching and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of power prediction, and in particular to a method for predicting photovoltaic power generation in cold wave weather under semi-supervised learning. Background Art

[0002] In recent years, due to the influence of multiple factors such as the intensification of the greenhouse effect, severe weather such as extreme precipitation and extreme weather has occurred frequently across the country, and heavy rain, strong winds, cold waves, etc. have affected the prediction of new energy power generation output. With the further rapid increase in the scale of new energy installed capacity, especially the proposal of the goal of building a new power system, the impact of major weather processes on the balance of the power system has become increasingly prominent. Therefore, how to achieve the prediction of photovoltaic power station output under major weather conditions has become a technical problem that needs to be solved urgently.

[0003] In order to realize the prediction of photovoltaic power generation under severe weather conditions, the existing technology extracts typical meteorological factors by performing grey correlation analysis on the environmental factors of photovoltaic power stations; then extracts the differential load characteristics of power generation according to the improved GMM clustering algorithm, and uses VMD variational mode decomposition to stabilize the power prediction data of photovoltaic power stations; finally, according to the KELM nuclear level limit conversion operation problem, the modal function of each scale is modeled and predicted, but there are the following technical problems: In the existing technical solutions, grey correlation analysis is used to extract key meteorological factors from meteorological data, and the GMM clustering algorithm is used to extract the differential load characteristics of power generation power. However, the existence of outliers and missing values ​​in the data is ignored. Direct use will lead to a decrease in data quality and affect the prediction accuracy.

[0004] Existing technical solutions ignore the problem of scarce relevant photovoltaic data and meteorological data under major weather conditions, lack adaptability to rare weather events, and affect the prediction generalization ability; the VMD variational mode decomposition used requires manual setting of the decomposition layer number, and it is difficult to effectively model short-term power fluctuations under major weather conditions; at the same time, the KELM used is a shallow model, which makes it difficult to capture complex time series relationships. Summary of the invention

[0005] In order to solve the above problems in the prior art, the present invention proposes a method for predicting photovoltaic power generation in cold wave weather under semi-supervised learning, which solves the problems of missing and abnormal data samples; solves the problem of scarcity of sample data under major weather conditions and improves the generalization ability of the model; and adopts a deep learning model to cope with short-term power fluctuations in major weather conditions and capture more complex timing relationships.

[0006] To achieve the purpose of the present invention, the present invention adopts the following technical solution: A method for predicting photovoltaic power generation in cold wave weather under semi-supervised learning according to the present invention comprises the following steps: (1) Obtain historical power generation data of photovoltaic power stations and meteorological data of the same period, define cold wave discrimination criteria based on the characteristics of sudden temperature drop, and divide historical data into regular weather and cold wave weather data sets; (2) Data preprocessing was performed on the cold wave samples. Outliers were detected and removed using the Grubbs test. Missing data were filled in using the piecewise cubic Hermite interpolation method to construct a small sample dataset of cold waves. (3) The SMOTE algorithm was used to expand the data of the small sample of cold waves, and K-medoids clustering was used to screen high-confidence samples. The labeled data was used to train and initialize the CNN-LSTM model. (4) Generate pseudo labels for unlabeled data by initializing the model, fuse the labeled data with high-confidence pseudo-label data, and use the whale optimization algorithm (WOA) to optimize the CNN-LSTM model parameters. Finally, output the PV output prediction results and perform error verification.

[0007] Furthermore, in step (2), the calculation process of the Grubbs test satisfies the following formula: ; ; Where: G is the Grubbs statistic of the data points in the sample under cold wave weather; is the i-th data point of the sample under cold wave weather; N is the total number of data points in the sample under cold wave weather; is the Grubbs test threshold under cold wave weather; t is the critical value distributed according to the sample degrees of freedom N-2 and the significance level.

[0008] Furthermore, in step (2), the implementation process of the piecewise cubic Hermite interpolation method includes: Constructing a time series and photovoltaic characteristic series , Compute the interpolation function: ; ; ; in: T is the time series in the sample under cold wave weather after removing outliers; P is a photovoltaic characteristic sequence in the sample under cold wave weather after removing outliers; is the nth time point in the sample under cold wave weather after removing outliers; is a photovoltaic characteristic quantity corresponding to the nth time point in the sample under cold wave weather after removing outliers; n is a certain location in the sample under cold wave weather after removing outliers; i is a position less than n in the sample under cold wave weather after removing outliers; is the number of samples less than 100% in cold wave weather after removing outliers. a certain point in time; In the sample under cold wave weather after removing outliers The derivative of the function curve at a time point; In the sample under cold wave weather after removing outliers Correction items at time points; In the sample under cold wave weather after removing outliers Basic interpolation of time points; In the sample under cold wave weather after removing outliers Fitted interpolation of time points.

[0009] Furthermore, in step (3), the SMOTE data expansion process specifically includes: calculating samples in the feature space With k nearest neighbor samples The Euclidean distance of: ; ; Generate new samples: ; in: is a sample point in a small sample data set under cold wave weather; The time corresponding to the sample point in the small sample data set under cold wave weather; The sample points of the small sample data set under cold wave weather The corresponding wind direction angle; The sample points of the small sample data set under cold wave weather The corresponding wind speed; The sample points of the small sample data set under cold wave weather The corresponding rainfall; The sample points of the small sample data set under cold wave weather The corresponding temperature; The sample points of the small sample data set under cold wave weather The corresponding humidity; The sample points of the small sample data set under cold wave weather The corresponding air pressure; The sample points of the small sample data set under cold wave weather The corresponding illumination amplitude; is the Euclidean distance between two sample points in the small sample data set under cold wave weather; A random number that controls the distance between the new sample and the original sample, with a value range of .

[0010] Furthermore, in step (3), the K-medoids clustering quality is evaluated using the silhouette coefficient: ; ; ; in: i is a sample in the expanded sample data set under cold wave weather; j is a sample in the expanded sample data set under cold wave weather; is the cluster where sample i is located in the expanded sample data set under cold wave weather; is the total number of samples i in the cluster where the samples are located in the sample data set under the expanded cold wave weather; is the average distance from sample i to other samples in the same cluster in the expanded sample data set under cold wave weather; is a characteristic quantity of sample i in the expanded sample data set under cold wave weather; is the average distance to the nearest cluster of sample i in the expanded sample data set under cold wave weather; is the silhouette coefficient value.

[0011] Furthermore, in step (4), the pseudo labels of the unlabeled data are generated by initializing the model, and the labeled data and the high-confidence pseudo-label data are fused, as follows: the unlabeled cold wave weather sample data is predicted using the initialized trained CNN-LSTM model to obtain the corresponding prediction output; the formula is as follows: ; ; ; in: is the xth feature map of the vth convolutional layer in the model; is the activation function in the model; The feature atlas of the input layer in the model is the xth feature map of the vth convolutional layer in the model and the The weight of the feature map; is the vth convolutional layer in the model. Additive bias of feature maps; is the vth pooling layer in the model feature maps; is the vth pooling layer in the model The multiplicative bias of the feature maps; is the pooling function used in the model; is the vth layer of the LSTM hidden layer in the model. A neuron.

[0012] Furthermore, in step (4), the confidence of all predicted samples is calculated, and the data with confidence higher than the set threshold is screened out, and its predicted label is assigned as a pseudo label; the formula is as follows: ; in: is the variance of the prediction sample; P is the number of samples of the prediction sample; is a predicted value of the predicted sample; k is the number of forward propagation of the unlabeled cold wave weather sample data input value; is the uncertainty threshold; True means the sample is a pseudo label; False means that the sample cannot be used as a pseudo label.

[0013] Furthermore, in step (4), the WOA optimization process is as follows: ; ; ; ; ; in: For the model Head whale solution; D is the dimension of the problem in the model; is the position of the first whale in the model in the Dth dimension; M is the number of whales in the model; is the current optimal solution of the model; Solution for the model The corresponding objective function value; For a certain whale in the model; For the model the new position of the first whale; For the model The current optimal solution for the first whale; A is the coefficient of control exploration behavior in the model; is the distance between the current whale and the optimal solution in the model; C is the coefficient of control development behavior in the model; r is the random function in the model, ranging from ; u is the coefficient that controls the balance between exploration behavior and development behavior in the model, which gradually decreases from the maximum value to 0.

[0014] Furthermore, in step (4), the error verification indicators include: ; ; ; in: The number of sample points predicted by the model; To use the model to predict The predicted value of sample points; For the The actual value of the sample points; is the average of the actual values.

[0015] Furthermore, the semi-supervised learning process specifically includes: using 30% labeled data to train the benchmark model in the initialization stage, retaining high-confidence samples with prediction variance σ²<0.05 in the pseudo-label generation stage, and fusing 50% of the original labeled data and 50% of the pseudo-label data for model optimization in the final training stage.

[0016] Beneficial effects: Compared with the prior art, the Grubbs test used in the present invention can effectively detect and remove outliers, and the segmented cubic Hermite interpolation method can accurately complete missing data and ensure data quality; in the process of small sample data expansion, the SMOTE method combined with K-medoids clustering can optimize data distribution and improve data reliability; in the process of photovoltaic power prediction, the WOA optimized CNN-LSTM model used can effectively extract time series features and improve prediction accuracy. The cold wave weather photovoltaic power generation prediction method proposed in the present invention has the advantages of accurate outlier detection, effective small sample expansion, and high prediction accuracy, which is of great significance for technical research in the fields of new energy power prediction and power grid dispatching. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of the present invention; Figure 2 It is a line graph of the daily maximum and minimum temperatures in November of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further explained and illustrated below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only used to illustrate and explain the present invention, and do not impose any limitation on the scope of implementation of the present invention.

[0019] The embodiment of the present invention provides a method for predicting photovoltaic power generation in cold wave weather under semi-supervised learning, comprising the following steps: (1) Obtain the historical power generation data of the photovoltaic power station from September to November in recent years and the meteorological data of the same period, including but not limited to core meteorological parameters such as temperature and humidity. Then, based on the characteristic trends and mutations of the meteorological data, define the criteria for distinguishing cold wave weather events. After the criteria are clarified, divide the historical data of the photovoltaic power station into normal weather and cold wave weather; like Figure 2 As shown, the criteria for identifying cold wave weather are: when the temperature drops by ≥8℃ and the lowest temperature is ≤5℃ within 24 hours, it is judged that the cold wave has begun; when the temperature continues to be >5℃ for more than 12 hours, it is judged that the cold wave has ended.

[0020] (2) Preprocess the selected cold wave sample data, verify the data through Grubbs test to identify the outliers in the data set, and then use the piecewise cubic Hermite interpolation method to supplement the missing values ​​caused by the above method to complete the establishment of a small sample data set under cold wave weather; (2.1) The Grubbs test is used to screen the samples under the cold wave weather. Assuming that the data is normally distributed, the deviation between the data point and the data mean is compared to determine whether it is an outlier, and the samples under the cold wave weather after the outliers are removed are obtained. The formula of the Grubbs test is as follows: ; ; Where: G is the Grubbs statistic of the data points in the sample under cold wave weather; is the i-th data point of the sample under cold wave weather; N is the total number of data points in the sample under cold wave weather; is the Grubbs test threshold under cold wave weather; t is the critical value distributed according to the sample degrees of freedom N-2 and the significance level.

[0021] (2.2) The missing values ​​of the samples under cold wave weather after removing the outliers are supplemented by the piecewise cubic Hermite interpolation method (PCHIP) to obtain a small sample data set under cold wave weather. The formula of the Hermite interpolation method is as follows: Constructing a time series and photovoltaic characteristic series , Compute the interpolation function: ; ; ; in: T is the time series in the sample under cold wave weather after removing outliers; P is a photovoltaic characteristic sequence in the sample under cold wave weather after removing outliers; is the nth time point in the sample under cold wave weather after removing outliers; is a photovoltaic characteristic quantity corresponding to the nth time point in the sample under cold wave weather after removing outliers; n is a certain location in the sample under cold wave weather after removing outliers; i is a position less than n in the sample under cold wave weather after removing outliers; is the number of samples less than 100% in cold wave weather after removing outliers. a certain point in time; In the sample under cold wave weather after removing outliers The derivative of the function curve at a time point; In the sample under cold wave weather after removing outliers Correction items at time points; In the sample under cold wave weather after removing outliers Basic interpolation of time points; In the sample under cold wave weather after removing outliers Fitted interpolation of time points.

[0022] (3) Based on the SMOTE model, the small sample data is expanded, the expanded samples are merged with the original small sample data, and the data quality is tested using K-medoids clustering. Data samples with higher confidence are preliminarily screened out, and part of them are used as labeled data to train and initialize the CNN-LSTM model; (3.1) For the small sample data set under cold wave weather, by analyzing the original data characteristics, the SMOTE algorithm is used to generate new synthetic sample data based on the K nearest neighbor strategy in the feature space, and the number of samples under cold wave weather is expanded to obtain the expanded sample data set under cold wave weather. Among them, the sample expansion formula based on SMOTE is as follows: ; ; Generate new samples: ; in: is a sample point in a small sample data set under cold wave weather; The time corresponding to the sample point in the small sample data set under cold wave weather; The sample points of the small sample data set under cold wave weather The corresponding wind direction angle; The sample points of the small sample data set under cold wave weather The corresponding wind speed; The sample points of the small sample data set under cold wave weather The corresponding rainfall; The sample points of the small sample data set under cold wave weather The corresponding temperature; The sample points of the small sample data set under cold wave weather The corresponding humidity; The sample points of the small sample data set under cold wave weather The corresponding air pressure; The sample points of the small sample data set under cold wave weather The corresponding illumination amplitude; is the Euclidean distance between two sample points in the small sample data set under cold wave weather; A random number that controls the distance between the new sample and the original sample, with a value range of .

[0023] (3.2) The K-medoids clustering algorithm is used to detect the data quality of the expanded cold wave weather sample data set. According to the distribution of samples in the feature space, the data is divided into several clusters, and the center points are selected as typical samples. The sample data with high confidence are preliminarily screened out, and some of them are labeled to obtain the cold wave weather samples of the labeled data. Among them, the silhouette coefficient value formula for evaluating the quality of the K-medoids clustering algorithm is as follows: ; ; ; in: i is a sample in the expanded sample data set under cold wave weather; j is a sample in the expanded sample data set under cold wave weather; is the cluster where sample i is located in the expanded sample data set under cold wave weather; is the total number of samples i in the cluster where the samples are located in the sample data set under the expanded cold wave weather; is the average distance from sample i to other samples in the same cluster in the expanded sample data set under cold wave weather; is a characteristic quantity of sample i in the expanded sample data set under cold wave weather; is the average distance to the nearest cluster of sample i in the expanded sample data set under cold wave weather; is the silhouette coefficient value.

[0024] (4) Use the preliminarily trained model to predict the unlabeled data, and then select samples with high confidence as pseudo-labels. Combine the labeled data and pseudo-label data, retrain the CNN-LSTM model optimized by WOA to predict the photovoltaic output, and verify the model accuracy through data indicators.

[0025] (4.1.1) Use the initialized and trained CNN-LSTM model to predict the unlabeled cold wave weather sample data and obtain the corresponding prediction output. The relevant formula of the CNN-LSTM model is as follows: ; ; ; in: is the xth feature map of the vth convolutional layer in the model; is the activation function in the model; the present invention selects the Relu function; It is the feature atlas of the input layer in the model; is the xth feature map of the vth convolutional layer in the model and the The weight of the feature map is the vth convolutional layer in the model Additive bias of feature maps; is the vth pooling layer in the model feature maps; is the vth pooling layer in the model The multiplicative bias of the feature maps; is the pooling function used in the model; is the vth layer of the LSTM hidden layer in the model. A neuron.

[0026] (4.1.2) After obtaining the corresponding model output, the confidence of all predicted samples is calculated based on the confidence evaluation mechanism, and the data with confidence higher than the set threshold is screened out, and its predicted label is assigned as a pseudo label. Among them, the uncertainty estimation formula based on the prediction result is as follows: ; in: is the variance of the prediction sample; P is the number of samples of the prediction sample; is a predicted value of the predicted sample; k is the number of forward propagation of the unlabeled cold wave weather sample data input value; is the uncertainty threshold; True means the sample is a pseudo label; False means that the sample cannot be used as a pseudo label.

[0027] (4.2.1) The selected high-confidence pseudo-label samples are merged with the original labeled data, and the CNN-LSTM model based on WOA optimization is used to find the optimal network parameters through iterative updates to predict the photovoltaic output. Among them, the formula of the CNN-LSTM model based on WOA optimization is as follows: ; ; ; ; ; in: For the model Head whale solution; D is the dimension of the problem in the model; is the position of the first whale in the model in the Dth dimension; M is the number of whales in the model; is the current optimal solution of the model; Solution for the model The corresponding objective function value; For a certain whale in the model; For the model the new position of the first whale; For the model The current optimal solution for the first whale; A is the coefficient of control exploration behavior in the model; is the distance between the current whale and the optimal solution in the model; C is the coefficient of control development behavior in the model; r is the random function in the model, ranging from ; u is the coefficient that controls the balance between exploration behavior and development behavior in the model, which gradually decreases from the maximum value to 0.

[0028] (4.2.2) The photovoltaic output prediction results of the CNN-LSTM model optimized by WOA are error checked. The present invention mainly uses indicators such as mean absolute error (MAE), root mean square error (RMSE), and determination coefficient for evaluation. The relevant calculation formula is as follows: ; ; ; in: The number of sample points predicted by the model; To use the model to predict The predicted value of sample points; For the The actual value of the sample points; is the average of the actual values.

Claims

1. A method for predicting photovoltaic power generation in cold weather under semi-supervised learning, characterized in that: The following steps are involved: (1) Obtain historical power generation data of photovoltaic power stations and meteorological data of the same period, define cold wave discrimination criteria based on the characteristics of sudden temperature drop, and divide historical data into regular weather and cold wave weather data sets; (2) Data preprocessing was performed on the cold wave samples. Outliers were detected and removed using the Grubbs test. Missing data were filled in using the piecewise cubic Hermite interpolation method to construct a small sample dataset of cold waves. (3) The SMOTE algorithm was used to expand the data of the small sample of cold waves, and K-medoids clustering was used to screen high-confidence samples. The labeled data was used to train and initialize the CNN-LSTM model. (4) Generate pseudo labels for unlabeled data by initializing the model, fuse the labeled data with high-confidence pseudo-label data, and use the whale optimization algorithm (WOA) to optimize the CNN-LSTM model parameters. Finally, output the PV output prediction results and perform error verification.

2. According to the method for predicting photovoltaic power generation in cold weather under semi-supervised learning in claim 1, it is characterized in that: In step (2), the calculation process of the Grubbs test satisfies the following formula: ； ； Where: G is the Grubbs statistic of the data points in the sample under cold wave weather; is the i-th data point of the sample under cold wave weather; N is the total number of data points in the sample under cold wave weather; is the Grubbs test threshold under cold wave weather; t is the critical value distributed according to the sample degrees of freedom N-2 and the significance level.

3. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1 is characterized in that: In step (2), the implementation process of the piecewise cubic Hermite interpolation method includes: Constructing a time series and photovoltaic characteristic series , Compute the interpolation function: ； ； ； in: T is the time series in the sample under cold wave weather after removing outliers; P is a photovoltaic characteristic sequence in the sample under cold wave weather after removing outliers; is the nth time point in the sample under cold wave weather after removing outliers; is a photovoltaic characteristic quantity corresponding to the nth time point in the sample under cold wave weather after removing outliers; n is a certain location in the sample under cold wave weather after removing outliers; i is a position less than n in the sample under cold wave weather after removing outliers; is the number of samples less than 100% in cold wave weather after removing outliers. a certain point in time; In the sample under cold wave weather after removing outliers The derivative of the function curve at a time point; In the sample under cold wave weather after removing outliers Correction items at time points; In the sample under cold wave weather after removing outliers Basic interpolation of time points; In the sample under cold wave weather after removing outliers Fitted interpolation of time points.

4. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1 is characterized in that: In step (3), the SMOTE data expansion process specifically includes: calculating samples in the feature space With k nearest neighbor samples The Euclidean distance of: ； ； Generate new samples: ； in: is a sample point in a small sample data set under cold wave weather; The time corresponding to the sample point in the small sample data set under cold wave weather; The sample points of the small sample data set under cold wave weather The corresponding wind direction angle; The sample points of the small sample data set under cold wave weather The corresponding wind speed; The sample points of the small sample data set under cold wave weather The corresponding rainfall; The sample points of the small sample data set under cold wave weather The corresponding temperature; The sample points of the small sample data set under cold wave weather The corresponding humidity; The sample points of the small sample data set under cold wave weather The corresponding air pressure; The sample points of the small sample data set under cold wave weather The corresponding illumination amplitude; is the Euclidean distance between two sample points in the small sample data set under cold wave weather; A random number that controls the distance between the new sample and the original sample, with a value range of .

5. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1, characterized in that: In step (3), the silhouette coefficient is used to evaluate the quality of K-medoids clustering: ； ； ； in: i is a sample in the expanded sample data set under cold wave weather; j is a sample in the expanded sample data set under cold wave weather; is the cluster where sample i is located in the expanded sample data set under cold wave weather; is the total number of samples i in the cluster where the samples are located in the sample data set under the expanded cold wave weather; is the average distance from sample i to other samples in the same cluster in the expanded sample data set under cold wave weather; is a characteristic quantity of sample i in the expanded sample data set under cold wave weather; is the average distance to the nearest cluster of sample i in the expanded sample data set under cold wave weather; is the silhouette coefficient value.

6. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1, characterized in that: In step (4), the pseudo labels of the unlabeled data are generated by initializing the model, and the labeled data and the high-confidence pseudo-label data are fused. Specifically, the unlabeled cold wave weather sample data is predicted using the initialized trained CNN-LSTM model to obtain the corresponding prediction output; the formula is as follows: ； ； ； in: is the xth feature map of the vth convolutional layer in the model; is the activation function in the model; It is the feature atlas of the input layer in the model; is the xth feature map of the vth convolutional layer in the model and the The weight of the feature map; is the vth convolutional layer in the model. Additive bias of feature maps; is the vth pooling layer in the model feature maps; is the vth pooling layer in the model The multiplicative bias of the feature maps; is the pooling function used in the model; is the vth layer of the LSTM hidden layer in the model. A neuron.

7. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1, characterized in that: In step (4), the confidence of all predicted samples is calculated, and the data with confidence higher than the set threshold is screened out, and its predicted label is assigned as a pseudo label; the formula is as follows: ； in: is the variance of the prediction sample; P is the number of samples of the prediction sample; is a predicted value of the predicted sample; k is the number of forward propagation of the unlabeled cold wave weather sample data input value; is the uncertainty threshold; True means the sample is a pseudo label; False means that the sample cannot be used as a pseudo label.

8. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1, characterized in that: In step (4), the WOA optimization process is as follows: ； ； ； ； ； in: For the model Head whale solution; D is the dimension of the problem in the model; is the position of the first whale in the model in the Dth dimension; M is the number of whales in the model; is the current optimal solution of the model; Solution for the model The corresponding objective function value; For a certain whale in the model; For the model the new position of the first whale; For the model The current optimal solution for the first whale; A is the coefficient of control exploration behavior in the model; is the distance between the current whale and the optimal solution in the model; C is the coefficient of control development behavior in the model; r is the random function in the model, ranging from ; u is the coefficient that controls the balance between exploration behavior and development behavior in the model, which gradually decreases from the maximum value to 0.

9. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1, characterized in that: In step (4), the error verification indicators include: ； ； ； in: The number of sample points predicted by the model; To use the model to predict The predicted value of sample points; For the The actual value of the sample points; is the average of the actual values.

10. The method for predicting photovoltaic power generation in cold weather under semi-supervised learning according to claim 1, characterized in that: The semi-supervised learning process specifically includes: using 30% labeled data to train the benchmark model in the initialization phase, retaining high-confidence samples with prediction variance σ²<0.05 in the pseudo-label generation phase, and integrating 50% of the original labeled data and 50% of the pseudo-label data for model optimization in the final training phase.

Images