New energy power generation multi-disaster impact analysis method and system in extreme weather

Abstract

The invention discloses a new energy power generation multi-disaster influence analysis method and system in extreme weather. The method comprises the following steps: constructing a random forest regression model according to meteorological data containing a target meteorological factor and normalized new energy power generation power data; an improved Bayesian optimization method is introduced, a decision coefficient of the random forest regression model on a verification set is used as a target function, hyper-parameters of the random forest regression model are optimized, the hyper-parameters comprise the number of decision trees, the maximum depth and the minimum split sample number, and a disaster type superposition influence rate model is obtained; and inputting real-time meteorological data into the disaster type superposition influence rate model, and outputting by the disaster type superposition influence rate model to obtain new energy power generation power corresponding to the real-time meteorological data. The influence of extreme weather on the new energy power generation efficiency is comprehensively predicted, and improved Bayesian optimization is introduced to optimize parameters in the random forest, so that the prediction accuracy is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy power generation, and particularly relates to a method, a system and a readable storage medium for analyzing the influence of multiple disasters on new energy power generation under extreme weather conditions. Background Art

[0002] The impact of extreme weather events on new energy power generation systems has become a key factor restricting their stable operation and development. For example, cold snaps may cause icing on wind turbines, fog may reduce solar radiation, and snowfall may cover photovoltaic panels, all of which seriously affect power generation efficiency and system safety. Existing research mainly focuses on the impact of single weather events on specific new energy power generation methods, lacking a systematic analysis method that comprehensively considers multiple extreme weather conditions. For example, the impact of cold snaps, strong winds, and snowfall on wind power generation is analyzed separately; the impact of extreme weather conditions such as sandstorms, heavy rains, and snowfall on photovoltaic power generation is discussed separately under extreme weather conditions.

[0003] The methods adopted above mostly analyze the impact of single weather factors on new energy power generation power. However, for the three extreme weather conditions of cold snaps, fog, and snowfall, they are very likely to occur simultaneously rather than individually. If the method of analyzing single weather elements is adopted, it is difficult to comprehensively evaluate and predict the comprehensive impact of extreme weather on the entire new energy power generation system. Summary of the Invention

[0004] The present invention provides a method and a system for analyzing the influence of multiple disasters on new energy power generation under extreme weather conditions, which are used to solve the technical problem that it is difficult to comprehensively evaluate and predict the comprehensive impact of extreme weather on the entire new energy power generation system by adopting the method of analyzing single weather elements.

[0005] In a first aspect, the present invention provides a method for analyzing the influence of multiple disasters on new energy power generation under extreme weather conditions, including: Obtain meteorological data and new energy power generation power data, wherein the meteorological data includes meteorological factors, and the new energy power generation power data includes new energy power generation power; Determine the linear correlation degree between the meteorological factors and the new energy power generation power according to the Pearson correlation coefficient, screen out the meteorological factors with the absolute value of the linear correlation degree greater than a preset threshold, and define them as target meteorological factors; Construct a random forest regression model according to the meteorological data including the target meteorological factors and the normalized new energy power generation power data; Introduce an improved Bayesian optimization method, use the determination coefficient of the random forest regression model on the validation set as the objective function, optimize the hyperparameters of the random forest regression model, and the hyperparameters include the number of decision trees, the maximum depth, and the minimum number of samples for splitting, to obtain a disaster superposition impact rate model; The real-time meteorological data is input into the disaster superposition impact rate model, and the disaster superposition impact rate model outputs the new energy power generation corresponding to the real-time meteorological data.

[0006] In a second aspect, the present invention provides a system for analyzing the impact of multiple disasters on renewable energy power generation under extreme weather conditions, comprising: an acquisition module configured to acquire meteorological data and renewable energy power generation data, wherein the meteorological data includes meteorological factors, and the renewable energy power generation data includes renewable energy power generation; a screening module configured to determine a linear correlation between the meteorological factor and the renewable energy power generation according to a Pearson correlation coefficient, screen out meteorological factors whose absolute value of the linear correlation is greater than a preset threshold, and define them as target meteorological factors; A construction module is configured to construct a random forest regression model based on the meteorological data including the target meteorological factor and the normalized new energy power generation data; An optimization module is configured to introduce an improved Bayesian optimization method, using the coefficient of determination of the random forest regression model on the validation set as an objective function, to optimize the hyperparameters of the random forest regression model, including the number of decision trees, the maximum depth, and the minimum number of split samples, to obtain a disaster superposition impact rate model; The output module is configured to input real-time meteorological data into the disaster superposition impact rate model, and the disaster superposition impact rate model outputs the new energy power generation power corresponding to the real-time meteorological data.

[0007] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the steps of the method for analyzing the impact of multiple disasters on new energy power generation under extreme weather conditions according to any embodiment of the present invention.

[0008] In a fourth aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the program instructions are executed by a processor, the processor executes the steps of the method for analyzing the impact of multiple disasters on renewable energy power generation under extreme weather conditions according to any embodiment of the present invention.

[0009] The method and system for analyzing the impacts of multiple disasters on new energy power generation under extreme weather of the present application are based on actual meteorological data and corresponding new energy power generation data, complete the identification of extreme weather and establish a mathematical model between extreme weather and new energy power generation, analyze the impacts of extreme weather on new energy power generation. In addition, a random forest model will be established, with various meteorological factors and historical power generation data as basic data, comprehensively predict the impacts of extreme weather on the efficiency of new energy power generation, and introduce improved Bayesian optimization to optimize the parameters in the random forest to improve the accuracy of prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0011] Figure 1 It is a flowchart of a method for analyzing the impacts of multiple disasters on new energy power generation under extreme weather provided by an embodiment of the present invention; Figure 2 It is a numerical graph of four meteorological elements collected by an embodiment of the present invention; Figure 3 It is a graph of new energy power generation data collected by an embodiment of the present invention; Figure 4 It is a fitting and prediction graph of a random forest for wind power generation according to an embodiment of the present invention; Figure 5 It is a fitting and prediction graph of a random forest for photovoltaic power generation according to an embodiment of the present invention; Figure 6 It is a structural block diagram of a system for analyzing the impacts of multiple disasters on new energy power generation under extreme weather provided by an embodiment of the present invention; Figure 7 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0013] Please refer to Figure 1 , which shows a flowchart of a method for analyzing the impacts of multiple disasters on new energy power generation under extreme weather of the present application.

[0014] As Figure 1 shown, the multi-hazard impact analysis method for new energy power generation under extreme weather specifically includes the following steps: Step S101: Obtain meteorological data and new energy power generation data. Among them, the meteorological data contains meteorological factors, and the new energy power generation data contains new energy power generation.

[0015] In this step, data collection mainly includes the collection of meteorological data and power generation data. Among them, meteorological data includes historical meteorological data of cold snaps, heavy fog, and snowfall, such as temperature, humidity, wind speed, visibility, snowfall, etc.; power generation data is the power values of wind power and photovoltaic power generation during the corresponding time period.

[0016] Data cleaning: Data cleaning mainly processes missing values and outliers in the collected data. If there are few missing data, you can choose to directly delete the rows with missing values and outliers; if there are a lot of missing data, use the mean value to fill them.

[0017] Step S102: Determine the linear correlation degree between the meteorological factors and the new energy power generation according to the Pearson correlation coefficient, screen out the meteorological factors whose absolute value of the linear correlation degree is greater than the preset threshold, and define them as target meteorological factors.

[0018] In this step, analyze the relationship between each meteorological element data and new energy power generation data, use the Pearson correlation coefficient to measure the relationship between two variables, and select the variables related to new energy power generation (|Pearson correlation coefficient|≥a, a takes 0.2 - 0.5). Its calculation formula is: , In the formula, is the Pearson correlation coefficient value, and are the corresponding meteorological data and new energy power generation data, and are their average values respectively. The Pearson correlation coefficient value is a value between -1 and 1. When it is close to 1 or -1, it indicates strong positive and negative correlations, and when it is close to 0, it indicates no linear correlation.

[0019] After completing the selection of meteorological factors, normalize these related variables and new energy power generation data for subsequent modeling processing. The processing process is: , In the formula, is the normalized value, and are the minimum and maximum values in the data respectively.

[0020] Step S103: Construct a random forest regression model based on the meteorological data containing the target meteorological factor and the normalized new energy power generation data.

[0021] In this step, before establishing the random forest regression model according to the obtained meteorological data and new energy power generation data, it is also necessary to identify extreme weather, and its judgment criteria include: For cold snaps, the cold snap criteria adopted in the north are: a temperature drop of more than 10°C in 24 hours, or a temperature drop of more than 12°C in 48 hours, and at the same time, the minimum temperature is lower than 4°C; the cold snap criteria adopted in the south are: a temperature drop of more than 8°C in 24 hours, or a temperature drop of more than 10°C in 48 hours, and at the same time, the minimum temperature is lower than 5°C. The Guangdong Meteorological Bureau defines the cold snap criteria as: a daily average temperature drop of more than 8°C in one day, or a sharp drop in the average temperature of more than 10°C in two days, and at the same time, the minimum temperature drops to 5°C or below.

[0022] For heavy fog, when the visibility is between 500 - 1000 meters, it is fog; when the visibility is between fifty - 500 meters, it is thick fog; when the visibility is less than 50 meters, it is called strong thick fog. Generally, the heavy fog we mentioned refers to thick fog and strong thick fog when the visibility is less than 500 meters.

[0023] For snowfall: Taking the 24 - hour precipitation as the division standard, among them, the precipitation of 0.1 - 2.4 mm is light snow, 1.3 - 3.7 mm is light to moderate snow, 2.5 - 4.9 mm is moderate snow, 3.8 - 7.4 mm is moderate to heavy snow, reaching 5.0 - 9.9 mm is heavy snow, 7.5 - 14.9 mm is heavy to blizzard, and the precipitation reaching or exceeding 10 mm is blizzard.

[0024] Specifically, constructing the random forest regression model includes: Sample selection: Randomly draw multiple sub - samples from the meteorological data and new energy power generation data. Among them, multiple sub - samples are allowed to be repeated, that is, a data point can appear multiple times in the same sub - sample; Constructing decision trees: For each sub - sample obtained by random sampling, independently construct a decision tree. Among them, when constructing each node of the decision tree, instead of selecting the optimal feature from all features, randomly select a part of the features, and then select the optimal feature from them for splitting; Splitting of decision trees: Use the selected features to split at each decision tree node. Commonly used splitting criteria include information gain or Gini impurity. Among them, the expression of Gini impurity is: , In the formula, is the Gini impurity, is the proportion of the category at the splitting node, is the total number of categories.

[0025] The expression for information gain is: , In the formula, is the dataset with splitting as the splitting attribute, is the attribute all possible values of, is when the attribute takes at this time subset of, is the dataset entropy before classification, represents the entropy of the subset when the attribute takes at this time subset of, represents the information gain obtained by partitioning the dataset on the attribute .

[0026] Result integration: For regression problems, calculate the average of the predicted values. Among them, for the regression task, the average formula is as follows: , In the formula, is the average of the predicted values, is the th decision tree's predicted value for the sample , is the total number of decision trees; Combine each regression decision tree to obtain the corresponding random forest model. Among them, each tree gives an independent predicted value for the sample . The final result is calculated by averaging the outputs of all trees, thereby improving the overall prediction accuracy and reducing the risk of overfitting. represents the integrated regression prediction process of the random forest for the input .

[0027] Step S104, introduce an improved Bayesian optimization method, use the coefficient of determination of the random forest regression model on the validation set as the objective function, optimize the hyperparameters of the random forest regression model. The hyperparameters include the number of decision trees, the maximum depth, and the minimum number of splitting samples, to obtain the disaster superposition influence rate model.

[0028] In this step, with the coefficient of determination of the random forest on the validation set being the maximum as the optimization goal, construct the objective function; Randomly sample hyperparameter combinations from the hyperparameter space ​ , evaluate the corresponding function values , construct an initial dataset, with the expression: , In the formula, is the initial dataset, is the initial number of sampling points, is the th sampling hyperparameter combination, is 's response value on the evaluation function.

[0029] Using the dataset , establish a Gaussian process regression model as an approximation of the objective function, with the expression: , In the formula, is the response value of the objective function under the hyperparameter combination , is the mean and variance predicted by the Gaussian process under the hyperparameter combination , is the average value under the hyperparameter combination , is the variance under the hyperparameter combination .

[0030] Select the expected improvement function as the acquisition function to measure the value of sampling new points under the current surrogate model; Based on the improved acquisition function, select the current optimal test point, with the expression: , In the formula, represents the currently selected optimal test point, that is, the hyperparameter combination that maximizes the acquisition function value, is the parameter that makes the function achieve the maximum value , is the numerical value of the acquisition function of the expected improvement, used to guide the selection of the next sampling point.

[0031] Apply the sampled to the training and performance evaluation of the random forest regression model to obtain the true objective function value , and update the dataset, with the expression: , In the formula, represents the true evaluation value of the objective function under the hyperparameter combination .

[0032] Repeat the modeling and sampling process, iteratively update the surrogate model and the dataset, and finally obtain the optimal hyperparameter combination. Then, construct the disaster superposition impact rate model according to the optimal hyperparameter combination. The expression of the optimal hyperparameter combination is: , where, represents the parameter that makes the function reach the minimum value . represents the optimal hyperparameter combination, that is, the parameter that makes the objective function achieve the minimum value.

[0033] It should be noted that the expression of the objective function is: , where, is the objective function value, is the actual power generation value at the th moment, is the model prediction value at the th moment, is the sample weight, is the physical factor, which is the wind speed for wind power generation and the radiation amount for photovoltaic power generation; is an indicator function, which takes the value of 1 when the condition is met and 0 when not. This method is used to avoid the situation that the power generation prediction is still positive when the physical conditions are insufficient. The optimization objective is to minimize the objective function.

[0034] The expression of the acquisition function is: , , where, is the optimal objective in the current observation point, is the exploration function, which is used to control the trade-off between exploration and exploitation, is the cumulative distribution function of the standard normal distribution, is the predicted standard deviation of the Gaussian process regression model at the point , is the probability density function of the standard normal distribution, is the standardized variable, which measures the improvement amplitude of the model prediction at the point ; is the uncertainty adjustment factor, which can be selected through parameter optimization, is the physical factor, which is the wind speed for wind power generation and the radiation amount for photovoltaic power generation; is an indicator function that takes the value of 1 when the condition is met and 0 when the condition is not met. is the predicted mean.

[0035] Step S105: Input the real-time meteorological data into the superimposed impact rate model of disaster types, and the superimposed impact rate model of disaster types outputs the new energy power generation corresponding to the real-time meteorological data.

[0036] In summary, comprehensively consider the impacts of various extreme weather conditions (such as cold snaps, fog, rainfall, etc.) on new energy power generation. The method of this application is based on actual meteorological data and corresponding new energy power generation data, completes the identification of extreme weather and establishes a mathematical model between extreme weather and new energy power generation, analyzes the impact of extreme weather on new energy power generation. In addition, a random forest model will be established, using various meteorological factors and historical power generation data as basic data to comprehensively predict the impact of extreme weather on new energy power generation efficiency, and introducing improved Bayesian optimization to optimize the parameters in the random forest to improve the accuracy of prediction.

[0037] In a specific embodiment, a total of 450 days of meteorological data and new energy power generation data are selected. The numerical graphs of each meteorological data and the new energy power generation data graph are as Figure 2 , Figure 3 shown. When training the random forest model, the ratio of the training set to the validation set is 8:2. After Bayesian optimization, it shows good results in both wind power generation and photovoltaic power generation. In photovoltaic power generation, the mean squared error (MSE) on the training set is 4.99, and the coefficient of determination is 0.997. The MSE on the test set is 56.78, is 0.969, and the optimal parameters are: n_estimators = 73, max_depth = 19, min_samples_split = 3; in wind power generation, the mean squared error (MSE) on the training set is 0.3878, and the coefficient of determination is 0.95. The MSE on the test set is 1.7574, is 0.84, and the optimal parameters are n_estimators (the number of decision trees in the random forest) = 50, max_depth (the maximum tree depth) = 13, min_samples_split (the minimum number of sample splits) = 2, which also shows good performance.

[0038] As Figure 4 and Figure 5 shown, after Bayesian optimization, the random forest model shows good performance in the simulation and prediction of wind power generation and photovoltaic power generation. In wind power generation, the mean squared error (MSE) on the training set is 0.3878, and the coefficient of determination is 0.95, and the MSE on the test set is 1.7574, is 0.84; during photovoltaic power generation, the mean squared error (MSE) on the training set is 4.99, and the coefficient of determination is 0.997, and the MSE on the test set is 56.78, is 0.969.

[0039] Please refer to Figure 6 , which shows the structural block diagram of a multi-disaster impact analysis system for new energy power generation under extreme weather conditions of the present application.

[0040] As Figure 6 shown, the multi-disaster impact analysis system 200 for new energy power generation under extreme weather conditions includes an acquisition module 210, a screening module 220, a construction module 230, an optimization module 240, and an output module 250.

[0041] Among them, the acquisition module 210 is configured to acquire meteorological data and new energy power generation power data, where the meteorological data includes meteorological factors, and the new energy power generation power data includes new energy power generation power; the screening module 220 is configured to determine the linear correlation degree between the meteorological factors and the new energy power generation power according to the Pearson correlation coefficient, screen out the meteorological factors with the absolute value of the linear correlation degree greater than a preset threshold, and define them as target meteorological factors; the construction module 230 is configured to construct a random forest regression model according to the meteorological data including the target meteorological factors and the normalized new energy power generation power data; the optimization module 240 is configured to introduce an improved Bayesian optimization method, use the coefficient of determination of the random forest regression model on the validation set as the objective function, optimize the hyperparameters of the random forest regression model, and the hyperparameters include the number of decision trees, the maximum depth, and the minimum number of split samples, to obtain a disaster superposition impact rate model; the output module 250 is configured to input real-time meteorological data into the disaster superposition impact rate model, and the disaster superposition impact rate model outputs the new energy power generation power corresponding to the real-time meteorological data.

[0042] It should be understood that Figure 6 the various modules described in Figure 1 correspond to the respective steps in the method described in reference Figure 6 . Therefore, the operations, features, and corresponding technical effects described above for the method also apply to

[0043] the various modules in As an implementation, the computer-readable storage medium of the present invention stores computer-executable instructions, which are set as follows: Obtain meteorological data and new energy power generation data, wherein the meteorological data contains meteorological factors, and the new energy power generation data contains new energy power generation; Determine the linear correlation degree between the meteorological factors and the new energy power generation according to the Pearson correlation coefficient, and screen out the meteorological factors whose absolute value of the linear correlation degree is greater than a preset threshold, and define them as target meteorological factors; Construct a random forest regression model according to the meteorological data containing the target meteorological factors and the normalized new energy power generation data; Introduce an improved Bayesian optimization method, use the coefficient of determination of the random forest regression model on the validation set as the objective function, and optimize the hyperparameters of the random forest regression model. The hyperparameters include the number of decision trees, the maximum depth, and the minimum number of samples for splitting, to obtain a disaster superposition impact rate model; Input the real-time meteorological data into the disaster superposition impact rate model, and the disaster superposition impact rate model outputs the new energy power generation corresponding to the real-time meteorological data.

[0044] The computer-readable storage medium may include a storage program area and a storage data area. Among them, the storage program area can store an operating system and application programs required for at least one function; the storage data area can store data created according to the use of the new energy power generation multi-disaster impact analysis system under extreme weather. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the computer-readable storage medium may optionally include a memory remotely set relative to the processor, and these remote memories can be connected to the new energy power generation multi-disaster impact analysis system under extreme weather through a network. Examples of the above networks include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0045] Figure 7 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, as Figure 7 shown. The device includes: a processor 310 and a memory 320. The electronic device may further include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330, and the output device 340 can be connected through a bus or other means. Figure 7Take the bus connection as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions, and modules stored in the memory 320, that is, to implement the method for analyzing the impact of multiple disasters on new energy power generation under extreme weather in the above method embodiment. The input device 330 can receive input digital or character information, and generate key signal inputs related to the user settings and function control of the system for analyzing the impact of multiple disasters on new energy power generation under extreme weather. The output device 340 may include display devices such as a display screen.

[0046] The above electronic device can execute the method provided by the embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method. For technical details not described in detail in this embodiment, reference can be made to the method provided by the embodiment of the present invention.

[0047] As an implementation manner, the above electronic device is applied to the system for analyzing the impact of multiple disasters on new energy power generation under extreme weather and is used for the client, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: Obtain meteorological data and new energy power generation power data, wherein the meteorological data includes meteorological factors, and the new energy power generation power data includes new energy power generation power; Determine the linear correlation degree between the meteorological factor and the new energy power generation power according to the Pearson correlation coefficient, screen out the meteorological factors whose absolute value of the linear correlation degree is greater than a preset threshold, and define them as target meteorological factors; Construct a random forest regression model according to the meteorological data including the target meteorological factor and the normalized new energy power generation power data; Introduce an improved Bayesian optimization method, use the determination coefficient of the random forest regression model on the validation set as the objective function, and optimize the hyperparameters of the random forest regression model, where the hyperparameters include the number of decision trees, the maximum depth, and the minimum number of samples for splitting, to obtain a disaster superposition impact rate model; Input the real-time meteorological data into the disaster superposition impact rate model, and the disaster superposition impact rate model outputs the new energy power generation power corresponding to the real-time meteorological data.

[0048] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiments.

[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A method for analyzing the impacts of multiple disasters on new energy power generation under extreme weather, characterized in that, Including: Obtain meteorological data and new energy power generation data, wherein the meteorological data contains meteorological factors, and the new energy power generation data contains new energy power generation; Determine the linear correlation degree between the meteorological factors and the new energy power generation according to the Pearson correlation coefficient, screen out the meteorological factors with the absolute value of the linear correlation degree greater than a preset threshold, and define them as target meteorological factors; Construct a random forest regression model according to the meteorological data containing the target meteorological factors and the normalized new energy power generation data; Introduce an improved Bayesian optimization method, use the determination coefficient of the random forest regression model on the validation set as the objective function, optimize the hyperparameters of the random forest regression model, and the hyperparameters include the number of decision trees, the maximum depth, and the minimum number of splitting samples, to obtain a disaster superposition influence rate model; Input the real-time meteorological data into the disaster superposition influence rate model, and the disaster superposition influence rate model outputs the new energy power generation corresponding to the real-time meteorological data.

2. The method for analyzing the impacts of multiple disasters on new energy power generation in extreme weather according to claim 1, wherein The constructing a random forest regression model according to the meteorological data containing the target meteorological factors and the normalized new energy power generation data includes: Sample selection: Randomly extract multiple sub-samples from the meteorological data and the new energy power generation data, wherein multiple sub-samples are allowed to be repeated, that is, a data point can appear multiple times in the same sub-sample; Construct decision trees: For each randomly sampled sub-sample, independently construct a decision tree. When constructing each node of the decision tree, instead of selecting the optimal feature from all features, randomly select a part of the features, and then select the optimal feature from them for splitting; Splitting of decision trees: Use the selected features to split at each decision tree node. Commonly used splitting criteria include information gain or Gini impurity. Among them, the expression of Gini impurity is: ， Wherein, is the Gini impurity, is the proportion of the th class in the splitting node, is the total number of classes; The expression of information gain is: ， In the formula, is the dataset with splitting as the splitting attribute, is the attribute for all possible values, is when the attribute takes at this time subset, is the dataset entropy before classification, represents the entropy of the subset when the attribute takes at this time subset, represents the information gain obtained by partitioning the dataset on the attribute ; Result integration: For regression problems, calculate the average value of the predicted values. For the regression task, its average value formula is as follows: ， Wherein, is the average value of the predicted values, is the -th decision tree's predicted value for the sample , is the total number of decision trees; Combine each regression decision tree to obtain the corresponding random forest model, where each tree gives an independent prediction value for the sample .​ 3. The method for analyzing the multi-hazard impacts on new energy power generation under extreme weather according to claim 1, wherein The introducing an improved Bayesian optimization method, using the determination coefficient of the random forest regression model on the validation set as the objective function, optimizing the hyperparameters of the random forest regression model, and the hyperparameters include the number of decision trees, the maximum depth, and the minimum number of splitting samples, to obtain a disaster superposition influence rate model includes: Taking the coefficient of determination of the random forest in the validation set with the maximum as the optimization objective to construct an objective function; Randomly sample hyperparameter combinations from the hyperparameter space and evaluate the corresponding function values respectively to construct an initial dataset, with the expression: ​ ， In the formula, is the initial data set, is the initial number of sampling points, is the th sampling hyperparameter combination, is the response value on the evaluation function; Using a dataset , a Gaussian process regression model is established as an approximation of the objective function, and the expression is: ， Wherein, is the response value of the objective function under the hyperparameter combination , is the mean and variance predicted by the Gaussian process under the hyperparameter combination , is the average value under the hyperparameter combination , is the variance under the hyperparameter combination . Select the expected improvement function as the acquisition function to measure the value of sampling new points under the current surrogate model; Based on the improved acquisition function, select the current optimal test point, and the expression is: ， In the formula, represents the currently selected optimal test point, that is, the hyperparameter combination that maximizes the acquisition function value, is the parameter that maximizes the function ; , is the value of the acquisition function for expected improvement, which is used to guide the selection of the next sampling point; Apply the sampled to the training and performance evaluation of the random forest regression model to obtain the true objective function value , and update the dataset, with the expression: ， In the formula, represents the true evaluation value of the objective function under the hyperparameter combination ; Repeat the modeling and sampling process, iteratively update the surrogate model and the data set, finally obtain the optimal hyperparameter combination, and construct a disaster superposition influence rate model according to the optimal hyperparameter combination. Among them, the expression of the optimal hyperparameter combination is: ， In the formula, represents the parameter that makes the function reach the minimum value , and represents the optimal hyperparameter combination, that is, the parameter that makes the objective function achieve the minimum value.

4. The method for analyzing the impacts of multiple disasters on new energy power generation under extreme weather according to claim 3, wherein, Wherein, The expression of the objective function is: ， In the formula, is the objective function value, is the actual power generation value at the th moment, is the th moment's model prediction value, is the sample weight, is the weight of each item, and the parameter optimization can be carried out by Bayesian optimization; is the physical factor, which is the wind speed during wind power generation and the radiation amount during photovoltaic power generation; is an index function, which takes the value of 1 when the condition is met and 0 when it is not met.

5. The method for analyzing the impacts of multiple disasters on new energy power generation under extreme weather according to claim 4, wherein The expression of the acquisition function is: ， ， Wherein, is the optimal target in the current observation point, is the exploration function, which is used to control the trade-off between exploration and exploitation, is the cumulative distribution function of the standard normal distribution, is the Gaussian process regression model at point the predicted standard deviation at, is the probability density function of the standard normal distribution, The standardized variable measures the improvement of the model prediction at point ; is the uncertainty adjustment factor, which can be selected by parameter optimization, is the physical factor, the wind speed is selected during wind power generation, and the radiation amount is selected during photovoltaic power generation; is an index function, which takes the value of 1 when the condition is met and 0 when the condition is not met, is the predicted mean.

6. A multi-hazard impact analysis system for new energy power generation under extreme weather, characterized in that, Including: An acquisition module configured to obtain meteorological data and new energy power generation data, wherein the meteorological data contains meteorological factors, and the new energy power generation data contains new energy power generation; A screening module, configured to determine the linear correlation degree between the meteorological factors and the new energy power generation according to the Pearson correlation coefficient, screen out the meteorological factors with the absolute value of the linear correlation degree greater than a preset threshold, and define them as target meteorological factors; A construction module, configured to construct a random forest regression model according to the meteorological data including the target meteorological factors and the normalized new energy power generation data; An optimization module, configured to introduce an improved Bayesian optimization method, use the determination coefficient of the random forest regression model on the validation set as the objective function, optimize the hyperparameters of the random forest regression model, where the hyperparameters include the number of decision trees, the maximum depth, and the minimum number of samples for splitting, to obtain a disaster superposition influence rate model; An output module, configured to input real-time meteorological data into the disaster superposition influence rate model, and the disaster superposition influence rate model outputs the new energy power generation corresponding to the real-time meteorological data.

7. An electronic device, characterized in that, Including: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1 to 5.

Images