Analysis method and system based on high impact meteorological-power event modeling data

Abstract

The invention discloses an analysis method and system based on high impact meteorological-power event modeling data, and belongs to the technical field of electric digital data processing. The method comprises the following steps: performing quantitative judgment on the fan state according to fan state parameters of each fan in a wind power plant to obtain a fan state quantitative judgment result, judging whether to perform fan dynamic processing or not based on the fan state quantitative judgment result, and obtaining each normal fan according to the fan dynamic processing result; performing quantitative judgment on wind speed fluctuation according to the wind speed fluctuation parameter of each normal fan in each monitoring time period in the wind power plant to obtain a wind speed fluctuation quantitative judgment result, and judging whether to perform power generation capacity dynamic correction and ensemble empirical mode decomposition method optimization or not based on the wind speed fluctuation quantitative judgment result; the optimized ensemble empirical mode decomposition method is used for meteorological-power data analysis, and the accuracy of high-impact meteorological-power event modeling data analysis is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric digital data processing, and particularly to an analysis method and system based on high-impact meteorological-electric power event modeling data. Background Art

[0002] With the progress of meteorological science, the accuracy of meteorological prediction has been improved. However, the proportion of meteorologically sensitive loads in the total load has increased, and extreme meteorological events have characteristics such as small probability, high severity consequences, and clustering. Traditional risk assessment methods also have limitations. Under this background, in order to improve the safety of power systems and achieve refined management, etc., it is necessary to carry out research on analysis methods for high-impact meteorological-electric power event modeling data.

[0003] There are various implementation methods for existing analysis methods of high-impact meteorological-electric power event modeling data: one is based on historical meteorological and electric power data, using a time series prediction network model for training, and adjusting parameters to optimize the model in combination with loss values; the second is to collect data through the power grid SCADA (Supervisory Control and Data Acquisition) system, and after processing with big data methods, mine the correlation between meteorology and power grid faults through clustering, classification, or association algorithms; the third is to construct a corresponding model for the impact of extreme meteorological disasters, propose a hybrid simulation method, and build a platform for implementation.

[0004] For example, the method and system for analyzing the operation data of a wind turbine set disclosed in the invention patent announcement with the publication number of CN105677791B includes: obtaining the operation data, status data, and unit fault data of the wind turbine set collected within a predetermined statistical time period as initial data; analyzing these data, extracting attributes related to the change of the operation state of the wind turbine set, and sampling the initial data to obtain sample data of multiple operation state categories; respectively performing statistical analysis on the sample data of multiple operation state categories according to the extracted attributes to obtain multiple operation characteristic quantities of the wind turbine set that meet the predetermined conditions of multiple operation state categories, and each operation characteristic quantity includes the value of a single attribute or the values of multiple attributes; drawing a change diagram of the operation characteristic quantities within a specified state time period according to the obtained multiple operation characteristic quantities to characterize the change of the operation characteristic quantities of the wind turbine set under the specified state.

[0005] For example, the invention patent announcement of the wind turbine health monitoring system based on big data analysis with the announcement number of CN110503131B includes: a feature extraction module, a fault classification module, a black powder rule base, a big data analysis platform, and a system interface module. The fault diagnosis method based on the EMD algorithm and data binning is adopted. First, the IMF signal is decomposed by the EMD algorithm, and its amplitude domain parameter characteristics are extracted as feature vectors and input into the SVM for fault classification, and it is verified by preliminary simulation and actual data. Big data analysis is adopted. Based on the open-source big data platform Spark, the distributed parallel structures of the EMD and statistical description feature operators are respectively implemented.

[0006] However, in the process of implementing the inventive technical solution in the embodiments of the present application, it is found that the above technology has at least the following technical problems: In the prior art, in the context of frequent occurrence of extreme weather and climate events such as typhoons, when using wind renewable energy to obtain the power generation of a wind farm, the extreme wind conditions may affect the accuracy of power generation collection, thereby affecting the accuracy of the power generation processing process, resulting in low analysis accuracy of the modeling data of high-impact meteorological-power events. Summary of the Invention

[0007] The present invention provides an analysis method and system based on high-impact meteorological-power event modeling data, solves the problem in the prior art that in the context of frequent occurrence of extreme weather and climate events such as typhoons, when using wind renewable energy to obtain the power generation of a wind farm, the extreme wind conditions may affect the accuracy of power generation collection, thereby affecting the accuracy of the power generation processing process, resulting in low analysis accuracy of the modeling data of high-impact meteorological-power events, and realizes the improvement of the analysis accuracy of high-impact meteorological-power event modeling data.

[0008] The present invention provides an analysis method based on high-impact meteorological-electricity event modeling data, including the following steps: quantitatively determining the fan status according to the fan status parameters of each fan in the wind farm to obtain the fan status quantitative determination result, determining whether to perform fan dynamic processing based on the fan status quantitative determination result, and obtaining each normal fan according to the fan dynamic processing result. Fan dynamic processing means processing the fan according to the fan status quantitative determination result to reduce the impact of the fan efficiency on the analysis accuracy of high-impact meteorological-electricity event modeling data; quantitatively determining the wind speed fluctuation according to the wind speed fluctuation parameters of each normal fan in each monitoring time period of the wind farm to obtain the wind speed fluctuation quantitative determination result, and determining whether to perform power generation dynamic correction and ensemble empirical mode decomposition method optimization based on the wind speed fluctuation quantitative determination result. Power generation dynamic correction and ensemble empirical mode decomposition method optimization mean respectively correcting the power generation of the normal fan and optimizing the white noise standard deviation in the ensemble empirical mode decomposition method according to the wind speed fluctuation quantitative determination result to reduce the impact of abnormal wind speed fluctuations on the analysis accuracy of high-impact meteorological-electricity event modeling data; using the optimized ensemble empirical mode decomposition method to perform meteorological-electricity data analysis. Meteorological-electricity data analysis means using the AI-driven LLM Agent large model to perform parametric modeling and analysis on the wind power generation data processed by the optimized ensemble empirical mode decomposition method, and constructing a multi-dimensional coupling index system.

[0009] The present invention also provides a system applying the analysis method based on high-impact meteorological-electricity event modeling data, including: a fan processing module, a power generation correction module, a data analysis module, and a meteorological electricity database; wherein, the fan processing module is used to quantitatively determine the fan status according to the fan status parameters of each fan in the wind farm to obtain the fan status quantitative determination result, determine whether to perform fan dynamic processing based on the fan status quantitative determination result, and obtain each normal fan according to the fan dynamic processing result; the power generation correction module is used to quantitatively determine the wind speed fluctuation according to the wind speed fluctuation parameters of each normal fan in each monitoring time period of the wind farm to obtain the wind speed fluctuation quantitative determination result, and determine whether to perform power generation dynamic correction and ensemble empirical mode decomposition method optimization based on the wind speed fluctuation quantitative determination result; the data analysis module is used to perform meteorological-electricity data analysis using the optimized ensemble empirical mode decomposition method. Meteorological-electricity data analysis means using the AI-driven LLM Agent large model to perform parametric modeling and analysis on the wind power generation data processed by the optimized ensemble empirical mode decomposition method, and constructing a multi-dimensional coupling index system.

[0010] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The present invention provides an analysis method based on high-impact meteorological-power event modeling data, thereby effectively reducing the impact of fan anomalies and wind speed fluctuations on the accuracy of high-impact meteorological-power event modeling data analysis. Furthermore, by dynamically processing the fans, dynamically correcting the power generation, and optimizing the ensemble empirical mode decomposition method, the accuracy and reliability of data analysis are improved, providing more accurate data support for the meteorological-power event modeling of wind farms.

[0011] 2. The present invention compares the fan status evaluation value with a preset threshold, and dynamically processes the fan and gives an anomaly prompt according to different status quantification determination results, thereby being able to accurately identify and process normal fans and abnormal fans, reducing the impact of fan status anomalies on the overall power generation efficiency. Furthermore, by dynamically adjusting the pitch angle and yaw angle and timely removing abnormal fans, the operation stability and efficiency of the wind farm are improved, and the overall performance of the wind farm and the accuracy of data analysis are optimized.

[0012] 3. The present invention compares the wind speed fluctuation evaluation value with the wind speed fluctuation threshold, and gives a fluctuation anomaly prompt and corrects the power generation of the fan according to the wind speed fluctuation quantification determination result, thereby being able to timely identify wind speed fluctuation anomalies and dynamically adjust relevant data, reducing the impact of wind speed fluctuations on power generation. Furthermore, by correcting white noise data and matching the power generation correction value, the power generation efficiency and data analysis accuracy of the wind farm are improved, and the operation management of the wind farm is optimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flowchart of the analysis method based on high-impact meteorological-power event modeling data provided by an embodiment of the present application.

[0014] Figure 2 It is a flowchart of the fully automatic analysis of AI-driven LLM Agent meteorological-energy data provided by an embodiment of the present application.

[0015] Figure 3 It is a mind map of the fan status quantification determination provided by an embodiment of the present application.

[0016] Figure 4 It is a mind map of the wind speed fluctuation quantification determination provided by an embodiment of the present application.

[0017] Figure 5 It is a schematic structural diagram of the analysis system based on high-impact meteorological-power event modeling data provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE INVENTION

[0018] Embodiments of the present application provide an analysis method and system based on high-impact meteorological-electricity event modeling data, which solve the problem in the prior art that in the context of frequent occurrence of extreme weather and climate events such as typhoons, when using wind renewable energy to obtain the power generation of a wind farm, the extreme wind conditions may affect the accuracy of power generation collection, thereby affecting the accuracy of the power generation processing process, resulting in low accuracy of the analysis of high-impact meteorological-electricity event modeling data. By quantifying and determining the fan state based on the fan state parameters of each fan in the wind farm, a fan state quantification determination result is obtained. Based on the fan state quantification determination result, it is determined whether to perform fan dynamic processing, and each normal fan is obtained according to the fan dynamic processing result; according to the wind speed fluctuation parameters of each normal fan in each monitoring time period of the wind farm, a wind speed fluctuation quantification determination result is obtained. Based on the wind speed fluctuation quantification determination result, it is determined whether to perform power generation dynamic correction and ensemble empirical mode decomposition method optimization; using the optimized ensemble empirical mode decomposition method for meteorological-electricity data analysis, the accuracy of high-impact meteorological-electricity event modeling data analysis is improved.

[0019] The technical solution in the embodiments of the present application is to solve the above problem that in the context of frequent occurrence of extreme weather and climate events such as typhoons, when using wind renewable energy to obtain the power generation of a wind farm, the extreme wind conditions may affect the accuracy of power generation collection, thereby affecting the accuracy of the power generation processing process, resulting in low accuracy of the analysis of high-impact meteorological-electricity event modeling data. The general idea is as follows: First, based on the operating state parameters of each fan in the wind farm, a quantification determination is performed to generate a fan state level, and dynamic processing is performed on the fan according to the determination result to screen out normal fans to eliminate the interference of low-quality data on modeling; then, a quantification evaluation is performed on the wind speed fluctuation parameters of normal fans in each monitoring time period. If the fluctuation exceeds the threshold, power generation dynamic correction and ensemble empirical mode decomposition method optimization are triggered, thereby reducing the impact of abnormal wind speed on data stability; finally, the optimized ensemble empirical mode decomposition method is used to decompose the meteorological-electricity coupled data, and parametric feature extraction and multi-dimensional analysis are performed on the IMF components through an AI-driven LLM Agent to construct a coupled index system covering meteorological sensitivity, equipment status, and grid response ability, improving the accuracy of high-impact meteorological-electricity event modeling data analysis.

[0020] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0021] As Figure 1As shown in the figure, it is a flowchart of an analysis method based on high-impact meteorological-electricity event modeling data provided by an embodiment of the present application. The method includes the following steps: Quantitatively determine the fan status based on the fan status parameters of each fan in the wind farm to obtain the fan status quantitative determination result. Based on the fan status quantitative determination result, determine whether to perform fan dynamic processing, and obtain each normal fan according to the fan dynamic processing result. Fan dynamic processing means processing the fan according to the fan status quantitative determination result to reduce the impact of the fan efficiency on the analysis accuracy of high-impact meteorological-electricity event modeling data; Quantitatively determine the wind speed fluctuation based on the wind speed fluctuation parameters of each normal fan in each monitoring time period in the wind farm to obtain the wind speed fluctuation quantitative determination result. Based on the wind speed fluctuation quantitative determination result, determine whether to perform power generation dynamic correction and ensemble empirical mode decomposition method optimization. Power generation dynamic correction and ensemble empirical mode decomposition method optimization mean respectively correcting the power generation of the normal fan and optimizing the white noise standard deviation in the ensemble empirical mode decomposition method according to the wind speed fluctuation quantitative determination result to reduce the impact of abnormal wind speed fluctuations on the analysis accuracy of high-impact meteorological-electricity event modeling data; Use the optimized ensemble empirical mode decomposition method to perform meteorological-electricity data analysis. Meteorological-electricity data analysis means using an AI (Artificial Intelligence)-driven LLM Agent (Large Language Model Agent) large model to perform parametric modeling and analysis on the wind power generation data processed by the optimized ensemble empirical mode decomposition method, and constructing a multi-dimensional coupling index system. The wind power generation data includes power generation, temperature, precipitation, relative humidity, wind speed, sunshine hours, etc. in the wind farm.

[0022] In addition, the meteorological electricity database is used to store relevant data of the analysis method based on high-impact meteorological-electricity event modeling data, including: critical wind energy utilization rate, critical wind speed variation coefficient, critical active power volatility, pitch angle adjustment values corresponding to the deviation ranges of each fan status, the first fan status threshold, the second fan status threshold, etc. The data in the meteorological electricity database can be directly queried from public databases such as the meteorological science data center and the power science research institute database, or can be obtained through cooperation with relevant meteorological departments and power companies.

[0023] In this embodiment, the present invention implements hierarchical dynamic processing based on the quantitative determination results of the fan state parameters, optimizes the fan operation state by intelligently adjusting the pitch angle and yaw angle, effectively isolates abnormal fans, and ensures the quality basis of the input data. Secondly, a quantitative determination is made for the wind speed fluctuations of normal fans. By dynamically correcting the power generation data and optimizing the white noise standard deviation parameter in the ensemble empirical mode decomposition method, the interference of abnormal wind speed fluctuations on data analysis is effectively suppressed. Finally, the semantic understanding ability of the LLM in the LLM Agent large model is used to comprehensively understand the relevant content of the optimized multi-source meteorological-electricity data parameterization scheme, deeply analyze the parameter relationships, data characteristics, etc. involved therein, automatically generate valuable analysis questions and directions, and during the understanding process, according to the AI-driven fully automatic analysis process of LLM Agent meteorological-energy data, use its knowledge reserve and reasoning ability to process data, mine potential associations, and construct a coupling index system based on the analysis results, which not only improves the operation efficiency and safety warning ability of the wind farm, but also provides reliable data support and decision-making basis for meteorological-electricity coupling research, realizing the maximization of the value mining of wind power data.

[0024] As Figure 2 shown, it is a fully automatic analysis flowchart of LLM Agent meteorological-energy data driven by AI provided by an embodiment of the present application. First, intention understanding and in-depth thinking are carried out based on the questions proposed by the LLM, and then a task plan is generated. During the execution of the task, real-time search and research of reference documents are carried out, and then data analysis and visualization operations are performed on the data processed by the ensemble empirical mode decomposition method, and finally conclusions are generated and reports are compiled. During the execution of the task, reflection and plan update are also carried out to ensure the effective progress of the task.

[0025] Figure 3 It is a mind map for quantitative determination of fan state provided by an embodiment of the present application. It includes: obtaining the fan state evaluation values of each fan by real-time monitoring of the fan state parameters of each fan, making a quantitative determination of the fan state according to the fan state evaluation values of each fan. If the fan state evaluation value of a certain fan is less than the first threshold of the fan state, the corresponding fan state quantitative determination result is recorded as the fan state being unqualified, and no additional processing is performed; if the fan state evaluation value of a certain fan is greater than or equal to the first threshold of the fan state and less than the second threshold of the fan state, the corresponding fan state quantitative determination result is recorded as the fan state to be optimized, and fan dynamic processing is performed; if the fan state evaluation value of a certain fan is greater than or equal to the second threshold of the fan state, the corresponding fan state quantitative determination result is recorded as the fan state being qualified, the abnormal fan is marked, and the preset personnel are notified to cut out the abnormal fan.

[0026] Specifically, the steps for quantitatively determining the fan status based on the fan status parameters of each fan in the wind farm include: obtaining the reference data of the fan status parameters from the preset meteorological power database, specifically including: critical wind energy utilization rate, critical wind speed variation coefficient, and critical active power volatility; performing a ratio approximation operation on the wind energy utilization rate, wind speed variation coefficient, and critical active power volatility of each fan respectively with the corresponding critical wind energy utilization rate, critical wind speed variation coefficient, and active power volatility. The ratio approximation operation is a ratio operation. Then, weight the results of the ratio approximation operation respectively using the fan status parameter weight ratios and couple them to obtain the fan status evaluation value of each fan. The fan status parameter weight ratios include the wind energy utilization rate weight ratio, the wind speed variation coefficient weight ratio, and the active power volatility weight ratio. The fan status evaluation value represents the quantitative data of the combined influence of the fan status parameters on the accuracy of the fan data. The fan status parameters include the wind energy utilization rate, the wind speed variation coefficient, and the active power volatility; obtaining the first fan status threshold and the second fan status threshold from the preset meteorological power database; comparing the fan status evaluation value of each fan with the first fan status threshold and the second fan status threshold respectively. If the fan status evaluation value of a certain fan is less than the first fan status threshold, record the corresponding fan status quantitative determination result as the fan status being unqualified; if the fan status evaluation value of a certain fan is greater than or equal to the first fan status threshold and less than the second fan status threshold, record the corresponding fan status quantitative determination result as the fan status to be optimized; if the fan status evaluation value of a certain fan is greater than or equal to the second fan status threshold, record the corresponding fan status quantitative determination result as the fan status being qualified.

[0027] The method for obtaining the fan status evaluation value of each fan is as follows: ; In the formula, represents the fan status evaluation value of the i-th fan, represents the wind energy utilization rate weight ratio, represents the wind speed variation coefficient weight ratio, represents the active power volatility weight ratio, represents the wind energy utilization rate of the i-th fan, which is the ratio of the kinetic energy of the air flow within the impeller range to the output electric energy during the current monitoring time period, represents the critical wind energy utilization rate, represents the wind speed variation coefficient of the i-th fan, which is a key indicator for measuring the intensity of wind speed fluctuations. Specifically, it is the ratio of the standard deviation of the wind speed to the average wind speed during the current monitoring time period. The wind speed can be directly measured by an anemometer, represents the critical wind speed variation coefficient, The active power fluctuation rate of the i-th wind turbine refers to the degree of fluctuation of the active power output of the wind turbine during the current monitoring period. The output active power of the wind turbine can be collected in real time using the power measurement equipment at the output end of the wind turbine. The active power fluctuation rate is the ratio of the difference between the maximum and minimum active power values during the current monitoring period to the average active power. represents the critical active power fluctuation rate, where i is the number of each wind turbine, i=1,2,3,...,N, and N is the total number of wind turbines.

[0028] 、 and These are the weight ratios corresponding to the wind energy utilization rate, wind speed variation coefficient, and active power fluctuation rate preset in the meteorological and power database, respectively, which represent the numerical values of the degree of influence of the wind energy utilization rate, wind speed variation coefficient, and active power fluctuation rate on the wind turbine status assessment value, and can be directly obtained from the meteorological and power database when used. These relationships are organized into a "mapping set", that is, a lookup table or corresponding rules. When the value of a wind turbine status parameter is actually monitored, this value can be searched according to the corresponding mapping set, and then the corresponding weight ratio can be obtained. This weight ratio is a number between 0 and 1, which represents the evaluation result of the degree of interference of the current wind turbine status parameter on the accuracy of power generation. For example, for wind energy utilization, there is a mapping set of wind energy utilization and the corresponding weight ratio of wind energy utilization. By inputting the value of wind energy utilization, we can get the corresponding weight ratio of wind energy utilization. For wind speed variation coefficient, there is a mapping set of wind speed variation coefficient and the corresponding weight ratio of wind speed variation coefficient. By inputting the value of wind speed variation coefficient, we can get the corresponding weight ratio of wind speed variation coefficient. For active power fluctuation rate, there is a mapping set of active power fluctuation rate and the corresponding weight ratio of active power fluctuation rate. By inputting the value of active power fluctuation rate, we can get the corresponding weight ratio of active power fluctuation rate. These mapping relationships can be many-to-one or one-to-one.

[0029] In this embodiment, there is a mutual correlation among the wind energy utilization rate, the wind speed variation coefficient, and the active power volatility. For example, the higher the wind speed variation coefficient, the greater the ratio of the wind speed standard deviation to the mean, indicating an increase in the disorder degree of the wind speed field. By enhancing vortex shedding, energy dissipation, and load pulsation, the turbulence intensity increases, resulting in a sharp fluctuation in the instantaneous input energy of the impeller and a significant increase in the active power volatility. The power volatility reflects the fluctuation of the output power of the wind turbine. The greater the power volatility, the more unstable the output power, and it is difficult for the wind turbine generator set to operate under the optimal conditions and continuously and efficiently convert wind energy into electrical energy, and the wind energy utilization rate may also be lower. When the wind speed variation coefficient increases, the instantaneous attack angle of the impeller frequently deviates from the optimal design range, the lift coefficient fluctuates more severely, and the wind energy capture efficiency decreases. The fan state evaluation value obtained through comprehensive analysis can accurately evaluate the working state of the fan, determine whether the fan is operating normally, whether adjustment or maintenance is required, effectively improve the productivity and stability of the wind farm, avoid power losses caused by unstable factors, and ensure the efficient operation of the wind farm system. By dynamically processing the fans based on the fan state evaluation values of each fan, the operating parameters of the fans can be optimized according to the changes in wind speed and fan state, ensuring that the fans are always in the best working state under different environmental conditions, thereby improving the working efficiency and power generation capacity of the fans.

[0030] Further, determining whether to perform dynamic processing on the fan based on the fan state quantization determination result, and obtaining each normal fan according to the fan dynamic processing result includes: if the fan state quantization determination result of a certain fan is that the fan state is qualified, no additional processing is performed; if the fan state quantization determination result of a certain fan is that the fan state needs to be optimized, the pitch angle and yaw angle are dynamically processed according to the fan state evaluation value and wind speed of the fan; if the fan state quantization determination result of a certain fan is that the fan state is unqualified, the fan is recorded as an abnormal fan and an abnormal fan prompt is issued; determining whether the number of fans with the fan state quantization determination result of unqualified fan state exceeds the preset fan number threshold. If so, the preset personnel are prompted to cut out the abnormal fans in batches according to the priority parameters of each abnormal fan, otherwise the preset personnel are directly prompted to cut out the abnormal fans; the fans not recorded as abnormal fans are recorded as normal fans.

[0031] Among them, the steps of dynamically processing the pitch angle and yaw angle according to the fan state evaluation value and wind speed of the fan include: when the wind speed of the fan is greater than the maximum value of the preset standard wind speed range in the meteorological power database and the fan state evaluation value in the current monitoring time period is greater than or equal to the fan state evaluation value in the previous monitoring time period, increase the pitch angle of the fan step by step according to the obtained pitch angle adjustment value until the fan state change value in the next monitoring time period exceeds the preset fan state evaluation value change threshold. The fan state change value is the difference between the fan state evaluation value in the next monitoring time period and the fan state evaluation value in the current monitoring time period. The pitch angle adjustment value is obtained by matching the current fan state deviation value with the pitch angle adjustment values corresponding to the preset fan state deviation ranges in the meteorological power database. The current fan state deviation value represents the difference between the fan state evaluation value in the current monitoring time period and the fan state evaluation value in the previous monitoring time period; when the wind speed of the fan is greater than the maximum value of the preset standard wind speed range and the fan state evaluation value in the current monitoring time period is less than the fan state evaluation value in the previous monitoring time period, decrease the pitch angle of the fan step by step according to the obtained pitch angle adjustment value until the fan state change value in the next monitoring time period exceeds the preset fan state evaluation value change threshold; when the wind speed of the fan is less than the minimum value of the preset standard wind speed range in the meteorological power database and the fan state evaluation value in the current monitoring time period is greater than or equal to the fan state evaluation value in the previous monitoring time period, increase the pitch angle of the fan step by step according to the obtained pitch angle adjustment value until the fan state change value in the next monitoring time period exceeds the preset fan state evaluation value change threshold; when the wind speed of the fan is less than the minimum value of the preset standard wind speed range and the fan state evaluation value in the current monitoring time period is less than the fan state evaluation value in the previous monitoring time period, decrease the pitch angle of the fan step by step according to the obtained pitch angle adjustment value until the fan state change value in the next monitoring time period exceeds the preset fan state evaluation value change threshold; when the wind speed of the fan is within the preset standard wind speed range in the meteorological power database, the yaw angle is dynamically adjusted according to the wind direction sensor data of the fan. The wind direction sensor data includes the real-time wind direction angle and the wind direction change rate. When dynamically adjusting the yaw angle, the difference between the wind direction angle and the current yaw angle of the fan is recorded as the initial target yaw angle in real time, and then the product of the wind direction change rate and the prediction time window obtained by a wind direction prediction model such as Kalman filter or neural network is added to the initial target yaw angle to obtain the target yaw angle, and the fan is adjusted to the target yaw angle in real time.

[0032] In this embodiment, when obtaining the pitch angle adjustment value, a mapping relationship table is formed by one-to-one correspondence between the state deviation ranges of each fan in the meteorological power database and the pitch angle adjustment value. The table records the state deviation range of each fan and its corresponding pitch angle adjustment value. These relationships can be one-to-one or many-to-one. When obtaining the pitch angle adjustment value, only need to input the fan state deviation value into the mapping relationship table, and the meteorological power database can quickly locate and return the pitch angle adjustment value corresponding to the fan state deviation value. The present invention realizes precise hierarchical control through the fan state quantization determination result, significantly improves the operation efficiency and safety of the wind farm, marks the unqualified fans in time and notifies the preset personnel to cut them out in batches to avoid the spread of faults and reduce the shutdown loss; for the fans to be optimized, the pitch angle and yaw angle are dynamically adjusted in combination with the wind speed and the fan state evaluation value to balance the power generation efficiency and mechanical load and extend the equipment life. Through the data-driven closed-loop control, not only the data quality of high-impact meteorological-electricity events modeling is ensured, but also the collaborative optimization of the power generation stability of the wind farm and the safety of the units is realized.

[0033] Among them, the steps of prompting the preset personnel to cut out the abnormal fans in batches according to the priority parameters of each abnormal fan include: obtaining the priority calculation value of each abnormal fan according to the priority parameters of each abnormal fan. The priority calculation value represents the quantitative data of the influence degree of the priority parameters on the fan operation priority. The priority parameters include the power generation contribution degree, average wind speed and power generation efficiency in the previous monitoring time period; sort the abnormal fans in ascending order according to the priority calculation value, and notify the preset personnel to cut out the number of abnormal fans equal to the fan quantity threshold according to the sorting order, and monitor the power change rate of the wind farm in real time. When the power change rate of the wind farm is lower than the preset power change threshold of the power station, continue to notify the preset personnel to cut out the number of abnormal fans equal to the fan quantity threshold until all the abnormal fans are shut down or repaired.

[0034] Specifically, the obtaining method of the priority calculation value of each abnormal fan is as follows: obtain the priority parameter reference data from the preset meteorological power database, specifically including: critical power generation contribution degree, critical average wind speed and critical power generation efficiency; perform the ratio approximation operation on the power generation contribution degree, critical average wind speed and power generation efficiency in the previous monitoring time period of each abnormal fan respectively with the corresponding critical power generation contribution degree, average wind speed and critical power generation efficiency, and then perform the weighting process on the ratio approximation operation results respectively by using the priority parameter weight ratio and couple them to obtain the priority calculation value of each abnormal fan. The priority parameter weight ratio includes the power generation contribution degree weight ratio, average wind speed weight ratio and power generation efficiency weight ratio.

[0035] Specifically, the calculation formula of the priority calculation value of each abnormal fan is: ; In the formula, represents the priority value of the jth abnormal wind turbine, Indicates the weight ratio of power generation contribution, represents the average wind speed weight ratio, represents the power generation efficiency weight ratio, It represents the contribution of the j-th abnormal wind turbine to the power generation, which is the proportion of the power generation of the abnormal wind turbine to the total power generation of the wind farm during the last monitoring period. represents the critical power generation contribution, The average wind speed of the jth abnormal wind turbine can be obtained by directly measuring the wind speed of each monitoring point in the previous monitoring period using an anemometer and calculating the average value. represents the critical mean wind speed, It represents the power generation efficiency of the jth abnormal wind turbine, which is the ratio of the actual power generation to the theoretical available wind energy in the last monitoring period. represents the critical power generation efficiency, where j is the number of each abnormal wind turbine, j = 1, 2, 3, ..., M, and M is the total number of abnormal wind turbines.

[0036] 、 and The weighted ratios for power generation contribution, average wind speed, and power generation efficiency, preset in the meteorological and power generation database, represent the degree of influence of power generation contribution, average wind speed, and power generation efficiency on the estimated priority value. These values can be directly retrieved from the meteorological and power generation database. These relationships are organized into a "mapping set," a lookup table or corresponding rules. When a priority parameter value is actually monitored, the corresponding value can be looked up in the corresponding mapping set to obtain the corresponding weighted ratio. This weighted ratio is a number between 0 and 1, representing the degree to which the current priority parameter interferes with the wind turbine's priority. For example, for power generation contribution, a mapping set exists between power generation contribution and its corresponding weighted ratio. Entering the power generation contribution value yields the corresponding weighted ratio. For average wind speed, a mapping set exists between average wind speed and its corresponding weighted ratio. Entering the average wind speed value yields the corresponding weighted ratio. For power generation efficiency, a mapping set exists between power generation efficiency and its corresponding weighted ratio. Entering the power generation efficiency value yields the corresponding weighted ratio. These mappings can be many-to-one or one-to-one.

[0037] In this embodiment, the power generation contribution degree, the average wind speed, and the power generation efficiency are interrelated. For example, under the same wind speed condition, the higher the power generation efficiency, the closer the actual power generation is to the theoretical maximum value, and the greater the power generation contribution degree. On the contrary, if the power generation efficiency of the wind turbine is low, then even if the wind speed is high, its power generation contribution degree may be limited; when the wind speed is lower than the cut-in wind speed, the wind energy captured by the wind turbine is small, and the power generation efficiency of the wind turbine is close to zero; when the wind speed is between the cut-in wind speed and the rated wind speed, the power generation efficiency increases with the increase of the wind speed; when the wind speed exceeds the rated wind speed, the wind turbine reaches the limit of the rated power, resulting in the power generation efficiency not increasing or even decreasing. The priority calculation value obtained through comprehensive analysis can determine the cut-out order for each abnormal wind turbine, and can cut out abnormal wind turbines to the greatest extent on the premise of the stable operation of the wind farm, reduce the impact of abnormal wind turbines on the accuracy of power generation, and maximize the overall power generation benefit of the wind farm.

[0038] As Figure 4 shown, it is a mind map for quantifying and judging wind speed fluctuations provided by an embodiment of the present application. It includes: obtaining the wind speed fluctuation evaluation value of each normal wind turbine in each monitoring time period by real-time monitoring of the wind speed fluctuation parameters of each normal wind turbine in each monitoring time period, and performing quantitative judgment of the wind speed fluctuation: comparing the wind speed fluctuation evaluation value of each normal wind turbine in each monitoring time period with the wind speed fluctuation threshold respectively. If the wind speed fluctuation evaluation value of a certain normal wind turbine in a certain monitoring time period is less than or equal to the wind speed fluctuation threshold, the corresponding wind speed fluctuation quantitative judgment result is recorded as qualified wind speed fluctuation and no additional processing is performed; if the wind speed fluctuation evaluation value of a certain normal wind turbine in a certain monitoring time period is greater than the wind speed fluctuation threshold, the corresponding wind speed fluctuation quantitative judgment result is recorded as unqualified wind speed fluctuation, and the power generation is corrected according to the wind speed fluctuation evaluation value of this normal wind turbine in this monitoring time period and the empirical mode decomposition method is optimized.

[0039] Specifically, the steps for quantitatively determining the wind speed fluctuation based on the wind speed fluctuation parameters of each normal wind turbine in each monitoring period of the wind farm are as follows: Obtain the reference data of wind speed fluctuation parameters from the preset meteorological power database, specifically including: critical wind speed variation coefficient, critical wind speed variation rate, and critical maximum wind speed fluctuation amplitude; perform a proportion approximation operation on the wind speed fluctuation parameters of each normal wind turbine in each monitoring period with the corresponding reference data of wind speed fluctuation parameters respectively, and then perform a weighting process on the proportion approximation operation using the wind speed fluctuation parameter weight ratios respectively and couple them to obtain the wind speed fluctuation evaluation values of each normal wind turbine in each monitoring period. The wind speed fluctuation parameter weight ratios include the wind speed variation coefficient weight ratio, the average wind speed variation rate weight ratio, and the maximum wind speed fluctuation amplitude weight ratio. The wind speed fluctuation evaluation value represents the quantitative data of the degree of influence of the wind speed fluctuation parameters on the power generation accuracy of the normal wind turbine. The wind speed fluctuation parameters include the wind speed variation coefficient, the average wind speed variation rate, and the maximum wind speed fluctuation amplitude; obtain the wind speed fluctuation threshold from the preset meteorological power database; compare the wind speed fluctuation evaluation values of each normal wind turbine in each monitoring period with the wind speed fluctuation threshold respectively. If the wind speed fluctuation evaluation value of a certain normal wind turbine in a certain monitoring period is less than or equal to the wind speed fluctuation threshold, record the corresponding wind speed fluctuation quantitative determination result as qualified for wind speed fluctuation; if the wind speed fluctuation evaluation value of a certain normal wind turbine in a certain monitoring period is greater than the wind speed fluctuation threshold, record the corresponding wind speed fluctuation quantitative determination result as unqualified for wind speed fluctuation.

[0040] The method for obtaining the wind speed fluctuation evaluation values of each normal wind turbine in each monitoring period is as follows: ; In the formula, represents the wind speed fluctuation evaluation value of the ath normal wind turbine in the kth monitoring period, represents the wind speed variation coefficient weight ratio, represents the average wind speed variation rate weight ratio, represents the maximum wind speed fluctuation amplitude weight ratio, represents the wind speed variation coefficient of the ath normal wind turbine in the kth monitoring period. The wind speed can be measured in real time using an anemometer. The ratio of the standard deviation of the wind speed to the average wind speed within this monitoring period is the wind speed variation coefficient, represents the critical wind speed variation coefficient, represents the average wind speed variation rate of the ath normal wind turbine in the kth monitoring period, which is obtained by comparing the absolute value of the wind speed difference between adjacent time monitoring points within this monitoring period with the time interval and then performing a mean operation on the calculation results, represents the critical average wind speed variation rate, represents the maximum wind speed fluctuation amplitude of the ath normal wind turbine in the kth monitoring period, which is the difference between the maximum wind speed and the minimum wind speed within this monitoring period, represents the critical maximum wind speed fluctuation amplitude, where k is the number of each monitoring time period, k = 1, 2, 3, ..., L, L is the total number of monitoring time periods, a is the number of each normal wind turbine, a = 1, 2, 3, ..., A, A is the total number of normal wind turbines.

[0041] 、 and These are the weight ratios corresponding to the wind speed variation coefficient, average wind speed variability, and maximum wind speed fluctuation amplitude preset in the meteorological and power database, respectively, representing the numerical values of the degree of influence of the wind speed variation coefficient, average wind speed variability, and maximum wind speed fluctuation amplitude on the wind speed fluctuation assessment value, which can be directly obtained from the meteorological and power database when used. These relationships are organized into a "mapping set," that is, a lookup table or corresponding rules. When the value of a certain wind speed fluctuation parameter is actually monitored, this value can be looked up in the corresponding mapping set, and then the corresponding weight ratio can be obtained. This weight ratio is a number between 0 and 1, which represents the evaluation result of the current wind speed fluctuation parameter on the power generation and the accuracy of the ensemble empirical mode decomposition method. For example, for the wind speed variation coefficient, there is a mapping set of the wind speed variation coefficient and the corresponding weight ratio of the wind speed variation coefficient. By inputting the value of the wind speed variation coefficient, the corresponding weight ratio of the wind speed variation coefficient can be obtained. For the average wind speed variability, there is a mapping set of the average wind speed variability and the corresponding weight ratio of the average wind speed variability. By inputting the value of the average wind speed variability, the corresponding weight ratio of the average wind speed variability can be obtained. For the maximum wind speed fluctuation amplitude, there is a mapping set of the maximum wind speed fluctuation amplitude and the corresponding weight ratio of the maximum wind speed fluctuation amplitude. By inputting the value of the maximum wind speed fluctuation amplitude, the corresponding weight ratio of the maximum wind speed fluctuation amplitude can be obtained. These mapping relationships can be many-to-one or one-to-one.

[0042] In this embodiment, the wind speed coefficient of variation, the average wind speed variation rate, and the maximum wind speed fluctuation amplitude all reflect the intensity and amplitude of wind speed fluctuations. The wind speed coefficient of variation and the wind speed variation rate can measure the instantaneous fluctuation degree of the wind speed, while the maximum wind speed fluctuation amplitude reflects the maximum change range of the wind speed. The three are interrelated. For example, the higher the wind speed coefficient of variation, the larger the wind speed standard deviation, which means that the wind speed is more likely to change significantly per unit time, and the average wind speed variation rate may be higher. At the same time, a high wind speed coefficient of variation indicates a dispersed wind speed distribution, such as the higher probability of extreme conditions such as gusts and calm winds, which may be accompanied by a larger wind speed fluctuation amplitude. If the maximum wind speed fluctuation amplitude is larger within a short period of time, the wind speed variation rate per unit time will necessarily increase. If the wind speed repeatedly crosses a large range over a long period of time, it means that the wind speed has a wide fluctuation range, and the average wind speed variation rate will also be larger. The wind speed fluctuation assessment value obtained through comprehensive analysis reflects the stability and efficiency of the fan under different wind speed conditions, and can help identify the impact of wind speed fluctuations on the accuracy of the fan's power generation. By dynamically correcting the power generation according to the wind speed fluctuation assessment value and optimizing the ensemble empirical mode decomposition method, the accuracy of power generation can be improved, and at the same time, the accuracy of the ensemble empirical mode decomposition method for modeling data analysis of high-impact meteorological-electricity events can be enhanced.

[0043] Furthermore, the steps for determining whether to perform dynamic power generation correction and optimize the ensemble empirical mode decomposition method based on the wind speed fluctuation quantization determination result are as follows: If the wind speed fluctuation quantization determination result of a normal fan in a certain monitoring time period is qualified for wind speed fluctuations, no additional processing is performed; if the wind speed fluctuation quantization determination result of a normal fan in a certain monitoring time period is unqualified for wind speed fluctuations, a fluctuation anomaly prompt is issued, and the difference between the wind speed fluctuation assessment value of the normal fan in this monitoring time period and the wind speed fluctuation threshold is recorded as the wind speed fluctuation deviation value; the white noise data is dynamically corrected according to the wind speed fluctuation deviation value, and the white noise data includes the white noise standard deviation of high-frequency data and the white noise standard deviation of low-frequency data; the wind speed fluctuation deviation value is matched with the power generation correction values corresponding to each wind speed fluctuation deviation value interval preset in the meteorological power database to obtain the power generation correction value, and the sum of the power generation correction value and the power generation of the fan is the corrected power generation.

[0044] Among them, the steps of dynamically correcting white noise data according to the wind speed fluctuation deviation value include: matching the wind speed fluctuation deviation value with the proportional factor data corresponding to each preset wind speed fluctuation deviation value interval in the meteorological power database to obtain the corresponding proportional factor data, where the proportional factor data includes a high-frequency proportional factor and a low-frequency proportional factor; obtaining the initial white noise standard deviation and the white noise standard deviation multiple range from the preset meteorological power database, where the white noise standard deviation multiple range includes a high-frequency multiple range and a low-frequency multiple range; obtaining the white noise standard deviation of the corrected high-frequency data according to the initial white noise standard deviation, the high-frequency multiple range, the high-frequency proportional factor, and the wind speed fluctuation deviation value; and obtaining the white noise standard deviation of the corrected low-frequency data according to the initial white noise standard deviation, the low-frequency multiple range, the low-frequency proportional factor, and the wind speed fluctuation deviation value. In this embodiment, the methods for obtaining the white noise standard deviation of the corrected high-frequency data and the white noise standard deviation of the corrected low-frequency data are as follows: respectively obtaining the high-frequency multiple adjustment value and the low-frequency multiple adjustment value by multiplying the high-frequency proportional factor and the low-frequency proportional factor by the wind speed fluctuation deviation value, and then respectively multiplying the high-frequency multiple adjustment value and the low-frequency multiple adjustment value by the sum of the corresponding minimum values of the high-frequency multiple range and the low-frequency multiple range and the initial white noise standard deviation to obtain the white noise standard deviation of the corrected high-frequency data and the white noise standard deviation of the corrected low-frequency data. The present invention dynamically matches the power generation correction value based on the wind speed fluctuation deviation value, effectively reducing the interference of abnormal wind speed on the power generation data and ensuring the accuracy of the output power. It not only reduces the negative impact of abnormal wind speed fluctuations on the modeling of high-impact meteorological-power events by dynamically correcting the power generation and optimizing the ensemble empirical mode decomposition parameters, but also provides a high-quality and highly consistent data basis for subsequent AI-driven LLM Agent analysis.

[0045] Such as Figure 5As shown in the figure, it is a schematic structural diagram of an analysis system based on high-impact meteorological-electricity event modeling data provided by an embodiment of the present application. The analysis system based on high-impact meteorological-electricity event modeling data provided by an embodiment of the present application includes: a fan processing module, a power generation correction module, a data analysis module, and a meteorological-electricity database; among them, the fan processing module is used to perform a quantitative determination of the fan state according to the fan state parameters of each fan in the wind farm, obtain the fan state quantitative determination result, determine whether to perform fan dynamic processing based on the fan state quantitative determination result, and obtain each normal fan according to the fan dynamic processing result; the power generation correction module is used to perform a quantitative determination of the wind speed fluctuation according to the wind speed fluctuation parameters of each normal fan in each monitoring time period of the wind farm, obtain the wind speed fluctuation quantitative determination result, and determine whether to perform power generation dynamic correction and ensemble empirical mode decomposition method optimization based on the wind speed fluctuation quantitative determination result; the data analysis module is used to perform meteorological-electricity data analysis using the optimized ensemble empirical mode decomposition method. The meteorological-electricity data analysis means using an AI-driven large LLM Agent model to perform parametric modeling and analysis on the wind power generation data processed by the optimized ensemble empirical mode decomposition method, and constructing a multi-dimensional coupling index system.

[0046] In summary, the embodiment of the present application first performs a quantitative determination based on the operating state parameters of each fan in the wind farm, generates the fan state level, and implements dynamic processing on the fan according to the determination result, screening out normal fans to eliminate the interference of low-quality data on modeling; then quantitatively evaluates the wind speed fluctuation parameters of normal fans in each monitoring time period, and if the fluctuation exceeds the threshold, triggers power generation dynamic correction and ensemble empirical mode decomposition method optimization, thereby reducing the impact of abnormal wind speed on data stability; finally, uses the optimized ensemble empirical mode decomposition method to decompose the meteorological-electricity coupling data, performs parametric feature extraction and multi-dimensional analysis on the IMF components through an AI-driven LLM Agent, and constructs a coupling index system covering meteorological sensitivity, equipment status, and grid response ability, realizing the improvement of the accuracy of high-impact meteorological-electricity event modeling data analysis.

[0047] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented 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.

[0048] The present invention is described with reference to flowchart illustrations and / or block diagram illustrations of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowchart illustrations and / or block diagram illustrations, and combinations of flows and / or blocks in the flowchart illustrations and / or block diagram illustrations, can be implemented by computer program instructions. These computer program instructions may 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 executed by the processor of the computer or other programmable data processing apparatus create means for implementing the functions specified in the flow Figure 1 one flow or more flows and / or blocks Figure 1 one block or more blocks.

[0049] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the functions specified in the flow Figure 1 one flow or more flows and / or blocks Figure 1 one block or more blocks.

[0050] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in the flow Figure 1 one flow or more flows and / or blocks Figure 1 one block or more blocks.

[0051] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0052] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. Analysis method based on high-impact meteorological-power event modeling data, characterized in that, It includes the following steps: Quantitatively determine the fan status based on the fan status parameters of each fan in the wind farm to obtain the fan status quantitative determination result. Determine whether to perform fan dynamic processing based on the fan status quantitative determination result, and obtain each normal fan according to the fan dynamic processing result. The fan dynamic processing means processing the fan according to the fan status quantitative determination result to reduce the impact of the fan efficiency on the accuracy of the high-impact meteorological-electricity event modeling data analysis; Quantitatively determine the wind speed fluctuation based on the wind speed fluctuation parameters of each normal fan in each monitoring time period of the wind farm to obtain the wind speed fluctuation quantitative determination result. Determine whether to perform power generation dynamic correction and ensemble empirical mode decomposition method optimization based on the wind speed fluctuation quantitative determination result. The power generation dynamic correction and ensemble empirical mode decomposition method optimization mean respectively correcting the power generation of the normal fan and optimizing the white noise standard deviation in the ensemble empirical mode decomposition method according to the wind speed fluctuation quantitative determination result to reduce the impact of abnormal wind speed fluctuations on the accuracy of the high-impact meteorological-electricity event modeling data analysis; Perform meteorological-electricity data analysis using the optimized ensemble empirical mode decomposition method. The meteorological-electricity data analysis means using the AI-driven LLM Agent large model to perform parametric modeling and analysis on the wind power generation data processed by the optimized ensemble empirical mode decomposition method, and constructing a multi-dimensional coupling index system.

2. The analysis method based on high-impact meteorological-power event modeling data according to claim 1, wherein: The step of quantitatively determining the fan status according to the fan status parameters of each fan in the wind farm includes: Obtain the fan status parameter reference data from the preset meteorological power database, specifically including: critical wind energy utilization rate, critical wind speed variation coefficient, and critical active power volatility; Perform a ratio approach operation on the wind energy utilization rate, wind speed variation coefficient, and critical active power volatility of each fan with the corresponding critical wind energy utilization rate, critical wind speed variation coefficient, and active power volatility respectively, and then perform a weighting process on the ratio approach operation results using the fan status parameter weight ratio and couple them to obtain the fan status evaluation value of each fan. The fan status parameter weight ratio includes the wind energy utilization rate weight ratio, the wind speed variation coefficient weight ratio, and the active power volatility weight ratio. The fan status evaluation value represents the quantitative data of the combined influence of the fan status parameters on the accuracy of the fan data. The fan status parameters include the wind energy utilization rate, the wind speed variation coefficient, and the active power volatility; Obtain the first fan status threshold and the second fan status threshold from the preset meteorological power database; Compare the fan status evaluation value of each fan with the first fan status threshold and the second fan status threshold respectively. If the fan status evaluation value of a certain fan is less than the first fan status threshold, record the corresponding fan status quantitative determination result as fan status unqualified; If the fan status evaluation value of a certain fan is greater than or equal to the first fan status threshold and less than the second fan status threshold, record the corresponding fan status quantitative determination result as fan status to be optimized; If the fan status evaluation value of a certain fan is greater than or equal to the second fan status threshold, then record the corresponding fan status quantization determination result as qualified fan status.

3. The analysis method based on high-impact meteorological-power event modeling data according to claim 2, wherein: The steps of determining whether to perform fan dynamic processing based on the fan status quantization determination result and obtaining each normal fan according to the fan dynamic processing result include: If the fan status quantization determination result of a certain fan is qualified fan status, then no additional processing is performed; If the fan status quantization determination result of a certain fan is fan status to be optimized, then perform dynamic processing on the pitch angle and yaw angle according to the fan status evaluation value and wind speed of this fan; If the fan status quantization determination result of a certain fan is unqualified fan status, then record this fan as an abnormal fan and issue a fan abnormality prompt; Judge whether the number of fans with the fan status quantization determination result of unqualified fan status exceeds the preset fan number threshold. If so, then prompt the preset personnel to cut out the abnormal fans in batches according to the priority parameters of each abnormal fan, otherwise directly prompt the preset personnel to cut out the abnormal fans; Record the fans that are not recorded as abnormal fans as normal fans.

4. The analysis method based on high-impact meteorological-power event modeling data according to claim 3, characterized in that: The steps of performing dynamic processing on the pitch angle and yaw angle according to the fan status evaluation value and wind speed of this fan include: When the wind speed of this fan is greater than the maximum value of the preset standard wind speed range and the fan status evaluation value in the current monitoring time period is greater than or equal to the fan status evaluation value in the previous monitoring time period, increase the pitch angle of this fan step by step according to the obtained pitch angle adjustment value until the fan status change value in the next monitoring time period exceeds the preset fan status evaluation value change threshold. The pitch angle adjustment value is obtained by matching the current fan status deviation value with the pitch angle adjustment values corresponding to the preset fan status deviation ranges in the meteorological power database. The current fan status deviation value represents the difference between the fan status evaluation value in the current monitoring time period and the fan status evaluation value in the previous monitoring time period; When the wind speed of this fan is greater than the maximum value of the preset standard wind speed range and the fan status evaluation value in the current monitoring time period is less than the fan status evaluation value in the previous monitoring time period, decrease the pitch angle of this fan step by step according to the obtained pitch angle adjustment value until the fan status change value in the next monitoring time period exceeds the preset fan status evaluation value change threshold; When the wind speed of this fan is less than the minimum value of the preset standard wind speed range and the fan status evaluation value in the current monitoring time period is greater than or equal to the fan status evaluation value in the previous monitoring time period, increase the pitch angle of this fan step by step according to the obtained pitch angle adjustment value until the fan status change value in the next monitoring time period exceeds the preset fan status evaluation value change threshold; When the wind speed of this fan is less than the minimum value of the preset standard wind speed range and the fan status evaluation value in the current monitoring time period is less than the fan status evaluation value in the previous monitoring time period, decrease the pitch angle of this fan step by step according to the obtained pitch angle adjustment value until the fan status change value in the next monitoring time period exceeds the preset fan status evaluation value change threshold; When the wind speed of this fan is within the preset standard wind speed range, then dynamically adjust the yaw angle according to the wind direction sensor data of this fan.

5. The analysis method based on high-impact meteorological-power event modeling data according to claim 3, characterized in that: The steps of prompting the preset personnel to cut out the abnormal wind turbines in batches according to the priority parameters of each abnormal wind turbine include: Obtain the priority calculation values of each abnormal wind turbine according to the priority parameters of each abnormal wind turbine. The priority calculation value represents the quantification data of the influence degree of the priority parameters on the operation priority of the wind turbine. The priority parameters include the power generation contribution degree, average wind speed, and power generation efficiency in the previous monitoring time period; Sort each abnormal wind turbine in ascending order according to the priority calculation value, and notify the preset personnel to cut out the abnormal wind turbine according to the sorting order.

6. The analysis method based on high-impact meteorological-power event modeling data according to claim 5, characterized in that: The acquisition method of the priority calculation value of each abnormal wind turbine is as follows: Obtain the priority parameter reference data from the preset meteorological power database, specifically including: critical power generation contribution degree, critical average wind speed, and critical power generation efficiency; Perform the ratio approximation operation on the power generation contribution degree, critical average wind speed, and power generation efficiency in the previous monitoring time period of each abnormal wind turbine with the corresponding critical power generation contribution degree, average wind speed, and critical power generation efficiency respectively. Then, perform the weighting process on the ratio approximation operation results respectively using the priority parameter weight ratio and couple them to obtain the priority calculation value of each abnormal wind turbine. The priority parameter weight ratio includes the power generation contribution degree weight ratio, average wind speed weight ratio, and power generation efficiency weight ratio.

7. The analysis method based on high-impact meteorological-electric power event modeling data according to claim 1, wherein: The steps of quantitatively determining the wind speed fluctuation according to the wind speed fluctuation parameters of each normal wind turbine in each monitoring time period in the wind farm include: Obtain the wind speed fluctuation parameter reference data from the preset meteorological power database, specifically including: critical wind speed variation coefficient, critical wind speed change rate, and critical maximum wind speed fluctuation amplitude; Perform the ratio approximation operation on the wind speed fluctuation parameters of each normal wind turbine in each monitoring time period with the corresponding wind speed fluctuation parameter reference data respectively. Then, perform the weighting process on the ratio approximation operation respectively using the wind speed fluctuation parameter weight ratio and couple them to obtain the wind speed fluctuation evaluation value of each normal wind turbine in each monitoring time period. The wind speed fluctuation parameter weight ratio includes the wind speed variation coefficient weight ratio, average wind speed change rate weight ratio, and maximum wind speed fluctuation amplitude weight ratio. The wind speed fluctuation evaluation value represents the quantification data of the influence degree of the wind speed fluctuation parameters on the power generation accuracy of the normal wind turbine. The wind speed fluctuation parameters include the wind speed variation coefficient, average wind speed change rate, and maximum wind speed fluctuation amplitude; Obtain the wind speed fluctuation threshold from the preset meteorological power database; Compare the wind speed fluctuation evaluation value of each normal wind turbine in each monitoring time period with the wind speed fluctuation threshold respectively. If the wind speed fluctuation evaluation value of a certain normal wind turbine in a certain monitoring time period is less than or equal to the wind speed fluctuation threshold, record the corresponding wind speed fluctuation quantitative determination result as qualified wind speed fluctuation; If the wind speed fluctuation evaluation value of a certain normal wind turbine in a certain monitoring time period is greater than the wind speed fluctuation threshold, record the corresponding wind speed fluctuation quantitative determination result as unqualified wind speed fluctuation.

8. The analysis method based on high-impact meteorological-power event modeling data according to claim 7, wherein: The steps of determining whether to perform dynamic power generation correction and optimized ensemble empirical mode decomposition method based on the wind speed fluctuation quantitative determination result include: If the wind speed fluctuation quantitative determination result of a certain normal wind turbine in a certain monitoring time period is qualified wind speed fluctuation, no additional processing is performed; If the quantization determination result of the wind speed fluctuation of a normal wind turbine in a certain monitoring period is that the wind speed fluctuation is unqualified, a fluctuation anomaly prompt is issued, and the difference between the wind speed fluctuation evaluation value and the wind speed fluctuation threshold of the normal wind turbine in this monitoring period is recorded as the wind speed fluctuation deviation value; Dynamically correct the white noise data according to the wind speed fluctuation deviation value, and the white noise data includes the white noise standard deviation of high-frequency data and the white noise standard deviation of low-frequency data; Match the wind speed fluctuation deviation value with the power generation correction value corresponding to each preset wind speed fluctuation deviation value interval in the meteorological power database to obtain the power generation correction value, and correct the power generation of the wind turbine according to the power generation correction value.

9. The analysis method based on high-impact meteorological-power event modeling data according to claim 8, wherein: The step of dynamically correcting the white noise data according to the wind speed fluctuation deviation value includes: Match the wind speed fluctuation deviation value with the proportional factor data corresponding to each preset wind speed fluctuation deviation value interval in the meteorological power database to obtain the corresponding proportional factor data, and the proportional factor data includes a high-frequency proportional factor and a low-frequency proportional factor; Obtain the initial white noise standard deviation and the white noise standard deviation multiple range from the preset meteorological power database, and the white noise standard deviation multiple range includes a high-frequency multiple range and a low-frequency multiple range; Obtain the white noise standard deviation of the corrected high-frequency data according to the initial white noise standard deviation, the high-frequency multiple range, the high-frequency proportional factor, and the wind speed fluctuation deviation value; Obtain the white noise standard deviation of the corrected low-frequency data according to the initial white noise standard deviation, the low-frequency multiple range, the low-frequency proportional factor, and the wind speed fluctuation deviation value.

10. A system applying the analysis method based on high-impact meteorological-electric power event modeling data as described in any one of claims 1-9, characterized in that: It includes a wind turbine processing module, a power generation correction module, a data analysis module, and a meteorological power database; Among them, the wind turbine processing module is used to perform quantization determination of the wind turbine state according to the wind turbine state parameters of each wind turbine in the wind farm, obtain the wind turbine state quantization determination result, determine whether to perform wind turbine dynamic processing based on the wind turbine state quantization determination result, and obtain each normal wind turbine according to the wind turbine dynamic processing result; The power generation correction module is used to perform quantization determination of the wind speed fluctuation according to the wind speed fluctuation parameters of each normal wind turbine in each monitoring period of the wind farm, obtain the wind speed fluctuation quantization determination result, and determine whether to perform dynamic power generation correction and ensemble empirical mode decomposition method optimization based on the wind speed fluctuation quantization determination result; The data analysis module is used to perform meteorological-power data analysis by using the optimized ensemble empirical mode decomposition method. The meteorological-power data analysis means using an AI-driven large model of the LLM Agent to perform parametric modeling and analysis on the wind power generation data processed by the optimized ensemble empirical mode decomposition method, and constructing a multi-dimensional coupling index system.

Images