A method and system for power dispatching based on wind power prediction

Abstract

The present invention relates to the field of wind power management technology, specifically to a power dispatching method and system based on wind power prediction, comprising the following steps: collecting the operating wind speed data and power output data of the wind farm, calculating the extreme value difference and average value difference of the time point in the time series according to the wind speed and corresponding power information at each time point, analyzing the current data change trend, and generating power fluctuation characteristic analysis results. In the present invention, by analyzing the extreme value difference and average value difference of the time series data, an in-depth analysis of the power fluctuation characteristics is achieved, which can help optimize the response strategy of the power system, ensure the optimization of the load distribution of the power grid and reduce the risks caused by power fluctuations. By calculating the local data density and fractal dimension, the complexity of power fluctuations is quantified, which not only enhances the adaptability to the actual operating conditions of the power grid, but also provides accurate data support for wind power fluctuation prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power management, and in particular to a power dispatching method and system based on wind power prediction. Background Art

[0002] Wind power management is a technical field that involves optimizing the output and operation of wind power generation systems. The field manages the energy production of wind power plants, including the measurement, prediction and adjustment of power output, real-time monitoring of wind speed, wind direction and other environmental factors, and optimization of wind turbine performance to maximize energy collection and reduce operation and maintenance costs. Ensure the stable operation of the power system, adjust the power supply to match the grid demand, and prevent equipment failure.

[0003] Among them, the power dispatching method is to guide the operation and dispatching of the power system by estimating the power output of wind farms in the future. In the energy structure dominated by wind power generation, it helps the power dispatching center to more accurately arrange the balance between power generation and consumption. In addition, the method can optimize the load distribution of the power grid and reduce the risks caused by power fluctuations, thereby improving the economy and reliability of the entire power system.

[0004] Existing technologies mainly rely on estimating future wind power output to guide the operation and dispatch of power systems. It is difficult to fully analyze the subtle changes and complex relationship between wind speed and power, which often leads to the power dispatch center's inability to react in time to sudden wind speed changes and the inability to accurately balance the relationship between power generation and consumption, increasing the risk of system operation. In addition, existing methods are difficult to effectively reduce the grid stability problems caused by power fluctuations in optimizing grid load distribution, thereby affecting the economy and reliability of the entire power system, making the power dispatch strategy not conducive to maximizing the use of wind power resources, and easily causing energy waste and reduced economic benefits. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose a power dispatching method and system based on wind power prediction.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a power dispatching method based on wind power prediction, comprising the following steps:

[0007] S1: Collect the operating wind speed data and power output data of the wind farm, calculate the extreme value difference and average value difference of the time point in the time series according to the wind speed and corresponding power information at each time point, analyze the current data change trend, and generate the power fluctuation characteristic analysis results;

[0008] S2: According to the power fluctuation characteristic analysis results, calculate the local data density at each time point, analyze the distribution characteristics of the local data, calculate the fractal dimension to quantify the complexity of power fluctuation, and integrate the data to establish a wind power fluctuation prediction model;

[0009] S3: Based on the wind power fluctuation prediction model, combined with the current wind speed and environmental parameters, simulate the power output scenario, calculate the future power fluctuation amplitude under various wind speed conditions, sample the simulated data, integrate all simulation results, and generate a prediction scenario sample set;

[0010] S4: Based on the predicted scenario sample set, analyze the energy storage demand and grid load in the simulated scenario, evaluate the operating status of the grid under various power output conditions, make power dispatch adjustments according to demand changes in the simulated scenario, and generate power dispatch optimization results.

[0011] As a further solution of the present invention, the acquisition step of analyzing the current data change trend is specifically:

[0012] S111: collecting wind farm power output data in a real-time monitoring manner, identifying the maximum and minimum values ​​of the power data within a specified time interval, and obtaining a maximum fluctuation range analysis result by analyzing the maximum power fluctuation of the wind farm within the specified time interval;

[0013] S112: Based on the maximum fluctuation range analysis result, according to the degree of deviation between the data point and its average value, analyzing the average fluctuation of the entire data set to obtain a power average fluctuation analysis result;

[0014] S113: According to the power average fluctuation analysis result, by comparing the benchmark fluctuation value of the operating data, the fluctuation behavior and stability of the wind farm in the current monitoring period are evaluated, the current data change trend is analyzed, and the power fluctuation characteristic analysis result is generated.

[0015] As a further solution of the present invention, the step of acquiring the distribution characteristics of the local data is specifically as follows:

[0016] S211: Based on the power fluctuation characteristic analysis result, the frequency of power data output by the wind farm at each time point is analyzed, using the formula:

[0017]

[0018] Calculation time The weighted local data density of , generate data density information;

[0019] in, is the total number of data points, Indicates when The value is 1 when the is a specified time point, It is Time labels for each data point;

[0020] for , using the formula:

[0021]

[0022] in, and It's in time The extreme value difference and the mean value difference;

[0023] S212: According to the data density information, compare the average data density of the operating data, analyze whether there is abnormal power fluctuation at each time point, and obtain distribution feature analysis results.

[0024] As a further solution of the present invention, the step of obtaining the complexity of the quantified power fluctuation is specifically:

[0025] S221: According to the distribution characteristic analysis result, through the fractal dimension analysis process, the formula is used:

[0026]

[0027] Calculating fractal dimension , identify repeated patterns and structural changes in data sets, and evaluate the complexity and self-similarity of data to generate data complexity analysis results;

[0028] in, is the size of the box covering the dataset, is in scale Next, considering the time point The local data density The minimum number of boxes required after adjustment is, It's at the time Power output data, is the coverage factor, and is the adjustment factor;

[0029] S222: Based on the data complexity analysis results, combined with power operation data and current environmental variables, a wind power fluctuation prediction model is established through data integration and model training.

[0030] As a further solution of the present invention, the step of obtaining the future power fluctuation amplitude is specifically as follows:

[0031] S311: The collected current wind speed and environmental parameters are uploaded to the wind power fluctuation prediction model through sensors and data interfaces to perform power output prediction and obtain time series prediction data;

[0032] S312: Based on the time series prediction data, combined with the operating power data and current environmental variables, the impact of wind speed and environmental factors on power fluctuations is estimated to obtain future power fluctuation amplitude analysis results.

[0033] As a further solution of the present invention, the step of sampling the simulated data is specifically as follows:

[0034] S321: According to the analysis result of the future power fluctuation amplitude, combined with the real-time wind speed and environmental parameters, a random sample of power fluctuation is collected, using the formula:

[0035] Calculate the total predicted power fluctuation , create a prediction dataset;

[0036] in, To identify the sampling points, is the total number of samples, is the scaling factor, It is A random sample, is the offset coefficient, is the global adjustment coefficient;

[0037] S322: According to the prediction data set, collect all the individual simulation results, including the power prediction under each wind speed and environmental condition, classify and store the results, and generate a prediction scenario sample set.

[0038] As a further solution of the present invention, the step of obtaining the power dispatch adjustment is specifically:

[0039] S411: Based on the predicted scenario sample set, identify the peak load and minimum load period in the scenario, compare the energy storage utilization rate and the load response mode, identify potential power demand pressure points, classify the scenario according to demand intensity, and obtain analyzed scenario data;

[0040] S412: Based on the analyzed scenario data, the operation status of the power grid under multiple power output conditions is evaluated, and the response capability and stability of the power grid are simulated by adjusting the power grid capacity and the influence of environmental factors, using the formula:

[0041]

[0042] Calculating the index of grid stability ;

[0043] in, is the grid capacity adjustment factor, is the current load of the grid, is the environmental factor adjustment coefficient, It is an indicator of environmental quality;

[0044] S413: Analyze the power demand and energy storage utilization in multiple scenarios according to the grid stability index, adjust the power generation plan and energy storage scheduling, optimize the response of the grid, and generate power scheduling optimization results.

[0045] A power dispatching system based on wind power prediction, the power dispatching system based on wind power prediction is used to execute the power dispatching method based on wind power prediction, the system comprises:

[0046] The wind speed and power analysis module collects the operating wind speed data and power output data of the wind farm, calculates the extreme value difference and average value difference of the time points in the time series, analyzes the current data change trend, and generates power fluctuation characteristic analysis results;

[0047] The local density calculation module analyzes the local data density at each time point based on the power fluctuation characteristic analysis result, calculates the distribution characteristics of the local data, and calculates the fractal dimension to quantify the complexity of the power fluctuation to obtain local characteristic information;

[0048] The fluctuation prediction model building module uses the local characteristic information, combines the current wind speed and environmental parameters, builds a wind power fluctuation prediction model, samples the simulated data, integrates all simulation results, and generates a prediction scenario sample set;

[0049] The power demand and load analysis module analyzes the energy storage demand and grid load in the simulation scenario based on the prediction scenario sample set, evaluates the operation status of the grid under various power output conditions, and obtains the grid operation status evaluation result;

[0050] The dispatch strategy optimization module performs power dispatch adjustment according to the power grid operation status evaluation result and generates a power dispatch optimization result.

[0051] Compared with the prior art, the advantages and positive effects of the present invention are:

[0052] In the present invention, by analyzing the extreme value difference and average value difference of time series data, an in-depth analysis of power fluctuation characteristics is achieved, which can help optimize the response strategy of the power system, ensure the optimization of power grid load distribution and reduce the risk caused by power fluctuations. By calculating the local data density and fractal dimension, the complexity of power fluctuations is quantified, which not only enhances the adaptability to the actual operating conditions of the power grid, but also provides accurate data support for wind power fluctuation prediction. By predicting future power output scenarios under variable environmental conditions and integrating simulation results to form a prediction scenario sample set, the flexibility and accuracy of power dispatching are enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0054] Figure 2 Flow chart of the acquisition steps for analyzing the current data change trend of the present invention

[0055] Figure 3 A flow chart of the steps of obtaining the distribution characteristics of local data analyzed in the present invention;

[0056] Figure 4 A flow chart of the steps for obtaining the complexity of quantifying power fluctuations for the present invention;

[0057] Figure 5 A flow chart of the steps for obtaining the future power fluctuation amplitude of the present invention;

[0058] Figure 6 A flow chart of the acquisition steps of sampling simulated data according to the present invention;

[0059] Figure 7 The present invention is a flowchart of the steps for obtaining power dispatch adjustment. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0061] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0062] See also Figure 1 The present invention provides a technical solution: a power dispatching method based on wind power prediction, comprising the following steps:

[0063] S1: Collect the operating wind speed data and power output data of the wind farm, calculate the extreme value difference and average value difference of the time point in the time series according to the wind speed and corresponding power information at each time point, analyze the current data change trend, and generate the power fluctuation characteristic analysis results;

[0064] S2: Based on the power fluctuation characteristics analysis results, calculate the local data density at each time point, analyze the distribution characteristics of the local data, and calculate the fractal dimension to quantify the complexity of power fluctuations, and integrate the data to establish a wind power fluctuation prediction model;

[0065] S3: Based on the wind power fluctuation prediction model, combined with the current wind speed and environmental parameters, simulate the power output scenario, calculate the future power fluctuation amplitude under various wind speed conditions, sample the simulated data, integrate all simulation results, and generate a prediction scenario sample set;

[0066] S4: Based on the predicted scenario sample set, analyze the energy storage demand and grid load in the simulated scenario, evaluate the operation status of the grid under various power output conditions, adjust the power dispatch according to the demand changes in the simulated scenario, and generate power dispatch optimization results;

[0067] The results of the power fluctuation characteristics analysis include the peak change, minimum value difference, and average fluctuation range of the time series. The wind power fluctuation prediction model includes the local maximum likelihood estimation parameters, the predicted error range, and the probability distribution type of the fluctuation pattern. The prediction scenario sample set is specifically a scenario set based on the highest power output, lowest power output, and average power output under different simulation conditions. The power dispatch optimization results include optimized energy storage configuration, adjusted response schedule, and improved load management plan.

[0068] See also Figure 2 , the specific steps for obtaining the current data change trend analysis are:

[0069] S111: collecting wind farm power output data in a real-time monitoring manner, identifying the maximum and minimum values ​​of the power data within a specified time interval, and obtaining a maximum fluctuation range analysis result by analyzing the maximum power fluctuation of the wind farm within the specified time interval;

[0070] To identify the maximum and minimum values ​​of power data, the formula is used:

[0071]

[0072] in, It represents the extreme value difference, that is, the maximum fluctuation amplitude of power output in a given time period. Representation dataset The maximum value in Representation dataset The minimum value in It is a power data set, collected by monitoring equipment at a wind farm, containing power output data points within a specific time range.

[0073] Collect power output data over a certain time range , the data between 9:00AM and 9:05AM is Kilowatts. Find the maximum and minimum values, that is, kW and kW, therefore, the extreme value difference kW. The result reflects the maximum fluctuation of wind power output in a short period of time.

[0074] S112: Based on the maximum fluctuation range analysis result, according to the degree of deviation between the data point and its average value, the average fluctuation of the entire data set is analyzed to obtain the power average fluctuation analysis result;

[0075] For analyzing the average fluctuation of the entire data set, the formula is used:

[0076]

[0077] in, It represents the mean difference, which calculates the average of the squared differences between the data points and the average power to reflect the overall volatility of the power data. Indicates the sequence number of the data point, is the total number of data points, i.e., the dataset The number of elements in , Representation dataset Middle The power value of the data point, It is a dataset The average power.

[0078] Using the Dataset , calculate the average power kW.

[0079] Calculate the square difference between the power of each data point and the average power, further sum and average to get the average difference kW². The result reflects the average fluctuation between the data points.

[0080] S113: Based on the power average fluctuation analysis result, by comparing the baseline fluctuation value of the operating data, the fluctuation behavior and stability of the wind farm in the current monitoring period are evaluated, the current data change trend is analyzed, and the power fluctuation characteristic analysis result is generated.

[0081] In the data monitoring of the wind farm, the current extreme value difference Difference between kW and average value kW². If the historical average extreme value difference is The difference between kW and the historical average is kW². By comparison, the current extreme value difference is larger than the historical average kW, indicating that the wind farm has experienced greater fluctuations in extreme conditions than historical data during the observation period, which may indicate a significant change in wind conditions or a reduction in equipment efficiency. Similarly, the current average difference is higher than the historical average. kW², further indicating an increase in overall volatility.

[0082] See also Figure 3 ,The specific steps for obtaining the distribution characteristics of local data are as follows:

[0083] S211: Based on the power fluctuation characteristics analysis results, analyze the power data frequency output by the wind farm at each time point, using the formula:

[0084]

[0085] Calculation time The weighted local data density of , generate data density information;

[0086] in, is the total number of data points, the number of wind farm power output data points obtained by the real-time monitoring system, is the Dirac function, when The time value is 1, otherwise it is 0, used to count specific time points The number of data points on , is a specified time point, It is Time labels for each data point;

[0087] for , using the formula:

[0088]

[0089] in, and It's in time The extreme value difference and the mean value difference.

[0090] In time The wind farm recorded power data three times, with a total of 100 time points. At these three time points, the extreme value differences of the power data were kW, kW, kW, the average value difference is kW², kW², kW², then:

[0091] kW

[0092] kW²

[0093]

[0094] therefore,

[0095]

[0096] The results show that at the 5th minute time point, the weighted local data density is 0.18255, which reflects the frequency of the data points and also takes into account the volatility of the data points, providing a more complex understanding of wind power fluctuations.

[0097] S212: comparing the average data density of the operating data according to the data density information, analyzing whether there is abnormal power fluctuation at each time point, and obtaining a distribution feature analysis result;

[0098] Based on the obtained weighted local data density (Take the 5th minute as an example), if the past running data shows the local data density The average value of is 0.05. Compared with this baseline, the local data density at the 5th minute is Significantly higher than the average level indicates that wind power output is unusually active at this point in time. This significant increase may reflect the strengthening of wind power or the improvement of equipment efficiency. It can more accurately reveal the power fluctuation trend at a specific point in time and provide key data for wind power forecasting.

[0099] See also Figure 4 , the steps to obtain the complexity of quantifying power fluctuations are as follows:

[0100] S221: According to the results of the distribution characteristic analysis, through the fractal dimension analysis process, the formula is used:

[0101]

[0102] Calculating fractal dimension , identify repeated patterns and structural changes in data sets, and evaluate the complexity and self-similarity of data to generate data complexity analysis results;

[0103] in, Used to quantify the complexity and self-similarity of data, is the box size covering the dataset, a measure used to observe the self-similarity of the dataset at different levels, is in scale Next, considering the time point The local data density The minimum number of boxes required after adjustment is, It's at the time The power output data is directly obtained from the real-time monitoring equipment of the wind farm, which affects the calculation of the number of boxes. is the coverage factor, which is set according to the size and complexity of the dataset to accommodate datasets of various sizes. and is the adjustment coefficient, which adjusts the calculation sensitivity of the fractal dimension according to the data characteristics. Based on the self-similar feature setting of the data set at different scales, Set according to the statistical characteristics and distribution of the data set. is the natural logarithm.

[0104] Setting parameters , for power output data Conduct fractal dimension analysis. , and set , , ,but:

[0105]

[0106]

[0107]

[0108] The results help to understand the power output fluctuation behavior of wind farms under different conditions, provide data support for power dispatching, ensure the stable operation of the power grid and optimize energy management strategies, thereby improving the operational efficiency and reliability of wind farms.

[0109] S222: Based on the data complexity analysis results, combined with power operation data and current environmental variables, a wind power fluctuation prediction model is established through data integration and model training;

[0110] By calculating the fractal dimension , which can quantify the complexity of wind farm power output data. This value reflects the self-similarity of data at different scales. For example, if the average fractal dimension of past operating data is , the currently calculated fractal dimension , indicating that during the current observation period, the volatility and complexity of wind power output have increased compared with historical data. This change may be caused by changes in environmental factors or fluctuations in equipment performance. By analyzing each time point The local data density And these data at all scales Number of boxes under The interaction of the data can further understand the distribution and fluctuation characteristics of the data, and combine the calculated fractal dimension with other relevant parameters (such as environmental change data), integrate this information through machine learning or statistical models, and establish a wind power fluctuation prediction model. This model combines historical and real-time data to predict future power output fluctuations. By real-time monitoring of the local data density of the current power output and the calculated fractal dimension, the prediction model can effectively predict possible power changes under specific conditions, provide early warnings for the power dispatch system, and help adjust strategies to cope with possible power fluctuations, thereby optimizing the overall stability of the power system and the energy management strategy of the wind farm.

[0111] See also Figure 5 , the specific steps for calculating the future power fluctuation amplitude are:

[0112] S311: The current wind speed and environmental parameters collected are uploaded to the wind power fluctuation prediction model through sensors and data interfaces to perform power output prediction and obtain time series prediction data;

[0113] When simulating the power output scenario of the wind power fluctuation prediction model, the model first receives real-time wind speed data and environmental parameters. These data are collected in real time through the sensor network and uploaded to the prediction model through the data interface. The preprocessing steps include data cleaning and formatting to ensure the quality and consistency of the input data. Next, the model uses a pre-trained algorithm to estimate the power output under current and future conditions based on the input wind speed and environmental parameters, including an analysis of the relationship between climate patterns, historical power output data and environmental changes. The simulation results are output in the form of a time series, showing the predicted power fluctuations under different wind speeds and environmental conditions. This provides decision support information for grid operations and ensures the efficiency and response speed of energy management.

[0114] S312: Based on the time series prediction data, combined with the operating power data and the current environmental variables, the influence of wind speed and environmental factors on power fluctuation is estimated to obtain the analysis result of the future power fluctuation amplitude;

[0115] For estimating the influence of wind speed and environmental factors on power fluctuation, the formula is used:

[0116]

[0117] in, is the wind speed adjustment coefficient, which is obtained by comparing the current wind speed data obtained from the wind speed sensor with the past power data through the statistical model. is the environmental adjustment coefficient, which is obtained by regression analysis of past data based on the temperature and humidity data collected by the current environmental monitoring equipment.

[0118] Setting parameters =1.05, indicating that the power output under current wind speed conditions is expected to be 5% higher than the historical average. = 0.95, indicating that the current environmental conditions may cause the power output to be slightly lower than expected by 5%;

[0119]

[0120] The results show that the power fluctuation amplitude expected under the current wind speed and environmental conditions is 0.5, 0.509775-0.5=0.009775, or an increase of 1.955% if the power fluctuation amplitude under the baseline conditions is 0.5. The slight increase shows that the impact caused by the current wind speed and environmental factors is relatively mild, but it still needs to be paid attention to in grid dispatch and wind farm operation management to prevent potential risks and adjustments.

[0121] See also Figure 6 , the specific steps for sampling the simulated data are:

[0122] S321: Based on the analysis results of future power fluctuation amplitude, combined with real-time wind speed and environmental parameters, random samples of power fluctuation are collected, using the formula:

[0123] Calculate the total predicted power fluctuation , create a prediction dataset;

[0124] in, To identify the sampling points, is the total number of samples, indicating the number of times sampling is performed. is the scaling factor used to adjust the random samples The influence of wind speed data is improved to improve the adaptability of the model to the actual impact of different wind speeds. By analyzing the relationship between historical wind speed data and power output, It is A random sample is simulated based on the actual wind speed data, representing a possible power output value under the current wind speed conditions. is the offset coefficient, which is used to adjust the baseline of each sample to ensure that all sample values ​​have a realistic possible offset and the reality of the model. It is obtained through statistical analysis of past data. It is a global adjustment coefficient that adjusts the amplitude of the entire model output according to changes in the real-time environment or system status.

[0125] Setting parameters , , , .like The value of is randomly generated and uniformly distributed between 0.45 and 0.56. is 1000, that is, 1000 independent samplings are performed, and the calculation steps are as follows:

[0126]

[0127]

[0128]

[0129]

[0130]

[0131] The results show that under the current wind speed and environmental conditions, the total power fluctuation amplitude predicted by the overall simulated 1,000 samples is 322.4, which provides key quantitative information for the power dispatching system to help dispatchers understand and prepare for possible power fluctuations.

[0132] S322: According to the prediction data set, collect all the individual simulation results, including the power prediction under each wind speed and environmental condition, classify and store the results, and generate a prediction scenario sample set;

[0133] After the sampling is completed, the simulation results are then integrated to form a comprehensive prediction scenario sample set. First, all individual simulation results, that is, the power fluctuation prediction for each sampling point, are collected and stored in a data structure. This data includes power output predictions under different wind speeds and environmental parameters. In order to integrate this data, each prediction result is labeled with its corresponding environmental and wind speed parameters. Then, these data are analyzed by statistical methods. In this way, the power fluctuation performance under various conditions can be understood from multiple angles. Finally, these statistical analysis results are compiled into a comprehensive sample set, which contains detailed predictions of possible future situations, providing decision support for power dispatching, ensuring that every possible scenario has been considered and analyzed, thereby providing data support for the operation of wind farms and the stable operation of power grids.

[0134] See also Figure 7 , the specific steps for obtaining power dispatch adjustment are:

[0135] S411: Based on the predicted scenario sample set, identify the peak load and minimum load period in the scenario, compare the energy storage utilization and load response mode, identify potential power demand pressure points, classify the scenario according to demand intensity, and obtain analyzed scenario data;

[0136] The analysis predicts the energy storage demand and grid load conditions of the simulated scenarios in the scenario sample set, including identifying the peak load and minimum load cycle in each simulated scenario, and determining possible pressure points or deficiencies by comparing the energy storage utilization and load response patterns of each scenario. Each scenario is labeled and classified by demand intensity. Such classification helps operators understand under what specific conditions additional energy storage support or load management measures may be required. For example, scenario analysis may reveal that the load on the grid exceeds the maximum output capacity of the energy storage system during high-demand periods, indicating the need to increase energy storage capacity or adjust load management strategies to avoid potential power outages.

[0137] S412: Based on the analyzed scenario data, evaluate the operation status of the power grid under multiple power output conditions, and simulate the response capability and stability of the power grid by adjusting the power grid capacity and the influence of environmental factors, using the formula:

[0138]

[0139] Calculating the index of grid stability ;

[0140] in, It is the grid capacity adjustment factor, which is calculated based on the grid design capacity and real-time load capacity. is the current load of the power grid, which is obtained from real-time monitoring of the power grid operation data. It is the environmental factor adjustment coefficient, which is adjusted according to the specific impact of current climate conditions and environmental conditions. It is an environmental quality indicator that comprehensively evaluates environmental parameters such as temperature and humidity and is obtained through environmental monitoring equipment.

[0141] Setting parameters , , , ;

[0142]

[0143]

[0144]

[0145]

[0146] The results show that under the current grid capacity and environmental conditions, the global stability index of the grid is 282.1, which provides key information for power dispatching to optimize the operation of the power system and ensure the stable operation of the grid under different power output conditions.

[0147] S413: Analyze the power demand and energy storage utilization in multiple scenarios according to the grid stability index, adjust the power generation plan and energy storage scheduling, optimize the grid response, and generate power scheduling optimization results.

[0148] The operation of adjusting power dispatch according to demand changes in simulated scenarios includes analyzing the power demand and energy storage utilization of each scenario, especially during peak and trough periods. These data points provide key information that needs to be considered in power dispatch, and optimize the response of the power grid by adjusting the power generation plan, energy storage dispatch and load management strategy. For example, in scenarios with predicted high demand, more energy storage resources can be dispatched in advance or backup power generation facilities can be started to ensure that the power grid maintains stable operation under different power demand conditions. At the same time, the power grid operation mode is adjusted according to the actual data, such as demand response and load reduction plans, to generate the final power dispatch optimization results. These results help ensure the efficiency and reliability of power grid operations and reduce the risks caused by power fluctuations.

[0149] A power dispatching system based on wind power prediction, the power dispatching system based on wind power prediction is used to execute the above-mentioned power dispatching method based on wind power prediction, the system comprises:

[0150] The wind speed and power analysis module collects the operating wind speed data and power output data of the wind farm, calculates the extreme value difference and average value difference of the time points in the time series, analyzes the current data change trend, and generates power fluctuation characteristic analysis results;

[0151] The local density calculation module analyzes the local data density at each time point based on the power fluctuation characteristic analysis results, calculates the distribution characteristics of the local data, and calculates the fractal dimension to quantify the complexity of the power fluctuation to obtain the local characteristic information;

[0152] The fluctuation prediction model building module uses local characteristic information, combined with current wind speed and environmental parameters, to build a wind power fluctuation prediction model, and samples the simulated data, integrating all simulation results to generate a prediction scenario sample set;

[0153] The power demand and load analysis module analyzes the energy storage demand and grid load in the simulation scenario based on the prediction scenario sample set, evaluates the operation status of the grid under various power output conditions, and obtains the grid operation status evaluation results;

[0154] The dispatch strategy optimization module adjusts power dispatch according to the evaluation results of the power grid operation status and generates power dispatch optimization results.

[0155] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A power dispatching method based on wind power prediction, characterized in that: The following steps are involved: Collect the wind speed data and power output data of the wind farm, calculate the extreme value difference and average value difference of the power time point in the time series according to the wind speed and corresponding power information at each time point, analyze the current data change trend, and generate the power fluctuation characteristic analysis results; According to the power fluctuation characteristic analysis results, the local data density of the power output data at each time point is calculated, the distribution characteristics of the local data are analyzed, and the fractal dimension is calculated to quantify the complexity of the power fluctuation, and the data is integrated to establish a wind power fluctuation prediction model; The steps of obtaining the distribution characteristics of the local data are specifically as follows: Based on the power fluctuation characteristic analysis results, the frequency of power data output by the wind farm at each time point is analyzed, using the formula: Calculation time The weighted local data density of , generate data density information; in, is the total number of data points, Indicates when The value is 1 when the is a specified point in time, It is Time labels for each data point; for , using the formula: in, and It's in time The extreme value difference and the mean value difference; According to the data density information, the average data density of the operating data is compared, and whether there is abnormal power fluctuation at each time point is analyzed to obtain a distribution characteristic analysis result; The steps for obtaining the complexity of the quantified power fluctuation are specifically as follows: According to the distribution characteristics analysis results, through the fractal dimension analysis process, the formula is adopted: Calculating fractal dimension , identify repeated patterns and structural changes in data sets, and evaluate the complexity and self-similarity of data to generate data complexity analysis results; in, is the size of the box covering the dataset, is in scale Next, considering the time point The local data density The minimum number of boxes required after adjustment is, It's at the time Power output data, is the coverage factor, and is the adjustment factor; Based on the data complexity analysis results, combined with power operation data and current environmental variables, a wind power fluctuation prediction model is established through data integration and model training; Based on the wind power fluctuation prediction model, combined with the current wind speed and environmental parameters, the power output scenario is simulated, the future power fluctuation amplitude under various wind speed conditions is calculated, and the simulated data is sampled, all simulation results are integrated, and a prediction scenario sample set is generated; Based on the predicted scenario sample set, the energy storage demand and grid load in the simulated scenario are analyzed, the operating status of the grid under various power output conditions is evaluated, and power dispatching adjustments are made according to demand changes in the simulated scenario to generate power dispatching optimization results.

2. The power dispatching method based on wind power prediction according to claim 1 is characterized in that: The acquisition steps for analyzing the current data change trend are specifically as follows: Collect wind farm power output data through real-time monitoring, identify the maximum and minimum values ​​of power data within a specified time interval, and obtain the maximum fluctuation range analysis results by analyzing the maximum power fluctuation of the wind farm within the specified time interval; Based on the maximum fluctuation range analysis result, according to the degree of deviation between the data point and its average value, the average fluctuation of the entire data set is analyzed to obtain the power average fluctuation analysis result; According to the power average fluctuation analysis result, by comparing the benchmark fluctuation value of the operating data, the fluctuation behavior and stability of the wind farm in the current monitoring period are evaluated, the current data change trend is analyzed, and the power fluctuation characteristic analysis result is generated.

3. The power dispatching method based on wind power prediction according to claim 1 is characterized in that: The specific steps for calculating the future power fluctuation amplitude are as follows: The collected current wind speed and environmental parameters are uploaded to the wind power fluctuation prediction model through sensors and data interfaces to perform power output prediction and obtain time series prediction data; Based on the time series forecast data, combined with the operating power data and current environmental variables, the impact of wind speed and environmental factors on power fluctuations is estimated to obtain future power fluctuation amplitude analysis results; For estimating the influence of wind speed and environmental factors on power fluctuation, the formula is used: in, is the wind speed adjustment factor, is the environmental adjustment factor.

4. The power dispatching method based on wind power prediction according to claim 3 is characterized in that: The acquisition step of sampling the simulated data is specifically as follows: According to the analysis results of the future power fluctuation amplitude, combined with the real-time wind speed and environmental parameters, random samples of power fluctuations are collected, using the formula: Calculate the total predicted power fluctuation , create a prediction dataset; in, To identify the sampling points, is the total number of samples, is the scaling factor, It is A random sample, is the offset coefficient, is the global adjustment coefficient; Based on the prediction data set, all individual simulation results are collected, including power prediction under each wind speed and environmental condition, and the results are classified and stored to generate a prediction scenario sample set.

5. The power dispatching method based on wind power prediction according to claim 4 is characterized in that: The steps for obtaining the power dispatch adjustment are specifically as follows: Based on the predicted scenario sample set, identifying the peak load and minimum load period in the scenario, comparing the energy storage utilization and the load response mode, identifying the potential power demand pressure points, classifying the scenario according to the demand intensity, and obtaining the analyzed scenario data; Based on the scenario data analyzed, the operation status of the power grid under multiple power output conditions is evaluated, and the response capability and stability of the power grid are simulated by adjusting the power grid capacity and the influence of environmental factors. The formula is used: Calculating the index of grid stability ; in, is the grid capacity adjustment factor, is the current load of the grid, is the environmental factor adjustment coefficient, It is an indicator of environmental quality; According to the grid stability index, the power demand and energy storage utilization in multiple scenarios are analyzed, the power generation plan and energy storage scheduling are adjusted, the response of the grid is optimized, and the power scheduling optimization result is generated.

6. A power dispatching system based on wind power prediction, characterized in that: According to the power dispatching method based on wind power prediction according to any one of claims 1 to 5, the system comprises: The wind speed and power analysis module collects the operating wind speed data and power output data of the wind farm, calculates the extreme value difference and average value difference of the power time point in the time series, analyzes the current data change trend, and generates power fluctuation characteristic analysis results; The local density calculation module analyzes the local data density of the power output data at each time point based on the power fluctuation characteristic analysis result, calculates the distribution characteristics of the local data, and calculates the fractal dimension to quantify the complexity of the power fluctuation to obtain local characteristic information; The fluctuation prediction model building module uses the local characteristic information, combines the current wind speed and environmental parameters, builds a wind power fluctuation prediction model, samples the simulated data, integrates all simulation results, and generates a prediction scenario sample set; The power demand and load analysis module analyzes the energy storage demand and grid load in the simulation scenario based on the prediction scenario sample set, evaluates the operation status of the grid under various power output conditions, and obtains the grid operation status evaluation result; The dispatch strategy optimization module performs power dispatch adjustment according to the power grid operation status evaluation result and generates a power dispatch optimization result.

Images