A high-permeability new energy power distribution network collaborative optimization method, system, device and storage medium

Abstract

The application discloses a kind of high-permeability new energy distribution network collaborative optimization method, system, equipment and storage medium, belong to information technology field, including: obtaining wind farm operating data and signal strength, processing and fitting signal attenuation trend, identifying thin fog environment path loss, obtaining wind power output probability distribution function, determining output fluctuation range, predicting output deviation scenario, calculating wireless signal fluctuation probability, extracting signal fluctuation characteristics;Channel fading model is established in combination with thin fog path loss, output wireless signal propagation delay, determine the potential delay interval of energy storage system scheduling instruction;Simulate scheduling disorder and power regulation instruction conflict, generate scheduling disorder evaluation data;Combining historical wind power output data judges tidal flow fluctuation, adjusts energy storage unit response order, generates power regulation instruction sequence, calculates adjusted tidal flow distribution data, generates energy storage and topology configuration optimization data, significantly improves wind farm operating reliability and economy.

Description

Technical Field

[0001] This invention relates to the field of information technology, specifically to a method, system, device, and storage medium for collaborative optimization of high-penetration new energy distribution networks. Background Technology

[0002] The application of virtual reality (VR) technology in the field of renewable energy microgrids is attracting increasing attention. By constructing immersive scenarios, it provides a new perspective for optimizing the operation of complex power grids, which is particularly crucial in scenarios with complex geographical environments and variable climates, such as mountainous wind power microgrids. The stable operation of microgrids is directly related to the efficient utilization of renewable energy and the reliability of regional energy supply. VR technology can significantly improve the decision-making efficiency of operation and maintenance personnel through visualization and interaction. However, existing solutions have significant shortcomings in handling uncertainties in complex scenarios. Traditional methods often rely on static models or single data analysis, which are difficult to cope with real-time changes caused by dynamic environments such as hilly areas and fog, leading to prediction deviations that are out of sync with actual operation and affecting the timeliness and accuracy of dispatch instructions. In hilly and foggy scenarios, the uncertainty of renewable energy output becomes the primary challenge. This uncertainty is first reflected in the fact that foggy weather often leads to deviations in wind power output predictions, which not only directly affects the power balance of the microgrid, but more complexly, due to the complex terrain and tree cover in mountainous areas, the already fog-affected wireless signal fluctuations are further aggravated. This signal instability directly leads to delays in the transmission of dispatch commands for the energy storage system. In turn, the command delay exacerbates the incoordination between microgrid topology configuration and energy storage dispatch, ultimately resulting in reduced operating efficiency or even threatening system stability.

[0003] Therefore, a series of closely related technical bottlenecks, from output prediction deviations to signal fluctuations and scheduling delays, collectively hinder the effective implementation of dynamic optimization schemes. How to integrate scenario analysis and probabilistic power flow data in a virtual reality environment to dynamically present optimization schemes and assist maintenance personnel in adjusting energy storage and topology configurations in real time has become a key issue in improving the operational efficiency and user trust of mountainous wind power microgrids. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is to address the issue that traditional methods rely heavily on static models or single data analysis, which are difficult to cope with real-time changes caused by dynamic environments such as hilly areas and light fog, resulting in prediction deviations that are out of sync with actual operation and affecting the timeliness and accuracy of scheduling instructions.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a collaborative optimization method for high-penetration renewable energy distribution networks, comprising:

[0007] Acquire wind farm operation data and signal strength, perform data processing on the wind farm operation data and signal strength, and obtain the signal attenuation trend;

[0008] The signal attenuation trend is fitted to obtain the signal attenuation characteristics. Based on the signal attenuation characteristics, the path loss in the fog environment is identified, and the wind power output probability distribution function is obtained.

[0009] The power output fluctuation range is obtained based on the wind power output probability distribution function, and the power output deviation scenarios are predicted. The wireless signal fluctuation probability of each scenario is calculated, and the signal fluctuation distribution characteristics are extracted.

[0010] A channel fading model is established based on the signal fluctuation distribution characteristics and the path loss in the foggy environment. The wireless signal propagation delay is output to obtain the transmission time series. Based on the transmission time series, the potential delay interval of the energy storage system scheduling command is obtained.

[0011] Simulate scheduling disorder and power regulation command conflicts within potential delay intervals, identify allocation deviations between microgrid topology configuration and energy storage scheduling, and generate scheduling disorder assessment data;

[0012] Power flow fluctuations are extracted from the scheduling disorder assessment data and judged by combining them with historical wind power output data;

[0013] Based on the judgment results, the response sequence of the energy storage units is adjusted, a power regulation command sequence is generated, the adjusted power flow distribution data is calculated, and energy storage and topology configuration optimization data are generated.

[0014] As a preferred embodiment of the high-penetration new energy distribution network collaborative optimization method described in this invention, the wind power output probability distribution function includes:

[0015] Obtain the signal sequence of the wind farm, preprocess the signal sequence, and establish a signal amplitude observation matrix;

[0016] The signal amplitude observation matrix is ​​normalized, and the signal amplitude probability distribution function is generated by fitting the least squares method.

[0017] The scattering coefficient is calculated based on the signal amplitude probability distribution function, and the electromagnetic wave energy loss matrix is ​​generated by combining the distance attenuation parameter.

[0018] Establish path loss mapping relationship based on electromagnetic wave energy loss matrix, calculate transmission path loss probability distribution, and generate wind power output probability function.

[0019] The beneficial effects of this preferred technical solution are that, through signal processing and modeling, the uncertainty of wind power output is quantified, providing a precise basis for optimized scheduling and improving the reliability and economy of system operation.

[0020] As a preferred embodiment of the high-penetration new energy distribution network collaborative optimization method described in this invention, the wind power output probability distribution function further includes:

[0021] Obtain the actual power output record of the wind farm, perform wavelet decomposition on the actual power output record of the wind farm, and obtain a smooth power output curve;

[0022] The smooth output curve is segmented according to a fixed time window, and the peak point set and valley point set within the time window are calculated to obtain the output fluctuation frequency characteristics.

[0023] A training sample matrix is ​​generated based on the power output fluctuation frequency characteristics and environmental parameter sequences. Time series features are extracted to obtain the wind power output probability function.

[0024] The beneficial effects of this preferred technical solution are that it accurately quantifies the uncertainty of wind power output and enhances the accuracy and reliability of dispatch optimization.

[0025] As a preferred embodiment of the high-penetration new energy distribution network collaborative optimization method described in this invention, the extraction of signal fluctuation distribution features includes:

[0026] The Monte Carlo method was used to randomly sample the wind power output probability function to obtain a set of output deviation prediction data.

[0027] Based on the output deviation prediction data set, determine the fluctuation amplitude threshold, divide the output fluctuation scenarios, and generate the probability value of scenario occurrence.

[0028] An electromagnetic wave propagation path map was established based on forest cover density map and topographic elevation data, and signal transmission loss distribution data was calculated.

[0029] Data processing is performed on the signal transmission loss distribution data to reconstruct and generate signal fluctuation characteristic sequence data.

[0030] The beneficial effect of this preferred technical solution is that it accurately characterizes signal fluctuation characteristics through random sampling and feature extraction, providing data support for subsequent optimization.

[0031] As a preferred embodiment of the high-penetration new energy distribution network collaborative optimization method described in this invention, the potential delay interval of the energy storage system dispatch command includes:

[0032] Set upper and lower limits for signal amplitude thresholds based on historical signal fluctuation statistics, calculate the mean and standard deviation of real-time signals, and determine whether signal fluctuations exceed the upper and lower limits for signal amplitude thresholds.

[0033] For signal segments that exceed the upper and lower limits of the signal amplitude threshold, a channel fading model is established by combining the signal fluctuation distribution characteristics and fog path loss.

[0034] Output the wireless signal propagation delay to determine the transmission time sequence of energy storage scheduling commands;

[0035] The transmission and reception timestamps of energy storage scheduling commands are obtained based on the wireless signal transmission delay, and a command transmission delay time series is generated.

[0036] Data processing is performed on the instruction transmission delay time series to generate a smooth delay sequence;

[0037] Calculate the statistical parameters of the smoothed delayed sequence and generate the probability distribution function of the delay time;

[0038] Error parsing logs are extracted from probability distribution functions, error type correlations are analyzed, delay influencing factors are identified, and potential delay intervals are generated.

[0039] As a preferred embodiment of the high-penetration renewable energy distribution network collaborative optimization method described in this invention, the scheduling disorder assessment data includes:

[0040] Obtain the timestamp information of the energy storage dispatch instruction sequence, extract the power regulation direction parameter and regulation amplitude parameter, and generate the instruction execution sequence;

[0041] Generate a perturbation execution sequence based on the instruction execution sequence;

[0042] The disturbance execution sequence is scanned to extract the adjustment direction sequence and determine the power conflict points of adjacent commands;

[0043] Based on the timing deviation and command conflict characteristics, output energy storage scheduling disorder assessment data.

[0044] As a preferred embodiment of the high-penetration new energy distribution network collaborative optimization method described in this invention, the energy storage and topology configuration optimization data includes:

[0045] Based on the energy storage dispatch disorder assessment data, the characteristics of power flow fluctuations caused by delays are extracted;

[0046] By combining historical wind power output data, it can be determined whether the power flow fluctuations exceed the preset range;

[0047] If the value exceeds the preset range, the energy storage unit's charging and discharging power data and response time data are acquired to generate an energy storage characteristic database.

[0048] Calculate the response time weighting coefficient based on the energy storage characteristic database, and generate a response time series mapping table;

[0049] A piecewise linear programming method is used to process the response time-series mapping table, set constraints, and generate power regulation commands.

[0050] Calculate power flow distribution based on power regulation commands, generate power flow distribution dataset, construct network status evaluation indicators, and generate energy storage and topology configuration optimization data.

[0051] This invention provides a high-penetration new energy distribution network collaborative optimization system.

[0052] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a high-penetration renewable energy distribution network collaborative optimization system, comprising:

[0053] The data processing module acquires wind farm operation data and signal strength, processes the wind farm operation data and signal strength, and obtains the signal attenuation trend.

[0054] The identification module fits the signal attenuation trend to obtain signal attenuation characteristics, identifies path loss in foggy environments based on the signal attenuation characteristics, and obtains the wind power output probability distribution function.

[0055] The calculation module obtains the power output fluctuation range based on the wind power output probability distribution function, predicts power output deviation scenarios, calculates the wireless signal fluctuation probability for each scenario, and extracts signal fluctuation distribution features.

[0056] The model building module establishes a channel fading model based on the signal fluctuation distribution characteristics and the path loss in the foggy environment, outputs the wireless signal propagation delay, obtains the transmission time series, and obtains the potential delay interval of the energy storage system scheduling command based on the transmission time series.

[0057] The simulation module simulates scheduling disorder and power regulation command conflicts within the potential delay interval, identifies the allocation deviation between microgrid topology configuration and energy storage scheduling, and generates scheduling disorder assessment data.

[0058] The judgment module extracts power flow fluctuations from the scheduling disorder assessment data and judges the power flow fluctuations by combining them with historical wind power output data.

[0059] The collaborative optimization module adjusts the response sequence of energy storage units based on the judgment results, generates a power regulation command sequence, calculates the adjusted power flow distribution data, and generates energy storage and topology configuration optimization data.

[0060] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the aforementioned high-penetration new energy distribution network collaborative optimization method.

[0061] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the aforementioned high-penetration new energy distribution network collaborative optimization method.

[0062] The beneficial effects of this invention are as follows: This invention discloses a collaborative optimization method for high-penetration new energy distribution networks. By collecting data such as wind speed and fog density, Rayleigh fading parameters and signal attenuation characteristics are extracted, path loss in foggy environments is identified, and the probability distribution function of wind power output is determined. A channel fading model is established based on signal fluctuation characteristics to analyze the potential delay of energy storage dispatch commands and assess dispatch disorder issues. To address power flow fluctuations caused by delays, this invention adjusts the response sequence of energy storage units, generates a power regulation scheme, and applies a multi-objective optimization algorithm to obtain an optimized scheme for energy storage and topology configuration. This invention effectively solves the problem of energy storage dispatch delay caused by unstable wireless communication in wind farms in hilly foggy environments, improving the reliability and economy of wind farm operation. Attached Figure Description

[0063] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 The above is a flowchart of a high-penetration new energy distribution network collaborative optimization method provided in one embodiment of the present invention. Detailed Implementation

[0065] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0066] Example 1, referring to Figure 1 This is one embodiment of the present invention, which provides a collaborative optimization method for high-penetration renewable energy distribution networks, comprising:

[0067] S100: Acquire wind farm operation data and signal strength, process the wind farm operation data and signal strength to obtain the signal attenuation trend;

[0068] In this embodiment of the invention, wind farm operation data under hilly fog conditions is obtained. The wind speed, fog density, and signal strength of the tree-covered area are collected by sensors, and the signal attenuation trend is obtained by filtering and noise reduction.

[0069] Specifically, at each wind turbine, height-adjustable anemometers measure wind speed data within the range from the ground to the rotor height. A spatial distribution sampling point set for wind speed is established based on the wind turbine blade rotation speed and the data synchronously recorded by the anemometers. Gaussian filtering is applied to this sampling point set to eliminate abnormal data fluctuations, with the filter window size set to three times the sampling time interval, resulting in a stable wind speed dataset. Atmospheric visibility and relative humidity parameters within the wind farm area are obtained from meteorological sensors. Fog concentration data is measured at different heights using a scattering-type concentration meter, and the scattering attenuation matrix of the fog on electromagnetic waves is obtained using Mie scattering theory. A three-dimensional topographic model is constructed based on geographic information measurement data of the forest-covered area near the wind turbine towers. LiDAR scanning is used to acquire topographic elevation point cloud data within the forest-covered area, and the regional shading intensity distribution map is obtained from the point cloud density. The stable wind speed dataset, the scattering attenuation matrix, and the regional shading intensity distribution map are synchronized in time and registered spatially. A weighted average method is used to construct a comprehensive environmental impact factor dataset. Wavelet transform processing was performed on the signal intensity data collected by wind turbine sensors. The db4 wavelet basis function (Dobessie series wavelet basis function) was selected to perform a three-level decomposition of the signal, and a soft thresholding method was used to remove high-frequency noise components. Singular value decomposition was then employed to extract features from the denoised signal data matrix, selecting the singular value vectors corresponding to eigenvalues ​​with a cumulative contribution rate of 95%. Based on the comprehensive environmental impact factor dataset and the signal feature vectors, a signal attenuation prediction model was constructed, and the signal attenuation trend curves for each monitoring area were calculated.

[0070] The collection of wind farm operation data relies on virtual reality technology, enabling real-time monitoring and analysis through an immersive interface. This ensures that operation and maintenance personnel can intuitively grasp the grid's operational status in complex environments. Data collection can be triggered via sensor networks or remote commands, with the specific method flexibly determined based on the actual scenario.

[0071] It should be noted that height-adjustable wind speed sensors were deployed at each wind turbine location in the wind farm to measure wind speed within the range from the ground to the rotor height. The sampling frequency was 10 Hz, and the sampling time was no less than 10 minutes. Based on the wind speed data recorded synchronously with the turbine blade rotation speed, a spatially distributed sampling point set was constructed. Gaussian filtering was used to eliminate random fluctuations caused by turbulence, with the filter window size being three times the sampling time interval, generating a stable wind speed dataset. The data smoothness was significantly improved, and the standard deviation was reduced by approximately 50%. Fog concentration data were collected at different heights using a scattering concentration meter. Combined with atmospheric visibility and humidity parameters provided by meteorological sensors, the scattering attenuation of electromagnetic waves was calculated based on Mie scattering theory, generating a scattering attenuation numerical matrix. Actual measurements showed that the fog concentration was high within 30 meters of the ground, resulting in electromagnetic wave energy loss accounting for approximately 30% of the total attenuation. Point cloud data was obtained by scanning the tree-covered area using lidar, with a point cloud density of 50 points per square meter within a 300-meter range. Based on point cloud density, regional occlusion intensity is calculated, generating an occlusion intensity distribution map that clearly reflects the impact of forest height and leaf area on signal propagation. Occlusion attenuation accounts for approximately 45% of the total loss. Temporal and spatial registration is performed on the wind speed dataset, scattering attenuation numerical matrix, and occlusion intensity distribution map. A weighted average method is used to integrate the data, generating a comprehensive environmental impact factor dataset. This embodiment of the invention does not impose excessive limitations on the registration algorithm; technicians can choose a suitable method according to actual needs. Signal intensity data is decomposed into three levels using the db4 wavelet basis function, and a soft thresholding denoising method is applied to remove high-frequency noise, improving the signal-to-noise ratio by approximately 6 dB. Noise-reduced signal features are extracted through singular value decomposition, and vectors corresponding to feature values ​​with a cumulative contribution rate of 95% are selected to generate a noise-reduced signal data matrix. The correlation coefficient between the reconstructed signal and the original signal reaches over 0.95.

[0072] S200: Fit the signal attenuation trend to obtain the signal attenuation characteristics, identify the path loss in the fog environment based on the signal attenuation characteristics, and obtain the wind power output probability distribution function.

[0073] S300: Obtain the power output fluctuation range based on the wind power output probability distribution function, predict power output deviation scenarios, calculate the wireless signal fluctuation probability for each scenario, and extract signal fluctuation distribution characteristics.

[0074] S400: Establish a channel fading model based on signal fluctuation distribution characteristics and path loss in foggy environments, output wireless signal propagation delay, obtain transmission time series, and obtain the potential delay interval of energy storage system scheduling commands based on the transmission time series.

[0075] S500: Simulates scheduling disorder and power regulation command conflicts within potential delay intervals, identifies allocation deviations between microgrid topology configuration and energy storage scheduling, and generates scheduling disorder assessment data;

[0076] S600: Extract power flow fluctuations from the dispatch disorder assessment data and combine them with historical wind power output data to judge the power flow fluctuations;

[0077] Specifically, dispatch disorder assessment data and microgrid topology data are collected to establish node power injection vectors and admittance matrices. The initial power flow distribution is obtained by solving the power flow equations using the Newton-Raphson iterative method. The power flow calculation results are reconstructed into a time series, and the power change rate sequence and voltage deviation sequence are calculated using the difference method. Abnormal fluctuation points are eliminated by median filtering. A sliding time window is used to segment the filtered sequence, with the window length set as a multiple of the sampling period. The mean and standard deviation within the window are calculated as statistical features. Wavelet transform is performed on the statistical feature sequence, and the db4 wavelet basis function is selected for three-level decomposition to extract high-frequency and low-frequency coefficients, reconstructing the fluctuation feature curve. Wind power output data is read from a historical database, and the power change trend curve is fitted using the least squares method to calculate the deviation sequence between the actual power and the trend curve. An adaptive threshold is set based on the deviation sequence, with the threshold parameter dynamically adjusted according to the standard deviation of the deviation, dividing the power curve into a steady-state segment and a fluctuation segment. Fast Fourier transform is used to perform spectral analysis on the fluctuation segment data, extracting the amplitude and phase characteristics of the main frequency components to construct a fluctuation feature vector. The fluctuation feature vector is compared with the preset fluctuation range, and the probability of fluctuation exceeding the limit is calculated by kernel density estimation to obtain the over-limit assessment result of the tidal fluctuation.

[0078] It should be noted that the data collected for scheduling disorder assessment and microgrid topology information are used to construct node power injection vectors and admittance matrices, with the number of nodes typically ranging from 10 to 50. The power flow equations are solved using the Newton-Raphson iterative method, converging after 4 to 6 iterations with an error less than 0.0001 per unit, generating an initial power flow distribution that reflects the power balance and network connectivity characteristics of each node, laying the foundation for subsequent fluctuation analysis. The power flow distribution data is then reconstructed as a time series, and the power change rate and voltage deviation sequences are calculated using a difference method. Under normal operating conditions, the power change rate does not exceed 20% of the rated power per minute, and the voltage deviation is controlled within ±5% of the nominal value. A 5-point median filter is used to eliminate instantaneous fluctuations caused by measurement noise, reducing the standard deviation of the filtered sequence by approximately 35%, generating a smoothed sequence that provides reliable data for feature extraction. A 10-minute sliding time window with a 5-minute step size and a 50% overlap is used for the smoothed sequence, calculating the mean, standard deviation, skewness, and kurtosis within the window to capture fast and slow fluctuation characteristics. A three-level decomposition was performed using the db4 basis function. High-frequency coefficients reflect short-term fluctuations above 0.1 Hz, while low-frequency coefficients correspond to long-term trends below 0.01 Hz, reconstructing the fluctuation characteristic curve. A 60-minute trend curve was extracted from historical wind power output data and fitted using the least squares method. The deviation sequence between actual power and trend was calculated, and an adaptive threshold of twice the standard deviation of the deviation was set to dynamically divide the steady-state and fluctuation segments to adapt to the fluctuation characteristics under different wind conditions. Fast Fourier Transform was applied to the fluctuation segment data to extract the amplitude and phase of the main frequency components. The frequency range included synoptic-scale variations from 0.001 to 0.01 Hz, turbulent fluctuations from 0.01 to 0.1 Hz, and noise components above 0.1 Hz. A fluctuation feature vector was constructed, and combined with a preset fluctuation range, a Gaussian kernel was selected through kernel density estimation, and the bandwidth was optimized through cross-validation to calculate the probability of fluctuation exceeding limits. The results show that the probability of exceeding limits increased from 5% to 15% under strong wind conditions, with an increased proportion of high-frequency components, reflecting the significant impact of turbulence on grid stability and providing a precise basis for optimized scheduling.

[0079] S700: Based on the judgment results, adjust the response sequence of the energy storage units, generate a power regulation command sequence, calculate the adjusted power flow distribution data, and generate energy storage and topology configuration optimization data.

[0080] It should be noted that by integrating multi-source data and dynamic modeling, the uncertainty of wind power output can be accurately quantified, energy storage scheduling and microgrid topology configuration can be optimized, significantly improving the operational reliability and economy of the new energy distribution network and effectively solving the scheduling delay problem in complex environments.

[0081] In this embodiment of the invention, step S200 includes the following sub-steps A1-A4;

[0082] In A1: Obtain the signal sequence of the wind farm, preprocess the signal sequence, and establish the signal amplitude observation matrix;

[0083] In A2: The signal amplitude observation matrix is ​​normalized, and the signal amplitude probability distribution function is generated by fitting the least squares method.

[0084] In A3: The scattering coefficient is calculated based on the signal amplitude probability distribution function, and the electromagnetic wave energy loss matrix is ​​generated by combining the distance attenuation parameter;

[0085] In A4: Establish the path loss mapping relationship based on the electromagnetic wave energy loss matrix, calculate the transmission path loss probability distribution, and generate the wind power output probability function.

[0086] In this embodiment of the invention, there are two ways to obtain the wind power output probability function;

[0087] Specifically, the first method for obtaining the wind power output probability function involves preprocessing the instantaneous signal data sequence collected from the wind farm, removing abnormal fluctuations through median filtering to obtain a stable signal amplitude time series, and establishing a signal amplitude observation matrix. Normalizing the signal amplitude observation matrix yields a standardized signal amplitude sequence, expressed as:

[0088] ;

[0089] in, For the normalized first Each signal amplitude value, The first in the original signal amplitude observation matrix Amplitude value, This represents the minimum value of the signal amplitude. This represents the maximum value of the signal amplitude.

[0090] Normalizing the signal amplitude value to between 0 and 1 facilitates subsequent probability distribution fitting calculations.

[0091] By fitting the Rayleigh probability density function using the least squares method, the fading depth parameter and the signal amplitude probability distribution function are extracted and expressed as:

[0092] ;

[0093] in, Here are the parameters of the Rayleigh distribution. The total number of observed data points. For the first The empirical cumulative distribution function value for each observation point Rayleigh distribution in the th The theoretical cumulative distribution function value for each observation point For the first One signal amplitude observation value.

[0094] The optimal parameters are estimated by fitting the probability distribution function of the signal amplitude using the least squares method and minimizing the squared error between the observed values ​​and the theoretical distribution.

[0095] An exponential function is used to fit the variation of the normalized signal amplitude sequence with propagation distance. The signal attenuation coefficient and characteristic distance parameters are extracted to establish a distance attenuation function. The Mie scattering coefficient is calculated based on the particle size distribution and dielectric constant of the fog particles. Combined with the distance attenuation function, an electromagnetic wave energy loss matrix is ​​constructed, yielding the scattering loss distribution function, expressed as:

[0096] ;

[0097] in, Distance Path loss at the location, For reference distance Path loss at the location, This is the path loss index. The Mie scattering coefficient is... For transmission distance, For reference distance.

[0098] A path loss mapping relationship is established based on the electromagnetic wave energy loss matrix and the scattering loss distribution function. The total loss probability distribution along the signal transmission path is estimated using a Bayesian probability estimation method. The path loss probability distribution is randomly sampled using the Monte Carlo method. Combined with the characteristics of the wind turbine power curve, the output power under different loss conditions is obtained. The wind power output probability density distribution function, i.e., the wind power output probability function, is obtained through kernel density estimation, expressed as:

[0099] ;

[0100] in, Wind power output is less than or equal to the threshold The cumulative probability, For the number of samples taken in Monte Carlo, For the first The wind power output value of the second sampling This is an indicator function; it returns 1 when the condition is met and 0 otherwise. The number of samples.

[0101] In this embodiment of the invention, after completing steps A1-A4, step S200 also includes steps A5-A7;

[0102] In A5: Obtain the actual power output record of the wind farm, perform wavelet decomposition on the actual power output record of the wind farm, and obtain a smooth power output curve;

[0103] In A6: The smooth output curve is segmented according to a fixed time window, and the peak point set and valley point set within the time window are calculated to obtain the output fluctuation frequency characteristics.

[0104] In A7: A training sample matrix is ​​generated based on the power output fluctuation frequency characteristics and environmental parameter sequences, and time series features are extracted to obtain the wind power output probability function;

[0105] Specifically, the second method for obtaining the wind power output probability function involves collecting wind turbine output data and meteorological environmental parameters through wind farm monitoring devices. The raw data is then processed using linear interpolation to fill in missing values, and median filtering is used to eliminate abnormal fluctuations, resulting in a preprocessed output time series. This preprocessed output time series is then subjected to wavelet decomposition, using the db4 wavelet basis function for a three-level decomposition. High-frequency coefficients are denoised using a soft thresholding method to reconstruct a smooth output curve. The smooth output curve is segmented using a fixed time window, with the window length set as an integer multiple of the sampling period. Output fluctuation frequency characteristics are statistically obtained from the peak and valley point sets within each window. An environmental parameter sequence is established based on wind speed and air pressure data collected from meteorological stations. The periodic characteristics of the wind speed sequence are extracted using Fourier transform, and the variation characteristics of the air pressure sequence are extracted using trend decomposition. The output fluctuation frequency characteristics and the environmental parameter sequence are aligned by timestamps to construct a training sample matrix. A long short-term memory network is then used to extract the temporal features of the sample data. The probability density estimator is trained based on the extracted time-series features. The probability distribution function of wind power output is obtained through the kernel density estimation method, and the probability density curve of output uncertainty is obtained, i.e., the wind power output probability function.

[0106] It should be noted that the analysis of signal attenuation characteristics and wind power output uncertainty relies on a virtual reality platform. Immersive scenarios visually present signal attenuation trends and power output fluctuation patterns, facilitating real-time monitoring of the power grid's operational status in complex environments by maintenance personnel. This invention generates high-precision path loss and power output distribution characteristics through multi-dimensional data processing and probabilistic modeling, providing a reliable basis for subsequent optimized scheduling. The specific implementation can be flexibly adjusted according to actual scenarios; this embodiment does not impose excessive limitations on specific algorithm details.

[0107] It should also be noted that the instantaneous signal data sequences collected through the wind farm sensor network include Rayleigh fading characteristics caused by fog and terrain obstruction. Median filtering was used to preprocess the signal sequences, with the filter window size set to five times the sampling period to remove outliers caused by equipment vibration or electromagnetic interference; these outliers typically exceeded the normal signal amplitude by more than three times. After filtering, the signal fluctuation amplitude was reduced by approximately 40%, generating a stable signal amplitude time series. A signal amplitude observation matrix was constructed to provide basic data for subsequent analysis. The signal amplitude observation matrix was normalized, with 90% of the normalized amplitude values ​​falling within the range of 0.3 to 0.8. The Rayleigh distribution probability density function was fitted using the least squares method to extract the fading depth parameter. The shape parameter ranged from 1.2 to 1.8; a larger parameter value indicated more severe signal fading. Further, the signal amplitude variation with propagation distance was fitted using an exponential function. Based on test data from 20 measurement points spaced 50 meters apart, the signal attenuation coefficient was calculated to be between 0.02 and 0.05, with a characteristic distance parameter of 300 to 500 meters, reflecting the propagation distance required for the signal strength to attenuate to its initial value of 1 / e. Based on the particle size distribution of the fog particles (2 to 8 micrometers), particle number density (500 to 2000 particles per cubic centimeter), and dielectric constant (real part 5 to 7), the scattering coefficient was calculated using Mie scattering theory, ranging from 0.01 to 0.03. Combined with the distance attenuation function, an electromagnetic wave energy loss matrix was constructed, generating a scattering loss distribution function. Based on the electromagnetic wave energy loss matrix, the total loss probability distribution along the transmission path was calculated using a Bayesian probability estimation method. The mean ranged from 35 to 50 dB, and the standard deviation ranged from 8 to 12 dB. The distribution curve exhibited a right-skewed characteristic, reflecting the combined impact of fog and terrain on signal propagation. The path loss probability distribution was randomly sampled 10,000 times using the Monte Carlo method. Combined with the characteristics of the wind turbine power curve, the output power distribution under different loss conditions was calculated. Kernel density estimation was used to generate the wind power output probability density function, which exhibited a bimodal characteristic. The main peak corresponded to 75% to 85% of the rated power, and the secondary peak corresponded to 45% to 55%, reflecting the significant fluctuations in output power under fog conditions. The standard deviation of the distribution increased by approximately 30%, and the 95% confidence interval width was approximately 50% of the rated power, providing a precise basis for subsequent scheduling optimization. Wind turbine output data and meteorological environmental parameters, including wind speed and air pressure sequences, were collected through wind farm monitoring devices. To address data missingness, with random missing values ​​accounting for 3% to 5% and continuous missing values ​​lasting 10 to 30 minutes, linear interpolation was used to fill in the missing values, with the interpolation error controlled within 5%. Median filtering was used to handle abnormal fluctuations, with a filtering window of 5 data points, reducing the standard deviation by approximately 40%, generating a preprocessed output time series to provide stable data for subsequent feature extraction. The preprocessed output time series was then decomposed into three levels using the db4 wavelet basis function to separate high-frequency rapid fluctuations and low-frequency trend components. A soft thresholding method was applied to process the high-frequency coefficients, with the threshold set at 1.5 times the standard deviation, reconstructing a smooth output curve.A fixed 10-minute time window with 50% overlap was used for segmented processing. The frequency of power output fluctuations was statistically analyzed using peak and trough data points within the window. The time intervals followed a log-normal distribution with a mean between 15 and 25 minutes. Wind speed and air pressure data were collected from meteorological stations. Fourier transform was used to extract the daily 24-hour wind speed cycle (amplitude 2-4 m / s) and the 365-day seasonal cycle (amplitude 3-6 m / s). Air pressure sequences were analyzed using trend decomposition to extract a linear trend rate of change of 0.1-0.3 hPa per hour, a fluctuation period of 12-24 hours, and an amplitude of 2-5 hPa. The power output fluctuation frequency features were aligned with the environmental parameter sequences by timestamp to construct a training sample matrix with a one-year time span and a 10-minute sampling interval. A Long Short-Term Memory (LSTM) network was used to process the training sample matrix. The network contained a three-layer structure, inputting 24-hour historical data to extract the temporal correlation features between power output fluctuations and environmental parameters. By training a probability density estimator using kernel density estimation, a wind power output probability density function is generated, exhibiting a bimodal characteristic. The primary peak corresponds to 70% to 80% of the rated power, and the secondary peak corresponds to 30% to 40%. The 95% confidence interval is approximately 50% of the rated power. This invention significantly improves output prediction accuracy and reduces the impact of uncertainties in foggy environments through multi-dimensional feature extraction and probabilistic modeling, providing reliable support for optimizing energy storage scheduling and topology configuration.

[0108] In this embodiment of the invention, step S300 includes the following sub-steps B1-B4;

[0109] In B1: The Monte Carlo method is used to randomly sample the wind power output probability function to obtain the output deviation prediction data set;

[0110] In B2: Determine the fluctuation amplitude threshold based on the power deviation prediction data set, divide the power fluctuation scenarios, and generate the probability value of scenario occurrence;

[0111] In B3: An electromagnetic wave propagation path map is established based on the forest cover density map and topographic elevation data, and signal transmission loss distribution data is calculated;

[0112] In B4: Data processing is performed on the signal transmission loss distribution data to reconstruct and generate signal fluctuation characteristic sequence data.

[0113] Specifically, the power output fluctuation range is obtained based on the wind power output probability distribution function. The Monte Carlo method is used to randomly sample the probability distribution, generating multiple sets of power output deviation prediction data. The mean and standard deviation of each set of data are used to obtain the fluctuation range characteristic values. The power output deviation prediction data is normalized, and the fluctuation amplitude threshold is determined using the quantile method. Based on the threshold, power output fluctuation scenarios are divided, and the occurrence probability and duration of each scenario are statistically obtained. A forest cover density map is used to obtain the spatial distribution of the obstructed area. Combined with terrain elevation data obtained from lidar, an electromagnetic wave propagation path map is established to obtain the signal transmission loss distribution under each scenario. Wireless signal data within the obstructed area is preprocessed. Outliers are removed by median filtering, and missing data points are filled using linear interpolation to obtain a continuous signal strength sequence. The Haar wavelet basis function is selected to perform a four-level decomposition on the preprocessed signal sequence. High-frequency coefficients are denoised using a soft thresholding method to reconstruct the signal fluctuation feature sequence. An adaptive segmentation method is used to divide the signal fluctuation feature sequence, obtaining the mean amplitude, median frequency, and duration of each signal segment, and constructing the probability density function of the signal fluctuation features.

[0114] It should be noted that the analysis of the wind power output probability distribution function, combined with virtual reality technology, uses an immersive scenario to demonstrate output fluctuations and signal transmission characteristics, helping operation and maintenance personnel to intuitively understand the grid operation status in complex environments. This invention uses methods such as Monte Carlo sampling, quantile analysis, and wavelet transform to accurately quantify output deviations and signal fluctuation characteristics, and combines terrain and vegetation data to generate propagation path models, improving the targeting and reliability of scheduling optimization.

[0115] It should also be noted that, based on the wind power output probability density function, the Monte Carlo method was used to conduct 10,000 random samplings, generating multiple sets of output deviation prediction data, covering various scenarios such as normal operation and reduced output operation. The mean and standard deviation of each set of data were calculated to determine the output fluctuation range. The standard deviation reflects the severity of the fluctuation, with the main peak corresponding to 70% to 80% of the rated power and the secondary peak corresponding to 40% to 50%. Through normalization, 90% of the data fluctuation amplitudes were distributed between 0.2 and 0.8. Using the quantile method, 0.3 and 0.7 were set as thresholds to divide the output fluctuations into small, medium, and large fluctuation scenarios. Small fluctuations accounted for 60%, and their duration was usually within 30 minutes, providing a basis for subsequent scenario analysis.

[0116] Topographic elevation data of forest-covered areas were acquired using lidar scanning. The average tree height ranged from 15 to 25 meters, and the canopy density ranged from 0.8 to 1.2 trees per square meter. An electromagnetic wave propagation path model was constructed by combining this data with a canopy density map. Based on power fluctuation scenarios, the signal transmission loss distribution was calculated for each scenario. The loss value was significantly affected by tree height and density, with a typical loss range of 30 to 45 dB. The propagation path model quantified the attenuation effect of the obstruction on electromagnetic waves through numerical simulation, generating loss distribution data to support signal fluctuation analysis.

[0117] Wireless signal data from obstructed areas is preprocessed by using median filtering to remove outliers caused by equipment jitter or external interference. Outliers exceeding 2.5 times the normal value are filtered within a window of 5 data points, reducing the signal standard deviation by approximately 35%. For missing data, linear interpolation is used to complete the signal, with the error controlled within 5%, generating a continuous signal strength sequence. A four-level decomposition using Haar wavelet basis functions separates high-frequency abrupt changes and low-frequency trend components. A soft thresholding method is applied to process the high-frequency coefficients, with the threshold set at 1.5 times the coefficient standard deviation, reconstructing a signal fluctuation characteristic sequence that preserves rapid changes.

[0118] An adaptive segmentation method was used to divide the signal fluctuation characteristic sequence. Segmentation points were identified when the difference between the means of adjacent data points exceeded twice the standard deviation, with durations ranging from 5 to 20 minutes and amplitude variations ranging from 20% to 60% of the peak value. The mean amplitude, median frequency, and duration of each segment were calculated to construct a probability density function for the signal fluctuation characteristics. Analysis showed that the signal fluctuation frequency was higher during the day than at night, with a period of 15 to 40 minutes. The amplitude variation range increased by approximately 25% during the day, and the proportion of high-frequency components rose to 35%, reflecting the dynamic impact of fog and terrain on signal transmission stability, providing accurate data support for optimized scheduling.

[0119] In this embodiment of the invention, step S400 includes the following sub-steps C1-C7;

[0120] In C1: Set the upper and lower limits of the signal amplitude threshold based on the historical signal fluctuation statistics, calculate the mean and standard deviation of the real-time signal, and determine whether the signal fluctuation exceeds the upper and lower limits of the signal amplitude threshold.

[0121] In C2: For signal segments that exceed the upper and lower limits of the signal amplitude threshold, a channel fading model is established by combining the signal fluctuation distribution characteristics and fog path loss.

[0122] In C3: output the wireless signal propagation delay to determine the transmission time sequence of energy storage scheduling commands;

[0123] In C4: The transmission and reception timestamps of energy storage scheduling instructions are obtained based on the wireless signal transmission delay, and an instruction transmission delay time series is generated;

[0124] In C5: Data processing is performed on the instruction transmission delay time series to generate a smooth delay sequence;

[0125] In C6: Calculate the statistical parameters of the smoothed delayed sequence and generate the delay time probability distribution function;

[0126] In C7: Extract error parsing logs from the probability distribution function, analyze the correlation between error types, identify factors affecting latency, and generate potential latency intervals.

[0127] Specifically, based on historical signal fluctuation statistics, upper and lower limits for signal amplitude thresholds are set. The mean and standard deviation of the real-time signal are calculated using a sliding window method to determine whether the signal fluctuation exceeds the upper and lower limits of the signal amplitude thresholds, as shown below:

[0128] ;

[0129] in, The standard deviation of the wind power output deviation prediction data. This represents the total number of deviation prediction data. For the first One predicted value for output deviation, The mean of all predicted deviations. This is the data sequence number.

[0130] For signal segments exceeding the upper and lower limits of the signal amplitude threshold, wavelet transform is used to extract the fluctuation amplitude and frequency features. The fluctuation probability distribution function is calculated using maximum likelihood estimation, and is expressed as:

[0131] ;

[0132] in, Let be the probability density function of the signal fluctuation amplitude. For a moment The signal amplitude value, The mean of the signal within the sliding window. The standard deviation of the signal within the sliding window. It is an exponential function.

[0133] A channel state feature vector is constructed by combining the fluctuation probability distribution function and the Rayleigh distribution parameters. The channel attenuation coefficient is calculated using path loss data in a foggy environment, and is expressed as:

[0134] ;

[0135] in, This is the channel attenuation coefficient. The base path loss at the reference distance, For transmission distance, For reference distance, For the first The weight coefficients of each feature vector. For the first Each channel state feature component Let be the dimension of the feature vector.

[0136] A recurrent neural network is trained using channel state feature vectors. The input features include signal amplitude, frequency, and attenuation coefficient, and the output is a predicted channel state sequence. The propagation delay of the wireless signal is calculated based on the predicted channel state sequence, and a delay distribution function is constructed using a probability density estimation method, expressed as:

[0137] ;

[0138] in, Let be the cumulative distribution function of transmission delay. This is the upper limit of latency. The integral variable represents the specific time delay value. Let be the mean parameter of the time delay distribution. Let be the standard deviation parameter of the time delay distribution. It is an exponential function.

[0139] Energy storage control commands are prioritized according to response time requirements, and the transmission time intervals for commands of different priorities are obtained by combining the delay distribution function. The Monte Carlo method is used to randomly sample the transmission time intervals, and the delay sequence of command transmission is obtained through the cumulative distribution function.

[0140] It should be noted that the channel fading model is constructed using virtual reality technology, which displays signal fluctuations and transmission delay characteristics in real time through an immersive interface, helping maintenance personnel quickly assess the wireless communication status. This step improves energy storage scheduling efficiency by extracting signal fluctuation characteristics and predicting delay distribution, combined with command priority optimization of transmission strategies.

[0141] It should also be noted that, based on historical signal data statistics, the standard deviation of signal amplitude is within the range of 2 to 5 dB. The upper threshold is set to the mean plus 3 times the standard deviation, and the lower threshold is the mean minus 2.5 times the standard deviation. A 60-second sliding window is used to calculate the mean and standard deviation of the real-time signal to determine whether it exceeds the upper and lower limits of the signal amplitude threshold. Signal segments exceeding the threshold reflect abnormal fluctuations caused by fog or obstruction, triggering subsequent feature extraction and delay analysis to ensure timely detection of communication anomalies. For signal segments exceeding the threshold, a three-level decomposition using the db4 wavelet basis function is performed to extract fluctuation amplitude and frequency features. Amplitude features are calculated through the wavelet coefficient energy distribution, with 90% of the energy concentrated in the first two levels. Frequency features are obtained through the zero-crossing rate of the reconstructed signal. Maximum likelihood estimation is used to calculate the fluctuation probability distribution function, reflecting the dynamic changes in signal strength and providing data support for channel modeling. Combining the fluctuation probability distribution function and Rayleigh distribution parameters, with shape parameters ranging from 1.2 to 1.8, a multi-dimensional channel state feature vector containing amplitude, frequency, and fog path loss is constructed. Path loss increases exponentially with distance, with an attenuation coefficient ranging from 0.2 to 0.5. A Long Short-Term Memory (LSTM) neural network was employed, receiving 300 seconds of historical features as input. The network consisted of two hidden layers with 64 neurons each. Training data covered various weather conditions throughout the year, achieving a channel state prediction accuracy of over 85% and generating a state prediction sequence. Based on this sequence, a delay distribution function was constructed using probability density estimation. Under normal conditions, the delay followed a normal distribution with a mean of 20 to 50 milliseconds. During channel degradation, the delay exhibited a right-skewed distribution with a mean increasing to 80 to 150 milliseconds. Energy storage commands were categorized into high, medium, and low priorities based on response time, requiring delays of less than 100 milliseconds, 200 milliseconds, and 500 milliseconds, respectively. A Monte Carlo method was used for 10,000 samplings to simulate transmission under different conditions. High-priority commands achieved a delay of less than 80 milliseconds for 95% of cases, medium-priority commands 150 milliseconds, and low-priority commands 400 milliseconds. Combining the cumulative distribution function, the tail characteristics of the delay distribution were analyzed, and the command transmission strategy was optimized to ensure that high-priority commands maintained a relatively small delay increase (approximately 30%) even under adverse conditions, thus improving scheduling reliability.

[0142] Specifically, the sending and receiving timestamps of energy storage scheduling commands are collected, and the time base is calibrated through a time synchronization server to obtain the actual delay time series of command transmission. Outlier detection is performed on the delay time series, identifying abnormal delay points using the three-standard-deviation method, and smoothing outliers using median filtering. Statistical feature extraction is performed on the smoothed delay series, calculating the delay mean, variance, skewness, and kurtosis, and constructing the probability distribution function of the delay time as follows:

[0143] ;

[0144] in, For delay time The probability density of occurrence For delay time variables, This is the mean parameter for the delay time. The standard deviation parameter of the delay time. It is an exponential function.

[0145] The system reads instruction parsing logs from the receiving database, extracts parsing error codes and descriptions, establishes an error type dictionary, and counts the frequency of each error type. It then uses association rule mining to analyze the relationships between error types, sets minimum support and confidence thresholds, and extracts frequent error patterns. A feature vector is constructed by combining latency features and error pattern features, and a random forest classifier is used to identify key latency-influencing factors. Based on the weights of the identified latency-influencing factors, a kernel density estimation method is used to obtain the confidence interval of the latency distribution, thereby determining the potential latency range of energy storage scheduling instructions.

[0146] It should be noted that the immersive interface displays latency distribution and error patterns, helping operations and maintenance personnel monitor communication performance in real time. This step precisely quantifies command transmission latency and its influencing factors, improving the reliability and real-time performance of scheduling commands, and providing support for power grid operation optimization in complex environments.

[0147] It should also be noted that by collecting the timestamps of the transmission and reception of energy storage scheduling commands and relying on a network time protocol server for time calibration, the synchronization accuracy is better than 1 millisecond, and the actual delay time series is calculated. The delay series exhibits periodic characteristics, reflecting the dynamic changes in network communication. Anomalies are detected using the three-standard-deviation method. Outliers are usually caused by network latency or communication interruptions, exceeding the normal value by more than three times. The normal delay range is 50 to 150 milliseconds, while outliers can reach 500 milliseconds or higher. A 5-point median filter is used to smooth outliers. After filtering, the standard deviation of the series is reduced by about 30%, providing stable data for subsequent analysis. Statistical characteristics such as mean, variance, skewness, and kurtosis are calculated for the smoothed delay series. Measured data show that under good communication conditions, the mean delay is about 80 milliseconds, the standard deviation is less than 20 milliseconds, the skewness is close to 0, and the kurtosis is between 2.8 and 3.2, approximating a normal distribution. A kernel density estimation method was employed, using a Gaussian kernel function. Bandwidth was determined using the Silverman empirical formula to generate a delay probability distribution function, clearly reflecting the delay distribution characteristics and providing a foundation for confidence interval calculation. Instruction parsing logs were extracted from the receiver database to construct an error type dictionary, including data validation errors, format parsing errors, and timeout errors, accounting for approximately 25%, 15%, and 10%, respectively. Data validation errors were mainly caused by bit flips, format parsing errors were related to instruction version mismatch, and timeout errors reflected network congestion. The Apriori algorithm (association analysis algorithm) was used for association rule mining, with a minimum support of 0.1 and a minimum confidence of 0.6. The correlation probability between data validation errors and timeout errors reached 0.7, indicating that error types are coupled when network quality deteriorates, increasing the risk of instruction transmission failure. Delay statistical features and error patterns were combined into a 10-dimensional feature vector, including delay mean, standard deviation, and error probability, which was then input into a random forest classifier for training. The classifier uses 10-fold cross-validation to optimize parameters, achieving an accuracy of over 85%. Weight analysis shows that network latency and signal quality are the main latency factors. Combined with kernel density estimation results, the latency interval at a 95% confidence level is calculated, ranging from the mean to 2.5 standard deviations above and below. The upper limit of high-priority instruction latency is controlled within 200 milliseconds. Seasonal analysis indicates that the mean latency increases by approximately 30% in foggy winter weather, with a heavier distribution tail, requiring additional time margin to ensure the real-time performance of scheduling instructions.

[0148] In this embodiment of the invention, step S500 includes the following sub-steps D1-D4;

[0149] In D1: Obtain the timestamp information of the energy storage scheduling instruction sequence, extract the power regulation direction parameter and regulation amplitude parameter, and generate the instruction execution sequence;

[0150] In D2: A perturbation execution sequence is generated based on the instruction execution sequence;

[0151] In D3: Scan the disturbance execution sequence, extract the adjustment direction sequence, and determine the power conflict points of adjacent commands;

[0152] In D4: Based on the timing deviation and command conflict characteristics, output energy storage scheduling disorder assessment data.

[0153] Specifically, the energy storage dispatch command sequence is timestamped and sorted to extract the power adjustment direction and amplitude parameters of the commands, establishing a command execution sequence database. Based on the potential delay interval, a time sampling interval is set, and multiple sets of delay time sequences are generated using the Monte Carlo method. The execution order after disturbance is obtained through command time sequence rearrangement. An command comparison matrix is ​​established for the perturbed execution order, and the timing deviation of command execution is statistically obtained based on the time interval and power change characteristics between adjacent commands. A fixed-time window scanning method is used to detect power adjustment command conflicts, setting the window sliding step size and overlap rate to extract the adjustment direction sequence of commands within the window. A command conflict rule table is constructed based on the adjustment direction sequence, setting judgment thresholds for opposite directions and overlapping amplitudes to identify conflict points in power adjustment commands. Topology connection data is obtained from the microgrid controller, and the capacity parameters and power limits of energy storage units are extracted to establish energy storage resource allocation constraints. Support vector regression is used to obtain the allocation deviation of energy storage dispatch. Combining the timing deviation and command conflict characteristics, a multilayer perceptron outputs the dispatch disorder assessment result.

[0154] It should be noted that timestamp information is obtained from the energy storage dispatch command sequence with an accuracy better than 1 millisecond. Power adjustment direction and amplitude parameters are extracted, with the direction including charging and discharging, and the amplitude range covering 10% to 90% of the rated power. A command execution sequence database is constructed based on timestamp sorting, recording the execution time, adjustment target, and priority of each command to ensure data integrity for subsequent analysis and provide a basis for out-of-order assessment. Based on the potential delay range, the sampling interval is set to 1 / 10 of the minimum delay; for a typical range of 50 to 200 milliseconds, the sampling interval is 5 milliseconds. The Monte Carlo method is used to generate 10,000 sets of delay time series to simulate the impact of communication delay on the command execution order. Disturbanced execution sequences are generated through time series rearrangement. Statistical analysis shows that 90% of the time series deviations are between 50 and 150 milliseconds, with the maximum deviation not exceeding three times the original interval, reflecting the degree of disturbance to the execution order caused by delay. A fixed 60-second time window is used to scan the disturbed execution sequences, with a sliding step size of 30 seconds and an overlap rate of 50%, to extract the adjustment direction sequence within the window. A command comparison matrix is ​​constructed, with matrix elements representing the time interval and power change of adjacent commands. The number of directional changes should not exceed four; excessive changes indicate control oscillation. Conflict rules are set: adjacent commands with opposite directions and an interval of less than 20 seconds constitute a directional conflict; amplitude overlap exceeding 70% constitutes an amplitude conflict. Power conflict points are identified, and the timing deviation of command execution is quantified. Topology connection data is obtained from the microgrid controller to extract the capacity and power limits of energy storage units. Charging and discharging power is controlled within 85% of rated power, and the state of charge is maintained between 20% and 90%. Support vector regression is used, with a radial basis function kernel selected. Parameters are optimized through cross-validation to calculate energy storage allocation deviation, achieving a correlation coefficient above 0.75. Combining timing deviation and conflict characteristics, a 16-dimensional feature vector is constructed and input into a multilayer perceptron with two hidden layers (32 and 16 neurons respectively), outputting a scheduling disorder assessment value. Results show that 15% of commands exhibit disorder, and 3% are severely disordered. Especially during rapid load changes, the superposition of delays and frequent command switching exacerbates the multi-unit collaborative scheduling deviation, requiring optimization of the scheduling strategy to improve system stability.

[0155] In this embodiment of the invention, step S700 includes the following sub-steps E1-E6;

[0156] In E1: Based on the energy storage scheduling disorder assessment data, extract the power flow fluctuation characteristics caused by delay;

[0157] In E2: Based on historical wind power output data, determine whether the power flow fluctuation exceeds the preset range;

[0158] In E3: If the value exceeds the preset range, the energy storage unit's charging and discharging power data and response time data are acquired to generate an energy storage characteristic database;

[0159] In E4: Calculate the response time weighting coefficient based on the energy storage characteristic database and generate a response time series mapping table;

[0160] In E5: a piecewise linear programming method is used to process the response timing mapping table, set constraints, and generate power regulation commands;

[0161] In E6: Calculate power flow distribution based on power regulation commands, generate power flow distribution dataset, construct network status evaluation indicators, and generate energy storage and topology configuration optimization data.

[0162] Specifically, the charging and discharging power and response time data of energy storage units are collected to construct an energy storage characteristic database. Response time weighting coefficients are obtained using the analytic hierarchy process (AHP). Energy storage units are prioritized based on these weighting coefficients, and a response time sequence mapping table is established using the scheduling disorder assessment results to generate a new response order. For the adjusted response order, piecewise linear programming is used to solve the power allocation problem, setting upper and lower power limits and ramp rate constraints to obtain power adjustment commands for each time period. The energy storage power adjustment commands are used to obtain a new power flow distribution. Node voltage and branch power are solved using power flow equations to generate a power flow distribution dataset. Network status assessment indicators are constructed based on the power flow distribution data, including voltage deviation, network loss, and energy storage lifetime indicators, with weighting coefficients set for each indicator. A multi-objective optimization function is established based on the assessment indicators, setting variable constraints for energy storage capacity and topology. A non-dominated sorting genetic algorithm is used to solve the optimization problem. Pareto optimality is assessed on the optimization results, and the optimal solution set satisfying the constraints is selected to generate an energy storage capacity configuration scheme and a network topology scheme. The multi-objective optimization function is expressed as follows:

[0163] ;

[0164] in, To synthesize the objective function value, , , These are the weighting coefficients for voltage deviation, grid loss, and energy storage life, respectively. The total number of nodes. For the first The actual voltage value of each node For reference voltage value, The total number of branch roads, For the first Power loss of the branch circuit This represents the total number of energy storage devices. For the first Capacity decay of an energy storage device For the first The rated lifespan of each energy storage device.

[0165] It should be noted that the charging and discharging power and response time data of the energy storage units were collected. The response time of lithium batteries is 50 to 100 milliseconds, flywheel energy storage is less than 20 milliseconds, and supercapacitors are about 10 milliseconds. A database of energy storage characteristics, including power, response speed, and reliability, was constructed. The analytic hierarchy process (AHP) was used to calculate the response time weight coefficients, considering response speed, power fluctuation, and scheduling reliability, with weights of 0.5, 0.3, and 0.2, respectively. Priority ranking was generated through a judgment matrix. When the scheduling disorder level exceeds 0.6, the response order needs to be completely rearranged; when the disorder level is between 0.3 and 0.6, only some units are adjusted to form a response time sequence mapping table to optimize command execution efficiency. Based on the response time sequence mapping table, piecewise linear programming was used to solve the power allocation problem, setting the upper limit of power to 90% of the rated power, the lower limit to 15%, and the ramp rate to 30% per minute to ensure safe operation of the energy storage. After generating power regulation commands, the power flow distribution is calculated using the Newton-Raphson method, with a convergence accuracy of 0.0001 per unit, completing the calculation within 5 iterations. Node voltages are maintained between 0.95 and 1.05 per unit, and the network loss rate is below 3%. A power flow distribution dataset is generated, providing a foundation for network condition assessment. Based on the power flow distribution data, network condition assessment indicators are constructed, including voltage deviation, network loss, and energy storage lifetime, with weights of 0.4, 0.3, and 0.3, respectively, reflecting operational quality, economy, and maintenance costs. A multi-objective optimization function is established, setting energy storage capacity and topology as constraint variables. A non-dominated sorting genetic algorithm is used, with a population size of 200, 500 generations, a crossover probability of 0.8, and a mutation probability of 0.1, to select 10 to 20 Pareto optimal solutions. In the optimized scheme, fast-response energy storage accounts for 30% of the total capacity, medium-response energy storage accounts for 50%, and slow-response energy storage accounts for 20%. In areas with high load density, fast-response energy storage should be prioritized to improve dispatch flexibility, while in areas with stable load, slow-response energy storage should be configured to reduce maintenance costs, thereby optimizing grid operation efficiency.

[0166] The above is an illustrative scheme of a high-penetration renewable energy distribution network collaborative optimization method according to this embodiment. It should be noted that the technical solution of this high-penetration renewable energy distribution network collaborative optimization system and the technical solution of the aforementioned high-penetration renewable energy distribution network collaborative optimization method belong to the same concept. Details not described in detail in the technical solution of the high-penetration renewable energy distribution network collaborative optimization system in this embodiment can be found in the description of the technical solution of the aforementioned high-penetration renewable energy distribution network collaborative optimization method.

[0167] This embodiment of a high-penetration new energy distribution network collaborative optimization system includes:

[0168] The data processing module acquires wind farm operation data and signal strength, processes the wind farm operation data and signal strength, and obtains the signal attenuation trend.

[0169] The identification module fits the signal attenuation trend to obtain signal attenuation characteristics, identifies path loss in foggy environments based on the signal attenuation characteristics, and obtains the wind power output probability distribution function.

[0170] The calculation module obtains the power output fluctuation range based on the wind power output probability distribution function, predicts power output deviation scenarios, calculates the wireless signal fluctuation probability for each scenario, and extracts signal fluctuation distribution features.

[0171] The model building module establishes a channel fading model based on the signal fluctuation distribution characteristics and the path loss in the foggy environment, outputs the wireless signal propagation delay, obtains the transmission time series, and obtains the potential delay interval of the energy storage system scheduling command based on the transmission time series.

[0172] The simulation module simulates scheduling disorder and power regulation command conflicts within the potential delay interval, identifies the allocation deviation between microgrid topology configuration and energy storage scheduling, and generates scheduling disorder assessment data.

[0173] The judgment module extracts power flow fluctuations from the scheduling disorder assessment data and judges the power flow fluctuations by combining them with historical wind power output data.

[0174] The collaborative optimization module adjusts the response sequence of energy storage units based on the judgment results, generates a power regulation command sequence, calculates the adjusted power flow distribution data, and generates energy storage and topology configuration optimization data.

[0175] This embodiment also provides a computer device applicable to a high-penetration renewable energy distribution network collaborative optimization method, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the high-penetration renewable energy distribution network collaborative optimization method proposed in the above embodiment.

[0176] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a collaborative optimization method for high-penetration new energy distribution networks as proposed in the above embodiments.

[0177] The storage medium proposed in this embodiment and the method for collaborative optimization of a high-penetration new energy distribution network proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0178] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0179] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A collaborative optimization method for high-penetration renewable energy distribution networks, characterized in that, include: Acquire wind farm operation data and signal strength, perform data processing on the wind farm operation data and signal strength, and obtain the signal attenuation trend; The signal attenuation trend is fitted to obtain the signal attenuation characteristics. Based on the signal attenuation characteristics, the path loss in the fog environment is identified, and the wind power output probability distribution function is obtained. The power output fluctuation range is obtained based on the wind power output probability distribution function, and the power output deviation scenarios are predicted. The wireless signal fluctuation probability of each scenario is calculated, and the signal fluctuation distribution characteristics are extracted. A channel fading model is established based on the signal fluctuation distribution characteristics and the path loss in the foggy environment. The wireless signal propagation delay is output to obtain the transmission time series. Based on the transmission time series, the potential delay interval of the energy storage system scheduling command is obtained. Simulate scheduling disorder and power regulation command conflicts within the potential delay interval, identify the allocation deviation between microgrid topology configuration and energy storage scheduling, and generate scheduling disorder assessment data; Power flow fluctuations are extracted from the scheduling disorder assessment data and judged by combining them with historical wind power output data; Based on the judgment results, the response sequence of the energy storage units is adjusted, a power regulation command sequence is generated, the adjusted power flow distribution data is calculated, and energy storage and topology configuration optimization data is generated. The probability distribution function of wind power output includes: Obtain the signal sequence of the wind farm, preprocess the signal sequence, and establish a signal amplitude observation matrix; The signal amplitude observation matrix is ​​normalized, and the signal amplitude probability distribution function is generated by fitting the least squares method. The scattering coefficient is calculated based on the signal amplitude probability distribution function, and the electromagnetic wave energy loss matrix is ​​generated by combining the distance attenuation parameter. Establish path loss mapping relationship based on electromagnetic wave energy loss matrix, calculate transmission path loss probability distribution, and generate wind power output probability function.

2. The high-penetration new energy distribution network collaborative optimization method as described in claim 1, characterized in that, The probability distribution function for wind power output also includes: Obtain the actual power output record of the wind farm, perform wavelet decomposition on the actual power output record of the wind farm, and obtain a smooth power output curve; The smooth output curve is segmented according to a fixed time window, and the peak point set and valley point set within the time window are calculated to obtain the output fluctuation frequency characteristics. A training sample matrix is ​​generated based on the power output fluctuation frequency characteristics and environmental parameter sequences. Time series features are extracted to obtain the wind power output probability function.

3. The high-penetration new energy distribution network collaborative optimization method as described in claim 1, characterized in that, Extracting signal fluctuation distribution features includes: The Monte Carlo method was used to randomly sample the wind power output probability function to obtain a set of output deviation prediction data. Based on the output deviation prediction data set, determine the fluctuation amplitude threshold, divide the output fluctuation scenarios, and generate the probability value of scenario occurrence. An electromagnetic wave propagation path map was established based on forest cover density map and topographic elevation data, and signal transmission loss distribution data was calculated. Data processing is performed on the signal transmission loss distribution data to reconstruct and generate signal fluctuation characteristic sequence data.

4. The high-penetration new energy distribution network collaborative optimization method as described in claim 3, characterized in that, The potential delay range for energy storage system dispatch commands includes: Set upper and lower limits for signal amplitude thresholds based on historical signal fluctuation statistics, calculate the mean and standard deviation of real-time signals, and determine whether signal fluctuations exceed the upper and lower limits for signal amplitude thresholds. For signal segments that exceed the upper and lower limits of the signal amplitude threshold, a channel fading model is established by combining the signal fluctuation distribution characteristics and fog path loss. Output the wireless signal propagation delay to determine the transmission time sequence of energy storage scheduling commands; The transmission and reception timestamps of energy storage scheduling commands are obtained based on the wireless signal transmission delay, and a command transmission delay time series is generated. Data processing is performed on the instruction transmission delay time series to generate a smooth delay sequence; Calculate the statistical parameters of the smoothed delayed sequence and generate the probability distribution function of the delay time; Error parsing logs are extracted from probability distribution functions, error type correlations are analyzed, delay influencing factors are identified, and potential delay intervals are generated.

5. The high-penetration new energy distribution network collaborative optimization method as described in claim 4, characterized in that, The scheduling disorder assessment data includes: Obtain the timestamp information of the energy storage dispatch instruction sequence, extract the power regulation direction parameter and regulation amplitude parameter, and generate the instruction execution sequence; Generate a perturbation execution sequence based on the instruction execution sequence; The disturbance execution sequence is scanned to extract the adjustment direction sequence and determine the power conflict points of adjacent commands; Based on the timing deviation and command conflict characteristics, output energy storage scheduling disorder assessment data.

6. The high-penetration new energy distribution network collaborative optimization method as described in claim 5, characterized in that, Energy storage and topology configuration optimization data include: Based on the energy storage dispatch disorder assessment data, the characteristics of power flow fluctuations caused by delays are extracted; By combining historical wind power output data, it can be determined whether the power flow fluctuations exceed the preset range; If the value exceeds the preset range, the energy storage unit's charging and discharging power data and response time data are acquired to generate an energy storage characteristic database. Calculate the response time weighting coefficient based on the energy storage characteristic database, and generate a response time series mapping table; A piecewise linear programming method is used to process the response time-series mapping table, set constraints, and generate power regulation commands. Calculate power flow distribution based on power regulation commands, generate power flow distribution dataset, construct network status evaluation indicators, and generate energy storage and topology configuration optimization data.

7. A high-penetration renewable energy distribution network collaborative optimization system, employing the high-penetration renewable energy distribution network collaborative optimization method as described in any one of claims 1 to 6, characterized in that, include: The data processing module acquires wind farm operation data and signal strength, processes the wind farm operation data and signal strength, and obtains the signal attenuation trend. The identification module fits the signal attenuation trend to obtain signal attenuation characteristics, identifies path loss in foggy environments based on the signal attenuation characteristics, and obtains the wind power output probability distribution function. The calculation module obtains the power output fluctuation range based on the wind power output probability distribution function, predicts power output deviation scenarios, calculates the wireless signal fluctuation probability for each scenario, and extracts signal fluctuation distribution features. The model building module establishes a channel fading model based on the signal fluctuation distribution characteristics and the path loss in the foggy environment, outputs the wireless signal propagation delay, obtains the transmission time series, and obtains the potential delay interval of the energy storage system scheduling command based on the transmission time series. The simulation module simulates scheduling disorder and power regulation command conflicts within the potential delay interval, identifies the allocation deviation between microgrid topology configuration and energy storage scheduling, and generates scheduling disorder assessment data. The judgment module extracts power flow fluctuations from the scheduling disorder assessment data and judges the power flow fluctuations by combining them with historical wind power output data. The collaborative optimization module adjusts the response sequence of energy storage units based on the judgment results, generates a power regulation command sequence, calculates the adjusted power flow distribution data, and generates energy storage and topology configuration optimization data.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the collaborative optimization method for a high-penetration new energy distribution network as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the collaborative optimization method for a high-penetration new energy distribution network as described in any one of claims 1 to 6.

Images