Wind power generation energy storage load intelligent prediction and power distribution management method

Abstract

The invention discloses a wind power generation energy storage load intelligent prediction and power distribution management method, and relates to the technical field of new energy power generation, and the method comprises the steps: generating a clock synchronization signal when a phase gradient quantity exceeds a stable threshold value, correcting the node voltage phase deviation of a power distribution network according to the clock synchronization signal, and outputting a whole network synchronization voltage waveform; inputting an energy storage charging and discharging control instruction into the power flow optimization model, and generating a voltage suppression control vector through a dynamic power deviation compensation algorithm; and extracting stability parameters in the wind power operation data, carrying out weight distribution and state matching on the energy storage charging and discharging control instruction and the voltage suppression control vector, and outputting a cooperative control instruction. According to the method, heterogeneous data such as wind speed, power and voltage are fused into multi-dimensional sequence parameters through a space-time dynamic coupling method, the phase gradient quantity is generated through phase field gradient extraction, a quantitative correlation model of wind speed fluctuation and power grid response is established, and the load prediction accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of new energy power generation technology, and in particular to a method for intelligent prediction and distribution management of wind power generation energy storage load. Background Technology

[0002] In recent years, with the increasing penetration of renewable energy in the power system, the volatility and intermittency of wind power generation have posed a severe challenge to the stable operation of the power grid. To address this issue, existing technologies mainly employ hybrid prediction methods based on physical models and data-driven approaches, combined with the rapid response capabilities of energy storage systems to achieve power balance. In the field of energy storage control, model predictive control and reinforcement learning algorithms are widely used in charging and discharging strategy optimization. Regarding grid synchronization control, traditional methods primarily rely on phase-locked loop (PLL) technology and droop control to achieve frequency and voltage stability by adjusting the inverter output. However, these methods still have significant limitations when dealing with the complex system dynamics under high-proportion renewable energy integration.

[0003] The shortcomings of existing technologies are mainly reflected in system coordinated control. First, traditional methods typically treat forecasting, energy storage control, and grid regulation as independent modules, lacking a unified dynamic coordination mechanism. Second, existing voltage regulation strategies are mostly based on local measurement information, lacking overall perception and coordinated optimization of the entire grid's status. These problems severely restrict the stable operation and absorption capacity of grids with a high proportion of renewable energy. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for intelligent prediction and distribution management of wind power generation and energy storage load to solve the problem of insufficient dynamic coupling of various modules in wind power and energy storage collaborative control.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for intelligent prediction and distribution management of wind power generation and energy storage loads, comprising,

[0008] Extract multidimensional sequence parameters from wind power operation data, perform phase field analysis on the multidimensional sequence parameters, and output the phase gradient.

[0009] When the phase gradient exceeds the stability threshold, a clock synchronization signal is generated, and the voltage phase offset of the distribution network nodes is corrected according to the clock synchronization signal, and the synchronous voltage waveform of the whole network is output.

[0010] The evolution direction of the phase field of the time-space crystal is analyzed on the synchronous voltage waveform of the entire network to generate energy storage charging and discharging control commands;

[0011] The energy storage charging and discharging control command is input into the power flow optimization model, and a voltage suppression control vector is generated through the dynamic power deviation compensation algorithm.

[0012] Stability parameters are extracted from wind power operation data, and weights are assigned and states are matched between energy storage charging and discharging control commands and voltage suppression control vectors to output coordinated control commands.

[0013] As a preferred embodiment of the wind power generation and energy storage load intelligent prediction and distribution management method of the present invention, the wind power operation data includes wind speed and wind direction measurements, generator output power, instantaneous bus voltage, original grid frequency signal, real-time SOC value of energy storage system, and temperature parameters.

[0014] As a preferred embodiment of the intelligent prediction and distribution management method for wind power generation and energy storage loads described in this invention, the specific steps for outputting the phase gradient are as follows:

[0015] Spatiotemporal dynamic coupling and energy state fusion of wind power operation data are performed to generate multidimensional sequence parameters;

[0016] The spatiotemporal evolution dynamics of multidimensional order parameters are analyzed and the phase field gradient is extracted to generate phase gradient quantities.

[0017] As a preferred embodiment of the intelligent prediction and distribution management method for wind power generation and energy storage loads described in this invention, the specific steps for outputting the network-wide synchronized voltage waveform are as follows:

[0018] When the phase gradient exceeds the stability threshold, the magnetic domain spin oscillation characteristics are simulated by the dynamic resonance algorithm to generate a clock synchronization signal.

[0019] Based on the clock synchronization signal, the voltage phase of each node in the distribution network is calibrated, and the synchronized voltage waveform of the entire network is output.

[0020] As a preferred embodiment of the intelligent prediction and distribution management method for wind power generation and energy storage loads described in this invention, the specific steps for analyzing the spatiotemporal crystal phase field evolution direction of the entire network synchronous voltage waveform are as follows:

[0021] The synchronous voltage waveform of the entire network is compared with the preset reference distribution network voltage in real time, and the instantaneous phase difference is output.

[0022] Based on the instantaneous phase difference, the energy flow trend is obtained through the evolution direction of the phase field of the spatiotemporal crystal, and combined with wind power operation data for collaborative decision-making to generate charging and discharging demand markers.

[0023] As a preferred embodiment of the intelligent prediction and distribution management method for wind power generation and energy storage loads described in this invention, the specific steps for generating energy storage charging and discharging control commands are as follows:

[0024] Based on the charging and discharging demand markers, power weight allocation and energy storage safety boundary constraint calculations are performed in conjunction with the instantaneous phase difference to generate charging and discharging power command values;

[0025] The charging and discharging power command values ​​are dynamically calibrated in direction and subject to multi-level safety constraints to generate energy storage charging and discharging control commands.

[0026] As a preferred embodiment of the intelligent prediction and distribution management method for wind power generation and energy storage load described in this invention, the specific steps of inputting energy storage charging and discharging control commands into the power flow optimization model and generating a voltage suppression control vector through a dynamic power deviation compensation algorithm are as follows.

[0027] A power flow optimization model is built based on a data input layer, an optimization calculation layer, and a vector generation layer.

[0028] The data input layer normalizes and aligns the energy storage charging and discharging control commands with the grid-wide synchronous voltage waveform in a spatiotemporal manner to generate a power flow optimization input matrix.

[0029] The optimization computation layer performs second-order cone programming to solve the power flow optimization input matrix, generating a set of node voltage corrections.

[0030] The vector generation layer performs virtual magnetic monopole field strength mapping on the set of node voltage corrections to generate voltage suppression control vectors.

[0031] As a preferred embodiment of the intelligent prediction and distribution management method for wind power generation and energy storage load described in this invention, the following steps are taken: Stability parameters are extracted from wind power operation data, and weights are allocated and state-matched between energy storage charging / discharging control commands and voltage suppression control vectors to output coordinated control commands.

[0032] Multi-scale principal component analysis was performed on wind power operation data to extract stability parameters;

[0033] Based on stability parameters, weight allocation and state matching are performed on energy storage charging and discharging control commands and voltage suppression control vectors, and a collaborative optimization control command set is obtained through a multi-objective dynamic fusion algorithm.

[0034] The collaborative optimization control instruction set is dynamically limited and priority is arbitrated, and then collaborative control instructions are output.

[0035] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the wind power generation energy storage load intelligent prediction and distribution management method as described in the first aspect of the present invention.

[0036] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the wind power generation energy storage load intelligent prediction and distribution management method as described in the first aspect of the present invention.

[0037] The beneficial effects of this invention are as follows: By using a spatiotemporal dynamic coupling method, heterogeneous data such as wind speed, power, and voltage are fused into multidimensional sequence parameters. Then, phase gradient quantities are generated through phase field gradient extraction, establishing a quantitative correlation model between wind speed fluctuations and grid response, thus improving the accuracy of load forecasting. The spatiotemporal crystal phase field evolution control technology unifies the node voltage phase through clock synchronization signals and generates energy storage commands by combining the spatiotemporal crystal phase field evolution direction, thereby reducing the voltage phase consistency error across the entire network and improving the time alignment accuracy between energy storage response and grid demand. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the 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.

[0039] Figure 1 This is a flowchart of a method for intelligent prediction and distribution management of wind power generation and energy storage loads.

[0040] Figure 2 A flowchart for generating multidimensional order parameters.

[0041] Figure 3 This is a flowchart for generating the synchronous voltage waveform for the entire network.

[0042] Figure 4 A flowchart for generating coordinated control commands. Detailed Implementation

[0043] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0044] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0046] Reference Figures 1-4 This is one embodiment of the present invention, which provides a method for intelligent prediction and distribution management of wind power generation energy storage load, including the following steps:

[0047] S1. Extract multidimensional sequence parameters from wind power operation data, perform phase field analysis on the multidimensional sequence parameters, and output the phase gradient.

[0048] S1.1 Wind power operation data includes wind speed and direction measurements, generator output power, instantaneous bus voltage, raw grid frequency signal, real-time SOC value and temperature parameters of energy storage system;

[0049] It should be noted that wind speed and direction measurements are obtained using an ultrasonic anemometer installed on the wind turbine; generator output power is measured by a DC-side power sensor on the converter; instantaneous bus voltage values ​​are acquired using a standard synchronous sampling protocol; the original grid frequency signal is obtained from the voltage signal through discrete Fourier transform; the real-time SOC value of the energy storage system is obtained by the BMS using the open-circuit voltage method, which is based on the terminal voltage measured by the battery in a static state and the preset voltage-SOC correspondence; and temperature parameters are obtained by temperature sensors embedded inside the energy storage battery pack.

[0050] S1.2. Perform spatiotemporal dynamic coupling and energy state fusion on wind power operation data to generate multidimensional sequence parameters;

[0051] It should be noted that the measured wind speed and direction values, generator output power, instantaneous bus voltage, raw grid frequency signal, real-time SOC value of the energy storage system, and temperature parameters are time-aligned to ensure that the timestamps of all data are synchronized. A sliding window analysis is used to analyze the spatial correlation between the measured wind speed and direction values ​​and the generator output power, establishing a power-wind speed correlation matrix. The dominant oscillation modes of the instantaneous bus voltage value and the raw grid frequency signal are extracted using an orthogonal decomposition method, constructing a voltage-frequency feature vector. Combining the energy state index of the real-time SOC value and temperature parameters of the energy storage system, a tensor product operation is performed on the power-wind speed correlation matrix and the voltage-frequency feature vector to generate a multidimensional order parameter that simultaneously contains spatiotemporal characteristics.

[0052] S1.3 Perform spatiotemporal evolution dynamic analysis and phase field gradient extraction on the multidimensional sequence parameters to generate phase gradient quantities.

[0053] It should be noted that a three-dimensional spatial coordinate system containing all monitoring points is established to record the spatial coordinates and measured values ​​of multi-dimensional sequence parameters such as wind speed and direction measurements, generator output power, instantaneous bus voltage, original grid frequency signal, real-time SOC value of energy storage system, and temperature parameters. Then, a difference method is used to analyze the changes in multi-dimensional sequence parameters between adjacent monitoring points, determining the spatial rate of change components of each monitoring point along the three coordinate axes. Simultaneously, the time series of multi-dimensional sequence parameters at each monitoring point is differentiated to obtain the time rate of change. Next, the spatial rate of change components are synthesized into a spatial gradient vector. The direction of the spatial gradient vector indicates the spatial growth direction of the sequence parameter, and the length of the spatial gradient vector reflects the intensity of spatial change. Then, the spatial gradient vector is coupled with the time rate of change. When the direction of the spatial gradient is consistent with the time change trend, the phase field gradient is enhanced; when the direction is opposite, the phase field gradient is weakened. Finally, the coupled phase field gradient is orthogonally decomposed to extract the feature vectors corresponding to the dominant modes, generating a phase gradient quantity containing amplitude and direction features. The amplitude of the phase gradient quantity reflects the intensity of energy fluctuations, and the direction of the phase gradient quantity indicates the energy transmission trend.

[0054] S2. When the phase gradient exceeds the stability threshold, a clock synchronization signal is generated, and the voltage phase offset of the distribution network nodes is corrected according to the clock synchronization signal, and the synchronous voltage waveform of the whole network is output.

[0055] S2.1 When the phase gradient exceeds the stability threshold, the magnetic domain spin oscillation characteristics are simulated through a dynamic resonance algorithm to generate a clock synchronization signal.

[0056] It should be noted that the phase gradient is compared with a preset stability threshold in real time. When the amplitude of the phase gradient continuously exceeds the stability threshold, the fluctuation characteristics of the phase gradient are pattern matched with a typical spin oscillation waveform to identify the dominant oscillation frequency. Then, the oscillation waveform parameters are adjusted according to the amplitude of the phase gradient to maintain a fixed proportional relationship between the oscillation period and the rate of change of the phase gradient. Next, the zero-crossing point of the oscillation waveform is extracted as a time reference to generate an equally spaced pulse sequence. Finally, the duty cycle of the pulse sequence is adjusted and the edges are sharpened to output a clock synchronization signal that meets the synchronization requirements of the power distribution network.

[0057] It should be noted that the stability threshold is set based on the statistical distribution characteristics of the phase gradient in historical wind power operation data, with an example value range of (0.15, 0.35).

[0058] S2.2. Based on the clock synchronization signal, calibrate the voltage phase of each node in the distribution network and output the synchronized voltage waveform of the entire network.

[0059] It should be noted that the rising edge of the clock synchronization signal is used as the reference trigger point. A phase detector compares the time difference between the zero-crossing point of the voltage waveform at each distribution network node and the reference point, outputting a phase error signal containing high-frequency jitter and low-frequency deviation. Subsequently, a proportional-integral (PI) structure is used to process the phase error signal. The proportional element responds quickly to phase abrupt changes, while the integral element eliminates steady-state phase errors. The response speed and stability are balanced by adjusting the proportional coefficient and the integral time constant. Next, the processed smoothed control voltage signal is input to the voltage-controlled oscillator (VCO). The amplitude of the control voltage signal is proportional to the integral amount of the phase error, and the rate of change is proportional to the instantaneous value of the phase error, driving the VCO to gradually correct the output frequency. When the phase error approaches zero, the VCO outputs a sinusoidal signal that is strictly synchronized with the clock synchronization signal. This sinusoidal signal serves as the phase reference, and the inverter output voltage phase is adjusted by a PWM modulator, ultimately achieving phase calibration of the voltage waveforms at each node of the distribution network, forming a synchronized voltage waveform for the entire network.

[0060] S3. Analyze the evolution direction of the spatiotemporal crystal phase field of the full network synchronous voltage waveform and generate energy storage charging and discharging control commands.

[0061] S3.1. Compare the synchronous voltage waveform of the entire network with the preset reference distribution network voltage in real time and output the instantaneous phase difference;

[0062] It should be noted that the zero-crossing characteristics of the synchronous voltage waveform of the entire network and the preset reference distribution network voltage waveform are collected simultaneously, and the time difference between the zero-crossing points of the two waveforms is recorded by a time-to-digital converter. Then, the ratio of the time difference to the power grid cycle is used to reflect the phase angle offset. Next, the time difference record is updated in each power grid cycle to form a continuous phase offset sequence. Finally, the phase offset obtained in the current power grid cycle is output as the instantaneous phase difference.

[0063] It should be noted that the specific process for setting the reference distribution network voltage is as follows: Based on power grid operation specifications, the amplitude of the standard sinusoidal waveform is set according to international standard voltage levels, and the frequency strictly adheres to the power frequency standard. The reference phase angle is based on the clock signal of the main substation's synchronous measurement unit, and periodic phase calibration is performed according to synchronous operation requirements. The voltage waveform characteristics conform to power quality standards, and its harmonic content meets industry specifications. The preset value of the reference distribution network voltage is verified through power system simulation to ensure that the voltage deviation at each node conforms to power system operation specifications under typical load conditions.

[0064] S3.2 Based on the instantaneous phase difference, the energy flow trend is obtained through the evolution direction of the phase field of the spatiotemporal crystal, and combined with wind power operation data for collaborative decision-making to generate charging and discharging demand markers;

[0065] It should be noted that the instantaneous phase differences of each node in the distribution network are arranged according to the spatial topology to form a discrete phase distribution matrix. The phase change rate between adjacent nodes is calculated using the central difference method to obtain a gradient vector field characterizing energy flow. The direction of the gradient vector indicates the dominant direction of energy flow, and the magnitude reflects the flow intensity. The generator output power change rate is extracted from the wind power operation data, and the correlation between the power change trend and the gradient vector direction is analyzed through vector dot product operation: when the dot product result exceeds the energy flow direction consistency threshold, it is determined to be an energy surplus state, and a negative charging and discharging demand mark is generated; when the dot product result is lower than the energy flow direction consistency threshold, it is determined to be an energy shortage state, and a positive charging and discharging demand mark is generated. The amplitude of the charging and discharging demand mark is determined by the product of the gradient vector magnitude and the power change rate, ensuring that the mark intensity matches the actual energy imbalance degree.

[0066] It should be noted that the energy flow direction consistency threshold is set based on the statistical correlation analysis of phase gradient and power change rate in historical wind power operation data, with an example value range of (0.15, 0.35).

[0067] S3.3. Based on the charging and discharging demand markings and combined with the instantaneous phase difference, power weight allocation and energy storage safety boundary constraint calculations are performed to generate charging and discharging power command values, as shown in the following expression.

[0068] P=sgn(D)·min(k·|Δθ|·|S1-S2|,P0·f);

[0069] Where P is the charging / discharging power command value, D is the logic signal for charging or discharging, k is the dynamic proportional coefficient with a value range of (0.3, 0.8), Δθ is the instantaneous phase difference, S1 is the real-time SCO value, S2 is the maximum allowable SCO value for energy storage, P0 is the rated power of the energy storage converter, f is the temperature attenuation coefficient with a value range of (0.8, 1.0), and sgn is the sign function.

[0070] It should be noted that the range of values ​​for the dynamic proportional coefficient k is obtained based on statistical analysis of historical operating data. The boundary value is determined by fitting the correlation curve of phase difference and power regulation under different operating conditions. The degree attenuation coefficient f is obtained by experimental calibration and fitting after testing the battery charge and discharge efficiency in the temperature range of -20℃ to 50℃.

[0071] It should be noted that the power regulation direction is determined based on the polarity of the charge / discharge demand marking, and the regulation intensity is determined based on the instantaneous phase difference amplitude. Then, the power regulation intensity is adjusted by referring to the dynamic proportional relationship determined by the real-time SOC value and temperature parameters of the energy storage system. Next, the adjusted power regulation intensity is compared with the rated power of the energy storage converter at a safety boundary, and the power value that meets the constraints is selected as the basic command. Finally, the basic command is adaptively modified based on the temperature decay relationship to ensure that the power command value is always within the allowable operating range of the energy storage device.

[0072] It should be noted that the process for determining the safety boundaries and constraints is as follows: The real-time SOC value of the energy storage system must be within the specified allowable charging and discharging range. During charging, the upper limit of SOC is the maximum allowable SOC value minus the safety margin; during discharging, the lower limit of SOC is the minimum allowable SOC value plus the safety margin. The absolute value of the power command must not exceed the product of the rated power of the energy storage converter and the short-term overload capacity coefficient. Rated power is attenuated based on real-time temperature to ensure that the upper limit of power equals the rated power of the energy storage converter multiplied by the temperature attenuation coefficient. Constraints include that the power command value must simultaneously not exceed the product of the rated power of the energy storage converter and the temperature attenuation coefficient, and not exceed the ratio of the energy storage capacity to the time constant. When the real-time SOC value of the energy storage system approaches the boundary, a ramp function is used to gradually reduce the power limit value.

[0073] S3.4 Perform dynamic direction calibration and multi-level safety constraints on the charging and discharging power command values ​​to generate energy storage charging and discharging control commands.

[0074] It should be noted that the sign characteristics of the charging and discharging power command values ​​are identified, with positive values ​​corresponding to the charging direction and negative values ​​corresponding to the discharging direction. Subsequently, based on the real-time SOC value change trend of the energy storage system, when the SOC value moves towards the storage boundary, the power limit value in the corresponding direction is automatically reduced. Then, the power limit value is dynamically adjusted in conjunction with temperature parameters, and the power limit is reduced accordingly when the temperature rises. Finally, three levels of verification are performed: directional consistency verification ensures that the power command is consistent with the current operating mode, power limit verification ensures that the command value does not exceed the adjusted power limit, and safe operation verification ensures that the power change rate is within the allowable range. The power command value that passes all verifications is used as the final energy storage charging and discharging control command output.

[0075] S4. Input the energy storage charging and discharging control command into the power flow optimization model, and generate the voltage suppression control vector through the dynamic power deviation compensation algorithm;

[0076] S4.1. Build a power flow optimization model based on the data input layer, optimization calculation layer and vector generation layer;

[0077] It should be noted that, firstly, a multi-source data interface is configured at the data input layer to receive the network-wide synchronous voltage waveform and energy storage charging and discharging control commands, and then time alignment and dimension normalization are performed. Subsequently, at the optimization calculation layer, a second-order cone programming problem is established with the goal of minimizing network loss and the constraint of node voltage safety. The interior point method is used to transform the constrained optimization problem into a sequential unconstrained optimization problem to obtain the voltage correction amount for each node. Finally, at the vector generation layer, the voltage correction amount is converted into a voltage suppression control vector with amplitude and direction characteristics through virtual magnetic monopole field strength mapping.

[0078] S4.2 The data input layer normalizes and aligns the energy storage charging and discharging control commands with the network-wide synchronous voltage waveform in time and space to generate a power flow optimization input matrix;

[0079] It should be noted that the power values ​​of the energy storage charging and discharging control commands and the instantaneous amplitude of the network-wide synchronous voltage waveform are collected, and data time alignment is ensured through GPS clock synchronization. Subsequently, the power values ​​of the energy storage charging and discharging control commands are converted into per-unit values ​​based on the rated power of the energy storage converter, and the instantaneous amplitude of the network-wide synchronous voltage waveform is converted into per-unit values ​​based on the rated voltage of the distribution network. Then, a structured array containing time series, power per-unit values, and voltage per-unit values ​​is constructed to ensure that the power-voltage data at each sampling time is completely corresponding. Finally, the integrity and consistency of the structured array are verified, and a power flow optimization input matrix that can be directly used for power flow optimization calculation is output.

[0080] S4.3 The optimization calculation layer performs second-order cone programming to solve the power flow optimization input matrix and generates a set of node voltage corrections;

[0081] It should be noted that, firstly, the power per-unit and voltage per-unit values ​​in the power flow optimization input matrix are converted into the standard form of second-order cone programming, constructing an optimization problem that includes the network loss minimization objective and node voltage safety constraints; the Lagrange multipliers and optimization variables are initialized to form an initial feasible solution. During the iteration process, the update direction of the original and dual variables is determined according to the current solution state, where the adjustment magnitude of the Lagrange multipliers is related to the degree of constraint deviation; then, the iteration step size is determined through step size search to ensure that the solution always meets the feasibility requirements, while tracking the changes of the original and dual residuals. The original residuals reflect the constraint satisfaction status, and the dual residuals reflect the progress of objective improvement. When both the original and dual residuals are at the preset adjustment convergence threshold, convergence is determined, and the optimized adjustment amounts of the voltage amplitude and phase angle of each node are output; finally, the optimized adjustment amounts of the voltage amplitude and phase angle are sorted by node number to form a complete set of node voltage correction amounts.

[0082] It should be noted that the convergence threshold is set based on numerical experiments and empirical analysis under typical power network operating conditions, with an example value range of (0.000001, 0.0005) per unit.

[0083] S4.4 The vector generation layer performs virtual magnetic monopole field strength mapping on the set of node voltage corrections to generate voltage suppression control vectors.

[0084] It should be noted that the voltage amplitude adjustment in the node voltage correction set is converted into the equivalent magnetic charge intensity of the virtual magnetic monopole, and the voltage phase angle adjustment is converted into the spatial azimuth angle of the magnetic monopole. Then, a virtual magnetic monopole array is arranged in a three-dimensional coordinate system, with each node's correction corresponding to a magnetic monopole unit. Next, based on the principle of electromagnetic field superposition, the composite field strength distribution at each spatial location point is obtained, and the field strength direction is determined by the interaction of each magnetic monopole unit. Then, the composite field strength distribution is vector-superimposed with the original node voltage correction, and the superimposed field strength control quantity reflects the optimal direction of voltage regulation. Finally, the field strength control quantity is converted into a unit vector with a definite amplitude and phase, and the output is a voltage suppression control vector that can be directly applied to inverter control.

[0085] S5. Extract stability parameters from wind power operation data, and perform weight allocation and state matching on energy storage charging and discharging control commands and voltage suppression control vectors to output coordinated control commands. S5.1. Perform multi-scale principal component analysis on wind power operation data to extract stability parameters;

[0086] It should be noted that the wind speed and direction measurements, generator output power, instantaneous bus voltage, raw grid frequency signal, real-time SOC value of energy storage system, and temperature parameters in the wind power operation data are divided into three datasets according to time scale: second-level, minute-level, and hour-level. Then, for each time-resolution dataset, mean shift processing is performed to align the distribution centers of each variable with the origin of the coordinate system, while fully preserving the fluctuation characteristics and correlation structure of the original data. Next, the processed data is arranged into an observation matrix according to the time series, with rows representing time sampling points and columns representing different variables. Then, matrix product operation is performed on the observation matrix and its transpose, and then the observation quantity is normalized to generate a covariance matrix reflecting the correlation between variables. After that, eigenvalue decomposition is performed on the covariance matrix to obtain a set of eigenvectors sorted by eigenvalue size. Finally, the dominant eigenvectors at different time scales are orthogonally combined to form a stability parameter that includes voltage fluctuation characteristics, frequency offset characteristics, and power oscillation characteristics.

[0087] S5.2 Based on stability parameters, weight allocation and state matching are performed on the energy storage charging and discharging control commands and voltage suppression control vectors, and a collaborative optimization control command set is obtained through a multi-objective dynamic fusion algorithm;

[0088] It should be noted that the weighting coefficients of the energy storage charging and discharging control command and the voltage suppression control vector are dynamically generated based on the mapping relationship between stability parameters and weighting coefficients. These weighting coefficients exhibit nonlinear characteristics as stability parameters change. The spatial relationship between the direction of the energy storage charging and discharging control command and the direction of the voltage suppression control vector is detected. Weight superposition is performed when the direction consistency condition is met; otherwise, voltage demand is adjusted. A multi-objective dynamic fusion algorithm orthogonally decomposes the weighted energy storage charging and discharging control command and the voltage suppression control vector, ultimately outputting a control command set. This collaboratively optimized control command set includes a power regulation component and a voltage compensation component. The power regulation component is determined by a weighted combination of the energy storage charging and discharging control command and the voltage suppression control vector, while the voltage compensation component reflects the orthogonal component of the voltage suppression control vector. The output of the collaboratively optimized control command set is synchronized to the execution terminal, and the response time meets the dynamic adjustment requirements of the power system.

[0089] S5.3 Perform dynamic amplitude limiting verification and priority arbitration on the collaborative optimization control instruction set, and output collaborative control instructions.

[0090] It should be noted that the dynamic limiting verification process limits the amplitude of the power regulation component in the collaborative optimization control command set based on the rated power of the energy storage converter and the real-time SOC value of the energy storage system, ensuring that the charging and discharging power command values ​​are always within the safe operating range of the energy storage equipment. Priority arbitration, based on the real-time frequency and voltage deviations of the power grid, arranges the execution order of the power regulation component, voltage compensation component, and virtual inertia flag in the collaborative optimization control command set according to the hierarchical rule of voltage stability taking precedence over frequency regulation, and frequency regulation taking precedence over economic dispatch. The final output collaborative control command retains the functions of power regulation, voltage compensation, and virtual inertia control, while also meeting equipment safety constraints and grid emergency response requirements. The transmission delay of the collaborative control command is controlled within the allowable time range of the power system's dynamic response.

[0091] This embodiment also provides a computer device applicable to the intelligent prediction and distribution management method for wind power generation and energy storage loads, including: 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 realize the intelligent prediction and distribution management method for wind power generation and energy storage loads as proposed in the above embodiment.

[0092] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0093] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the method for intelligent prediction and distribution management of wind power generation energy storage load as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0094] In summary, this invention establishes a quantitative correlation model between wind speed fluctuations and grid response by fusing heterogeneous data such as wind speed, power, and voltage into multi-dimensional sequence parameters through a spatiotemporal dynamic coupling method, and then generating phase gradient quantities through phase field gradient extraction, thereby improving the accuracy of load forecasting. The spatiotemporal crystal phase field evolution control technology unifies the node voltage phase through clock synchronization signals and generates energy storage commands by combining the spatiotemporal crystal phase field evolution direction, thereby reducing the voltage phase consistency error across the entire grid and improving the time alignment accuracy between energy storage response and grid demand.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit 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 method for intelligent prediction and distribution management of wind power generation and energy storage loads, characterized in that: include, Extract multidimensional sequence parameters from wind power operation data, perform phase field analysis on the multidimensional sequence parameters, and output the phase gradient. When the phase gradient exceeds the stability threshold, a clock synchronization signal is generated, and the voltage phase offset of the distribution network nodes is corrected according to the clock synchronization signal, and the synchronous voltage waveform of the whole network is output. The evolution direction of the phase field of the time-space crystal is analyzed on the synchronous voltage waveform of the entire network to generate energy storage charging and discharging control commands; The energy storage charging and discharging control command is input into the power flow optimization model, and a voltage suppression control vector is generated through the dynamic power deviation compensation algorithm. Stability parameters are extracted from wind power operation data, and weights are assigned and states are matched between energy storage charging and discharging control commands and voltage suppression control vectors to output coordinated control commands.

2. The method for intelligent prediction and distribution management of wind power generation and energy storage load as described in claim 1, characterized in that: The wind power operation data includes wind speed and direction measurements, generator output power, instantaneous bus voltage, raw grid frequency signal, real-time SOC value of energy storage system, and temperature parameters.

3. The method for intelligent prediction and distribution management of wind power generation and energy storage load as described in claim 2, characterized in that: The specific steps for generating the output phase gradient are as follows. Spatiotemporal dynamic coupling and energy state fusion of wind power operation data are performed to generate multidimensional sequence parameters; The spatiotemporal evolution dynamics of multidimensional order parameters are analyzed and the phase field gradient is extracted to generate phase gradient quantities.

4. The method for intelligent prediction and distribution management of wind power generation and energy storage load as described in claim 3, characterized in that: The specific steps for outputting the network-wide synchronized voltage waveform are as follows. When the phase gradient exceeds the stability threshold, the magnetic domain spin oscillation characteristics are simulated by the dynamic resonance algorithm to generate a clock synchronization signal. Based on the clock synchronization signal, the voltage phase of each node in the distribution network is calibrated, and the synchronized voltage waveform of the entire network is output.

5. The method for intelligent prediction and distribution management of wind power generation and energy storage load as described in claim 4, characterized in that: The specific steps for analyzing the spatiotemporal crystal phase field evolution direction of the entire network synchronization voltage waveform are as follows: The synchronous voltage waveform of the entire network is compared with the preset reference distribution network voltage in real time, and the instantaneous phase difference is output. Based on the instantaneous phase difference, the energy flow trend is obtained through the evolution direction of the phase field of the spatiotemporal crystal, and combined with wind power operation data for collaborative decision-making to generate charging and discharging demand markers.

6. The method for intelligent prediction and distribution management of wind power generation and energy storage load as described in claim 5, characterized in that: The specific steps for generating energy storage charging and discharging control commands are as follows: Based on the charging and discharging demand markers, power weight allocation and energy storage safety boundary constraint calculations are performed in conjunction with the instantaneous phase difference to generate charging and discharging power command values; The charging and discharging power command values ​​are dynamically calibrated in direction and subject to multi-level safety constraints to generate energy storage charging and discharging control commands.

7. The method for intelligent prediction and distribution management of wind power generation and energy storage load as described in claim 6, characterized in that: The steps for inputting energy storage charging and discharging control commands into the power flow optimization model and generating a voltage suppression control vector through a dynamic power deviation compensation algorithm are as follows: A power flow optimization model is built based on a data input layer, an optimization calculation layer, and a vector generation layer. The data input layer normalizes and aligns the energy storage charging and discharging control commands with the grid-wide synchronous voltage waveform in a spatiotemporal manner to generate a power flow optimization input matrix. The optimization computation layer performs second-order cone programming to solve the power flow optimization input matrix, generating a set of node voltage corrections. The vector generation layer performs virtual magnetic monopole field strength mapping on the set of node voltage corrections to generate voltage suppression control vectors.

8. The method for intelligent prediction and distribution management of wind power generation and energy storage load as described in claim 7, characterized in that: Stability parameters are extracted from wind power operation data, and weights are allocated and state-matched between energy storage charging and discharging control commands and voltage suppression control vectors to output coordinated control commands. The specific steps are as follows: Multi-scale principal component analysis was performed on wind power operation data to extract stability parameters; Based on stability parameters, weight allocation and state matching are performed on energy storage charging and discharging control commands and voltage suppression control vectors, and a collaborative optimization control command set is obtained through a multi-objective dynamic fusion algorithm. The collaborative optimization control instruction set is dynamically limited and priority is arbitrated, and then collaborative control instructions are output.

9. 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 wind power generation and energy storage load intelligent prediction and distribution management method according to any one of claims 1 to 8.

10. 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 wind power generation and energy storage load intelligent prediction and distribution management method according to any one of claims 1 to 8.

Images