Virtual power plant cooperative control intelligent scheduling system and method

Abstract

The invention discloses a virtual power plant cooperative control intelligent scheduling system and method, and relates to the technical field of intelligent power grids. Intelligent sensors are deployed, a virtual power plant system is combined, power grid related data are collected in real time, a power grid data set S is constructed, and feature index extraction and comprehensive calculation are performed according to the power grid data set S; and the load score prediction and evaluation module predicts the comprehensive power grid load score ZHpred (t + delta t) at a future time point t + delta t through a load score prediction model in combination with historical data and real-time data, calculates a power grid load fluctuation index BD, evaluates power grid load fluctuation, and determines the power grid load score ZH when the power grid load fluctuation is abnormal. A particle swarm optimization algorithm is used to generate an optimal scheduling scheme, power output of power generation equipment and energy storage equipment is adjusted, supply-demand balance of the power grid is recovered, resource utilization efficiency is improved, stable operation of the power grid is maintained, and intelligent self-healing and efficient management of the power grid are realized.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a virtual power plant collaborative control intelligent dispatching system and method. Background Art

[0002] Virtual power plants integrate multiple distributed energy resources, such as wind power, solar power, energy storage systems and traditional power facilities, and achieve coordinated scheduling through a central control system to optimize the balance between electricity supply and demand. Specifically, the intelligent scheduling of virtual power plants is not limited to the load management of power equipment, but also includes real-time monitoring of dynamic changes in electricity demand and production, ensuring that the power grid can maintain efficient and stable operation under changing supply and demand conditions.

[0003] At present, in most power systems, when faced with grid overload, equipment failure or external environmental interference, the grid usually takes emergency shutdown or load shedding measures. Although these methods can temporarily relieve the pressure on the grid and ensure safety, they have significant shortcomings. First, emergency shutdown and load shedding often lead to local or widespread power outages, affecting people's livelihood and industrial production, and causing serious losses to the social economy. Secondly, these measures are difficult to respond to the complex interactions of multiple factors in the grid in a timely manner, often leading to instability in the power restoration process, and then causing secondary power fluctuations or power system collapse. The current dispatching system has not yet fully utilized modern intelligent algorithms for refined and dynamic adjustments, resulting in the failure to achieve "fine-tuning" dispatching when the grid load fluctuates violently, avoiding the risk of power outages.

[0004] The above problems arise mainly because traditional power grid dispatching technology and equipment are difficult to cope with the changing power supply and demand conditions, especially when facing power grid load fluctuations or equipment failures, they lack sufficient flexibility and real-time response capabilities. In the traditional mode, the power dispatching system relies on preset rules and manual intervention, and it is difficult to make real-time load balance adjustments. With the increase in the proportion of renewable energy, the impact of volatile power sources such as wind and solar energy on the power grid is increasing, which makes the imbalance between power demand and supply more frequent. Once the power grid fails or the load suddenly surges, traditional shutdown and cutting strategies are often difficult to restore balance in time, resulting in large-scale power outages, equipment damage and excessive tension in the power grid. These abnormal phenomena not only bring inconvenience and losses to residents, enterprises and infrastructure, but also pose a serious threat to the long-term stability of the power system. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a virtual power plant collaborative control intelligent scheduling system and method, which solves the problems in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a virtual power plant collaborative control intelligent dispatching system, including a data acquisition module, a data analysis module, a load score prediction and evaluation module, a self-healing dispatching and adjustment module, and a resource allocation module;

[0007] The data acquisition module is used to collect power grid related data in real time through the wind power station and power grid monitoring system of the virtual power plant, construct a power grid data set S, and pre-process the power grid data set S;

[0008] The data analysis module is used to perform summary calculation based on the power grid data set S to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, and perform comprehensive calculation based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain the comprehensive power grid load score ZH;

[0009] The load score prediction and evaluation module is used to construct a load score prediction model, combine historical power grid data, and predict the comprehensive power grid load score ZH at the future time point t+Δt. pred (t+Δt), and combined with the comprehensive grid load score ZH at the current time point t real (t), obtaining a power grid load fluctuation index BD, and presetting a load fluctuation threshold BDYZ, comparing and analyzing the power grid load fluctuation index BD with the load fluctuation threshold BDYZ, and evaluating the power grid load fluctuation state;

[0010] The self-healing scheduling and adjustment module is used to collect power grid operation data and construct an objective function J(x) when the power grid load fluctuation state is evaluated as an abnormal state, and to obtain the optimal scheduling plan for power grid resources in combination with a particle swarm algorithm;

[0011] The resource allocation module is used to adjust the output of each energy point according to the optimal scheduling plan and optimize the allocation of power grid resources.

[0012] Preferably, the data acquisition module is used to deploy intelligent sensor groups at different nodes of the power grid and integrate the intelligent sensor groups into the virtual power plant system to collect power grid dispatch related data in real time, including wind speed V wind , power generation P gen , power grid load P load and the grid frequency f grad , wherein the smart sensor group includes an ultrasonic anemometer, a wind power station inverter, a regional smart meter, a grid load monitor, and a power quality analyzer;

[0013] The wind speed V wind The wind speed is obtained by real-time monitoring using an ultrasonic anemometer;

[0014] The power generation P gen Obtained through wind power station inverters;

[0015] The grid load P load Obtained through regional smart meters and grid load monitors;

[0016] The grid frequency f grad Real-time monitoring of the grid acquisition through the use of power quality analyzers;

[0017] According to the wind speed V wind , power generation P gen , power grid load P load and the grid frequency f grad , construct a power grid data set S, and preprocess the power grid data set S, where the preprocessing includes data cleaning, denoising and data normalization.

[0018] Preferably, the data analysis module includes a feature extraction unit and a comprehensive power grid load score calculation unit;

[0019] The feature extraction unit is used for the power grid data set S to perform summary calculations to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL;

[0020] The wind speed fluctuation coefficient FS is obtained in the following way:

[0021]

[0022] Among them, V wind (i) represents the wind speed at time point i, represents the mean value of wind speed, n represents the total number of time points of data collection, i = {1, 2, 3, ..., n};

[0023] The grid balance coefficient DW is obtained in the following way:

[0024]

[0025] Where P gen (i) represents the actual power generation at time point i, P load (i) represents the grid load at time point i.

[0026] Preferably, the grid frequency stability coefficient PL is obtained in the following manner:

[0027]

[0028] In the formula, f grad (i) represents the grid frequency at time point i, f nominalrepresents the rated frequency of the power grid, and n represents the total number of time points for data collection.

[0029] Preferably, the comprehensive power grid load score calculation unit is used to perform summary calculation using a weighted algorithm based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain a comprehensive power grid load score ZH. The comprehensive power grid load score ZH is obtained in the following manner:

[0030] ZH=α 1 FS+α 2 DW+α 3 PL+C;

[0031] In the formula, α 1 , α 2 and α 3 They respectively represent the weight coefficients of the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, and C represents the first correction constant.

[0032] Preferably, the load score prediction and assessment module includes a comprehensive power grid load score prediction unit and a load fluctuation assessment unit;

[0033] The comprehensive power grid load scoring prediction unit is used to collect historical power grid dispatch related data, construct a time series set D, and construct a load scoring prediction model based on a linear regression algorithm, and input the time series set D into the load scoring prediction model to train the load scoring prediction model, wherein the load scoring prediction model is specifically expressed in the form of:

[0034] ZH pred (t+Δt)=β 0 +β 1 ·ZH(t-1)+β 2 ·ZH(t-2)+…+β k · ZH(tk)+∈;

[0035] In the formula, β 0 represents the intercept, β 1 , β 2 and β k represents the regression coefficient of the historical comprehensive grid load score ZH, ∈ represents the random error term, ZH pred (t+Δt) represents the comprehensive grid load score at the future time point t+Δt;

[0036] The load score prediction model parameters are fitted by minimizing the mean square error (MSE), and the trained load score prediction model is used to obtain the comprehensive grid load score ZH at the future time point t+Δt. pred (t+Δt);

[0037] The load fluctuation evaluation unit is used to evaluate the comprehensive grid load score ZH at the future time point t+Δt. pred (t+Δt) and the comprehensive grid load score ZH at the current time point t real (t), obtaining the grid load fluctuation index BD, wherein the grid load fluctuation index BD is obtained in the following manner:

[0038]

[0039] In the formula, ZH pred (t+Δt) represents the comprehensive grid load score at the future time point t+Δt, ZH real (t) represents the comprehensive grid load score at the current time point t;

[0040] The load fluctuation threshold BDYZ is preset, and the load fluctuation threshold BDYZ is compared and analyzed with the grid load fluctuation index BD to evaluate the grid load fluctuation state. The specific evaluation contents are as follows:

[0041] If the grid load fluctuation index BD ≥ load fluctuation threshold BDYZ, the grid load fluctuation is determined to be abnormal, triggering the intelligent alarm mechanism, generating alarm information, and sending the alarm information to the virtual power plant monitoring personnel to perform self-healing and dispatching adjustments of the grid;

[0042] If the grid load fluctuation index BD is less than the load fluctuation threshold BDYZ, the grid load fluctuation is determined to be normal. At this time, no adjustment or scheduling is required, and the grid fluctuation situation should be continuously monitored.

[0043] Preferably, the self-healing scheduling and adjustment module includes an objective function construction unit and a scheduling algorithm execution unit;

[0044] The objective function construction unit is used to divide the power grid into several areas when the power grid load fluctuates abnormally, and obtain a power grid area set R, wherein the power grid area set R includes real-time collection of power grid operation data, wherein the power grid operation data includes the real-time load L of each area, the maximum load L max , the output power P of the power generation equipment f , Output power range Unit dispatch cost C, output power of energy storage equipment P E And the output power range

[0045] According to the power grid operation data, an objective function J(x) is constructed and constraints are set. The objective function J(x) is:

[0046]

[0047] Where R represents the total number of power grid regions, r = {1, 2, 3, ..., R}, Lr represents the actual load of region r, represents the maximum load of region r, G represents the total number of power generation equipment, j = {1, 2, 3, ..., G}, C j represents the unit dispatch cost of the jth power generation equipment, P f，j represents the output power of the jth power generation equipment;

[0048] The constraints include:

[0049]

[0050] in, and They represent the minimum output power and maximum output power of the j-th power generation equipment, and represent the minimum output power and maximum output power of the sth energy storage device, respectively. represents the maximum load of region r, U represents the total number of energy storage devices, and s = {1, 2, 3, ..., U}.

[0051] Preferably, the scheduling algorithm execution unit is used to obtain the optimal scheduling scheme of power grid resources using a particle swarm optimization algorithm according to the objective function J(x), wherein the optimal scheduling scheme acquisition process is as follows:

[0052] N scheduling schemes are randomly generated to form a scheduling scheme group, and the power allocation value and power change step size of each scheduling scheme are randomly generated, wherein the scheduling scheme is specifically expressed as follows:

[0053] X k =[P f，1 , P f，2 ,…,P f，j , P E，1 , P E，2 ,…,P E，s ];

[0054] Where P f，j represents the output power of the jth power generation equipment, P E，s represents the output power of the sth energy storage device, X k represents the kth scheduling scheme, k = {1, 2, 3, ..., N};

[0055] According to the N randomly generated scheduling plans, the objective function value J(x) of each scheduling plan is obtained, and the objective function value of each scheduling plan is used as the personal historical optimal solution. The objective function values ​​J(x) of all scheduling plans are compared, and the scheduling plan with the lowest objective function value J(x) is selected as the global optimal solution.

[0056] According to the power allocation value and power change step of each scheduling scheme, the scheduling scheme is adjusted, and the objective function value of the adjusted scheduling scheme is calculated. If the objective function value of the adjusted scheduling scheme If the value is less than the personal historical optimal solution, the adjusted scheduling scheme is updated to be the personal historical optimal solution, and the objective function value of each adjusted scheduling scheme is analyzed. Select the objective function value The lowest scheduling plan is updated as the global optimal solution, and the iterative update of the global optimal solution is repeated until the global optimal solution no longer changes. At this time, the scheduling plan represented by the global optimal solution is recorded as the optimal scheduling plan, where the optimal scheduling plan includes the power output values ​​of each power generation equipment and energy storage equipment.

[0057] Preferably, the resource allocation module is used to obtain the optimal power output value of each power generation equipment and energy storage device according to the optimal scheduling plan, and generate control instructions based on the obtained optimal power output value of each power generation equipment and energy storage device, and send the control instructions to each power generation equipment and energy storage device to adjust the output power of the equipment.

[0058] Preferably, a virtual power plant collaborative control intelligent scheduling method comprises the following steps:

[0059] Step 1: Through the wind power station and power grid detection system of the virtual power plant, collect power grid related data in real time, build a power grid data set S, and pre-process the power grid data set S;

[0060] Step 2: Based on the power grid data set S, perform summary calculation to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, and perform comprehensive calculation based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain the comprehensive power grid load score ZH;

[0061] Step 3: Construct a load score prediction model, combine historical grid data, and predict the comprehensive grid load score ZH at the future time point t+Δt pred (t+Δt), and combined with the comprehensive grid load score ZH at the current time point t real (t), obtaining a power grid load fluctuation index BD, and presetting a load fluctuation threshold BDYZ, comparing and analyzing the power grid load fluctuation index BD with the load fluctuation threshold BDYZ, and evaluating the power grid load fluctuation state;

[0062] Step 4: When the grid load fluctuation state is evaluated as an abnormal state, the grid operation data is collected, and the objective function J(x) is constructed, and the optimal scheduling plan of the grid resources is obtained by combining the particle swarm algorithm;

[0063] Step 5: According to the optimal dispatching plan, adjust the output of each energy point and optimize the allocation of power grid resources.

[0064] The present invention provides a virtual power plant collaborative control intelligent dispatching system, which has the following beneficial effects:

[0065] (1) Through real-time data collection and analysis, the operation status of the power grid is comprehensively evaluated. The wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL are used as key characteristic indicators to dynamically calculate the comprehensive power grid load score ZH. In combination with the power grid load fluctuation index BD, the load fluctuation status of the power grid is judged in real time. When the power grid is abnormal, the system can quickly trigger the self-healing scheduling and adjustment module, and dynamically adjust the resource allocation plan according to the current operation data of the power grid to ensure the balance of power supply and demand in the power grid and avoid frequency anomalies, power outages and equipment failures caused by load fluctuations, thereby greatly improving the stability and reliability of the power grid operation.

[0066] (2) The particle swarm optimization algorithm is used to optimize the dispatch of power grid resources. By constructing the objective function and setting multi-dimensional constraints, the optimal power allocation to different regions, power generation equipment and energy storage equipment is achieved. Compared with the traditional dispatching method, the particle swarm optimization algorithm can quickly converge to the global optimal solution and dynamically respond to changes in the power grid operating environment, significantly improving the intelligence level and efficiency of dispatching. In addition, the regional division design performs differentiated dispatching according to the characteristics and needs of different regions, improves resource utilization efficiency, and reduces the computational complexity of the overall power grid dispatching.

[0067] (3) The volatility of wind power generation is monitored and dynamically adjusted in real time through the virtual power plant system. The wind speed fluctuation coefficient FS and the grid balance coefficient DW are used to accurately evaluate the impact of wind power generation. The energy storage equipment is combined to adjust the charge and discharge to smooth the volatility of wind power generation, thereby improving the utilization rate of renewable energy such as wind power. By optimizing the power allocation of energy storage equipment and power generation equipment, the dispatching cost is further reduced, and economical and efficient power dispatch is achieved. In addition, the grid load prediction function based on weighted comprehensive scoring can identify load anomalies in advance and adjust the dispatching plan, avoiding the problem of unstable grid operation caused by the fluctuation of new energy in traditional methods, and promoting the widespread application of new energy in smart grids. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a block diagram of a virtual power plant collaborative control intelligent dispatching system of the present invention.

[0069] Figure 2 A schematic flow chart of a virtual power plant collaborative control intelligent scheduling method according to the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] Example 1

[0072] See also Figure 1 , the present invention provides a virtual power plant collaborative control intelligent dispatching system, including a data acquisition module, a data analysis module, a load score prediction and evaluation module, a self-healing dispatching and adjustment module, and a resource allocation module;

[0073] The data acquisition module is used to collect power grid related data in real time through the wind power station and power grid monitoring system of the virtual power plant, construct a power grid data set S, and pre-process the power grid data set S;

[0074] The data analysis module is used to perform summary calculation based on the power grid data set S to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, and perform comprehensive calculation based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain the comprehensive power grid load score ZH;

[0075] The load score prediction and evaluation module is used to construct a load score prediction model, combine historical power grid data, and predict the comprehensive power grid load score ZH at the future time point t+Δt. pred (t+Δt), and combined with the comprehensive grid load score ZH at the current time point t real (t), obtaining a power grid load fluctuation index BD, and presetting a load fluctuation threshold BDYZ, comparing and analyzing the power grid load fluctuation index BD with the load fluctuation threshold BDYZ, and evaluating the power grid load fluctuation state;

[0076] The self-healing scheduling and adjustment module is used to collect power grid operation data and construct an objective function J(x) when the power grid load fluctuation state is evaluated as an abnormal state, and to obtain the optimal scheduling plan for power grid resources in combination with a particle swarm algorithm;

[0077] The resource allocation module is used to adjust the output of each energy point according to the optimal scheduling plan and optimize the allocation of power grid resources.

[0078] In the embodiment, through the collaborative work of the data acquisition module, the data analysis module, the load score prediction and evaluation module, the self-healing scheduling and adjustment module and the resource allocation module, the problems of insufficient ability to cope with load fluctuations, slow response speed and low resource utilization efficiency in traditional power grid scheduling are effectively solved. The data acquisition module uses the monitoring system of the virtual power plant to collect power grid-related data in real time, build a complete power grid data set S, and improve the accuracy and consistency of the data through preprocessing technology; the data analysis module quantifies and comprehensively evaluates the operating status of the power grid based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, obtains the comprehensive power grid load score ZH, and provides accurate operating status evaluation; the load score prediction and evaluation module uses the load score prediction model, combines historical data and real-time data, predicts the comprehensive power grid load score ZH at future time points and calculates the power grid load fluctuation index BD, so as to achieve early warning of power grid load fluctuations. The self-healing scheduling and adjustment module is automatically triggered when the fluctuation is abnormal, and uses the particle swarm optimization algorithm to quickly generate the optimal scheduling plan, dynamically adjust the power output of power generation equipment and energy storage equipment, and restore the balance of power supply and demand; the resource allocation module performs real-time optimization and regulation according to the scheduling plan to improve resource utilization efficiency. The coordinated operation of the entire system significantly improves the stability, flexibility and economy of power grid operation, and realizes the intelligent self-healing and efficient management of the power grid.

[0079] Example 2

[0080] Please refer to Figure 1 Specifically: the data acquisition module is used to deploy intelligent sensor groups at different nodes of the power grid and integrate the intelligent sensor groups into the virtual power plant system to collect power grid dispatch related data in real time, including wind speed V wind , power generation P gen , power grid load P load and the grid frequency f grad , wherein the smart sensor group includes an ultrasonic anemometer, a wind power station inverter, a regional smart meter, a grid load monitor, and a power quality analyzer;

[0081] The virtual power plant system is an intelligent energy management system that integrates multiple distributed energy sources. It integrates energy equipment distributed in different geographical locations, such as wind power generation, solar power generation and energy storage equipment, into a "virtual power production and management entity" through information and communication technology (ICT) to achieve unified monitoring, scheduling and optimization.

[0082] The wind speed V wind The wind speed is obtained by real-time monitoring using an ultrasonic anemometer;

[0083] The power generation P gen Obtained through wind power station inverters;

[0084] The grid load P load Obtained through regional smart meters and grid load monitors;

[0085] The grid frequency f grad Real-time monitoring of the grid acquisition through the use of power quality analyzers;

[0086] According to the wind speed V wind , power generation P gen , power grid load P load and the grid frequency f grad , construct a power grid data set S, and preprocess the power grid data set S, where the preprocessing includes data cleaning, denoising and data normalization.

[0087] In the embodiment, the intelligent sensor group is deployed at different nodes of the power grid and integrated into the virtual power plant system, so as to realize the real-time monitoring and data collection of the power grid operation status, which has important beneficial effects. The wind speed V is monitored in real time by the ultrasonic anemometer. wind , accurately reflects the volatility of wind power generation, provides data support for the calculation of wind speed fluctuation coefficient FS and new energy scheduling, and collects power generation power P through the inverter of the wind power station gen , can grasp the operating status of wind power generation equipment in real time, ensure the efficient use of power generation resources, and realize the grid load capacity P load The refined monitoring provides comprehensive data support for evaluating the balance of power supply and demand; the power quality analyzer monitors the power grid frequency in real time grad , timely reflect the power grid frequency fluctuations, improve the safety and reliability of power grid operation, and at the same time, through the wind speed V wind , power generation P gen , power grid load P load and the grid frequency f grad , construct a power grid data set S, and perform preprocessing such as data cleaning, denoising and normalization, which improves the quality and applicability of the data, avoids the impact of noise data and abnormal data on subsequent analysis and optimization, enhances the intelligence level and data-driven ability of the power grid dispatching system, and lays a solid foundation for accurate analysis and efficient dispatching.

[0088] Example 3

[0089] Please refer to Figure 1 ,Specifically: the data analysis module includes a feature extraction unit and a comprehensive ,grid load score calculation unit;

[0090] The feature extraction unit is used for the power grid data set S to perform summary calculations to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL;

[0091] The wind speed fluctuation coefficient FS is obtained in the following way:

[0092]

[0093] Among them, V wind (i) represents the wind speed at time point i, represents the mean value of wind speed, n represents the total number of time points for data collection, i = {1, 2, 3, ..., n}, the larger the wind speed fluctuation coefficient FS is, the stronger the fluctuation of power output is, which leads to the instability of power generation. This instability will affect the supply and demand balance of the power grid and cause abnormal fluctuations in the power grid frequency;

[0094] The grid balance coefficient DW is obtained in the following way:

[0095]

[0096] Where P gen (i) represents the actual power generation at time point i, P load (i) represents the grid load at time point i, where the actual power generation power P at time point i is gen (n) obtained through the power station SCADA system;

[0097] The power station SCADA system is a monitoring and data acquisition system, which is widely used in power systems, such as automatic control and operation monitoring of power stations and substations, and is responsible for real-time monitoring, control and management of internal equipment and operation status of power stations.

[0098] The grid frequency stability coefficient PL is obtained in the following manner:

[0099]

[0100] In the formula, f grad (i) represents the grid frequency at time point i, f nominal represents the rated frequency of the power grid, n represents the total number of time points of data collection, wherein the rated frequency of the power grid f nominal Obtained through the national power system safety standards, the grid frequency stability coefficient PL is an indicator used to describe the degree of grid frequency fluctuation and reflects the stability of grid operation. The smaller the grid frequency stability coefficient PL is, the more stable the grid frequency is and the more reliable the grid operation is.

[0101] The comprehensive power grid load score calculation unit is used to perform summary calculation using a weighted algorithm based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain a comprehensive power grid load score ZH. The comprehensive power grid load score ZH is obtained in the following manner:

[0102] ZH=α 1 FS+α 2 DW+α 3 PL+C;

[0103] In the formula, α 1 , α 2 and α 3 They represent the weight coefficients of wind speed fluctuation coefficient FS, grid balance coefficient DW and grid frequency stability coefficient PL respectively. C represents the first correction constant, where the weight coefficient is set by the customer according to the actual situation. 0<α 1 <1,0<α 2 <1,0<α 3 <1, α 1 +α 2 +α 3 =1.

[0104] In the embodiment, the feature extraction unit and the comprehensive power grid load score calculation unit of the data analysis module comprehensively improve the analysis and evaluation capabilities of the power grid state, and effectively solve the problems of poor real-time performance, insufficient feature analysis, and low scheduling accuracy in traditional power grid scheduling. The feature extraction unit calculates the wind speed fluctuation coefficient FS, the power grid balance coefficient DW, and the power grid frequency stability coefficient PL by real-time aggregation of the power grid data set S, and comprehensively evaluates the volatility of wind power generation, the power grid supply and demand balance, and the frequency stability. The wind speed fluctuation coefficient FS accurately reflects the fluctuation characteristics of wind power generation, and provides data support for the optimal utilization of new energy power generation. The power grid balance coefficient DW can intuitively measure the matching degree between power generation and load demand, and timely identify the supply and demand imbalance problem. The power grid frequency stability coefficient PL is used to describe the frequency fluctuation of power grid operation and reflect the overall stability of the power grid. The comprehensive power grid load score calculation unit summarizes the above-mentioned feature indicators through a weighted algorithm to generate a comprehensive power grid load score ZH, which provides an accurate quantitative basis for subsequent load prediction and scheduling optimization. wind , power grid load P load and grid frequency f grad Comprehensive analysis and scoring of multi-dimensional characteristics such as energy efficiency and power generation have enabled dynamic evaluation of the grid’s operating status, greatly improving the real-time, accuracy and stability of grid dispatch. At the same time, it has effectively responded to the volatility challenges brought about by the access of new energy sources and ensured the reliability and economy of the grid.

[0105] Example 4

[0106] Please refer to Figure 1 ,Specifically: the load scoring prediction and evaluation module includes a comprehensive power grid load scoring prediction unit and a load fluctuation evaluation unit;

[0107] The comprehensive power grid load scoring prediction unit is used to collect historical power grid dispatch related data, construct a time series set D, and construct a load scoring prediction model based on a linear regression algorithm, and input the time series set D into the load scoring prediction model to train the load scoring prediction model, wherein the load scoring prediction model is specifically expressed in the form of:

[0108] ZH pred (t+Δt)=β 0 +β 1 ·ZH(t-1)+β 2 ·ZH(t-2)+…+β k · ZH(tk)+∈;

[0109] In the formula, β 0 represents the intercept, β 1 , β 2 and β k represents the regression coefficient of the historical comprehensive grid load score ZH, ∈ represents the random error term, ZH pred (t+Δt) represents the comprehensive grid load score at the future time point t+Δt;

[0110] The load score prediction model parameters are fitted by minimizing the mean square error (MSE), and the trained load score prediction model is used to obtain the comprehensive grid load score ZH at the future time point t+Δt. pred (t+Δt);

[0111] The load fluctuation evaluation unit is used to evaluate the comprehensive grid load score ZH at the future time point t+Δt. pred (t+Δt) and the comprehensive grid load score ZH at the current time point t real (t), obtaining the grid load fluctuation index BD, wherein the grid load fluctuation index BD is obtained in the following manner:

[0112]

[0113] In the formula, ZH pred (t+Δt) represents the comprehensive grid load score at the future time point t+Δt, ZH real (t) represents the comprehensive grid load score at the current time point t;

[0114] The load fluctuation threshold BDYZ is preset, and the load fluctuation threshold BDYZ is compared and analyzed with the grid load fluctuation index BD to evaluate the grid load fluctuation state. The specific evaluation contents are as follows:

[0115] If the grid load fluctuation index BD ≥ load fluctuation threshold BDYZ, the grid load fluctuation is determined to be abnormal, triggering the intelligent alarm mechanism, generating alarm information, and sending the alarm information to the virtual power plant monitoring personnel to perform self-healing and dispatching adjustments of the grid;

[0116] If the grid load fluctuation index BD is less than the load fluctuation threshold BDYZ, the grid load fluctuation is determined to be normal. At this time, no adjustment or scheduling is required, and the grid fluctuation situation should be continuously monitored.

[0117] In the embodiment, the time series set D is constructed by collecting historical power grid dispatch data, and a load score prediction model is established by using a linear regression algorithm. The load score prediction model parameters are fitted by minimizing the mean square error MSE to generate a comprehensive power grid load score ZH at the future time point t+Δt. pred (t+Δt), this process effectively combines historical trends with the current grid operation status, providing data-driven prediction capabilities for grid operation. The load fluctuation assessment unit uses the comprehensive grid load score ZH at the future time point t+Δt to calculate the load fluctuation score ZH. pred (t+Δt) and the comprehensive grid load score ZH at the current time point t real (t), obtain the grid load fluctuation index BD, and compare and analyze it with the preset load fluctuation threshold BDYZ to quickly determine the grid fluctuation state. When the grid load fluctuation index BD is greater than or equal to the load fluctuation threshold BDYZ, the system can trigger the intelligent alarm mechanism, generate alarm information in time and notify the dispatcher, start grid self-healing and dispatch adjustment. When the grid load fluctuation index BD is within the normal range, the system continuously monitors the grid fluctuation to ensure the smooth operation of the grid. A complete closed loop from accurate prediction of load fluctuation to real-time evaluation and dynamic response is realized, which not only improves the efficiency and flexibility of grid dispatch, but also effectively avoids power outages, frequency fluctuations and equipment losses caused by load fluctuations, ensuring the efficient and stable operation of the grid.

[0118] Example 5

[0119] Please refer to Figure 1 ,Specifically: the self-healing scheduling and adjustment module includes an objective function building unit and a scheduling algorithm execution unit;

[0120] The objective function construction unit is used to divide the power grid into several areas when the power grid load fluctuates abnormally, and obtain a power grid area set R, wherein the power grid area set R includes real-time collection of power grid operation data, wherein the power grid operation data includes the real-time load L of each area, the maximum load L max , the output power P of the power generation equipment f , Output power range Unit dispatch cost C, output power of energy storage equipment P EAnd the output power range

[0121] According to the power grid operation data, an objective function J(x) is constructed and constraints are set. The objective function J(x) is:

[0122]

[0123] Where R represents the total number of power grid regions, r = {1, 2, 3, ..., R}, L r represents the actual load of region r, represents the maximum load of region r, G represents the total number of power generation equipment, j = {1, 2, 3, ..., G}, C j represents the unit dispatch cost of the jth power generation equipment, P f，j represents the output power of the jth power generation equipment;

[0124] The constraints include:

[0125]

[0126] in, and They represent the minimum output power and maximum output power of the j-th power generation equipment, and represent the minimum output power and maximum output power of the sth energy storage device, respectively. represents the maximum load of region r, U represents the total number of energy storage devices, and s = {1, 2, 3, ..., U}.

[0127] The scheduling algorithm execution unit is used to obtain the optimal scheduling scheme of power grid resources using a particle swarm optimization algorithm according to the objective function J(x), wherein the optimal scheduling scheme acquisition process is as follows:

[0128] Randomly generate N scheduling schemes to form a scheduling scheme group, and the generated scheduling scheme must meet the constraints, and randomly generate a power allocation value and a power change step size for each scheduling scheme, wherein the scheduling scheme is specifically expressed in the form of:

[0129] X k =[P f，1 , P f，2 ,…,P f，j , P E，1 , P E，2 ,…,P E，s ];

[0130] Where P f，j represents the output power of the jth power generation equipment, P E，srepresents the output power of the sth energy storage device, X k represents the kth scheduling scheme, k = {1, 2, 3, ..., N};

[0131] According to the N randomly generated scheduling plans, the objective function value J(x) of each scheduling plan is obtained, and the objective function value of each scheduling plan is used as the personal historical optimal solution. The objective function values ​​J(x) of all scheduling plans are compared, and the scheduling plan with the lowest objective function value J(x) is selected as the global optimal solution.

[0132] According to the power allocation value and power change step of each scheduling scheme, the scheduling scheme is adjusted, and the objective function value of the adjusted scheduling scheme is calculated. If the objective function value of the adjusted scheduling scheme If the value is less than the personal historical optimal solution, the adjusted scheduling scheme is updated to be the personal historical optimal solution, and the objective function value of each adjusted scheduling scheme is analyzed. Select the objective function value The lowest scheduling plan is updated as the global optimal solution, and the iterative update of the global optimal solution is repeated until the global optimal solution no longer changes. At this time, the scheduling plan represented by the global optimal solution is recorded as the optimal scheduling plan, where the optimal scheduling plan includes the power output values ​​of each power generation equipment and energy storage equipment.

[0133] In the embodiment, the objective function construction unit and the scheduling algorithm execution unit in the self-healing scheduling and adjustment module realize the efficient resource scheduling optimization of the power grid under the condition of abnormal load fluctuation. First, the objective function construction unit collects the power grid operation data of each power grid area based on the area divided by the power grid, including the real-time load L, the maximum load L max , the output power P of the power generation equipment f , Output power range Unit dispatch cost C, output power of energy storage equipment P E And the output power range Dynamically construct the objective function J(x) and multi-dimensional constraints. This regional-based scheduling optimization method can flexibly formulate scheduling strategies according to the actual operating characteristics of different regions, significantly reduce the global calculation complexity, and improve scheduling efficiency. Secondly, the scheduling algorithm execution unit adopts the particle swarm optimization algorithm to randomly generate multiple scheduling schemes for global search, combined with the dynamic adjustment of the objective function value and the real-time verification of the constraints, iteratively update the personal historical optimal solution and the global optimal solution, and finally output the optimal power allocation value of each power generation equipment and energy storage equipment. Compared with the traditional scheduling method, the global search capability of the particle swarm optimization algorithm can quickly converge to the optimal scheduling plan, avoid falling into the local optimum, and significantly improve the accuracy and economy of the scheduling plan. Overall, this module not only realizes the dynamic optimization allocation of power grid resources, but also effectively improves the self-healing ability of the power grid under abnormal load fluctuations, ensuring the stability, economy and reliability of power grid operation.

[0134] Example 6

[0135] Please refer to Figure 1 Specifically: the resource allocation module is used to obtain the optimal power output value of each power generation equipment and energy storage equipment according to the optimal scheduling plan, and generate control instructions based on the obtained optimal power output value of each power generation equipment and energy storage equipment, and send the control instructions to each power generation equipment and energy storage equipment to adjust the output power of the equipment.

[0136] In the embodiment, by executing the optimal scheduling scheme, the optimal power output value of each power generation equipment and energy storage equipment is accurately obtained, and these values ​​are converted into specific control instructions and sent to the device terminal, and the output power of the equipment is dynamically adjusted, so as to achieve the optimal allocation of power grid resources. The core advantages of this module are reflected in the following aspects: First, by executing the optimal scheduling scheme, the resource utilization efficiency is significantly improved, the working potential of the power generation equipment and energy storage equipment is maximized, and resource waste and scheduling redundancy are avoided. Secondly, the control instructions are dynamically generated and real-time adjustments are implemented to ensure the balance of supply and demand of the power grid. Especially in the case of load fluctuations or abnormal conditions, the power grid frequency fluctuations caused by insufficient or overloaded loads are reduced through rapid response capabilities, thereby improving the operation stability of the power grid. Thirdly, the module fully considers the power range limitations of different equipment, the charging and discharging status of the energy storage equipment, and the scheduling cost and other multi-dimensional constraints in the allocation process, taking into account economy and reliability, and reducing the overall scheduling cost. Finally, through the automatic control instruction generation and equipment response mechanism, the module effectively reduces the need for manual intervention, improves the scheduling efficiency and intelligence level, adapts to the high requirements of modern smart grids for real-time adjustment, improves the flexibility and economy of power grid scheduling, and also provides a strong guarantee for the stable operation of the system.

[0137] Example 7

[0138] Please refer to Figure 2 Specifically: A virtual power plant collaborative control intelligent scheduling method includes the following steps:

[0139] Step 1: Through the wind power station and power grid detection system of the virtual power plant, collect power grid related data in real time, build a power grid data set S, and pre-process the power grid data set S;

[0140] Step 2: Based on the power grid data set S, perform summary calculation to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, and perform comprehensive calculation based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain the comprehensive power grid load score ZH;

[0141] Step 3: Construct a load score prediction model, combine historical grid data, and predict the comprehensive grid load score ZH at the future time point t+Δt pred (t+Δt), and combined with the comprehensive grid load score ZH at the current time point t real (t), obtaining a power grid load fluctuation index BD, and presetting a load fluctuation threshold BDYZ, comparing and analyzing the power grid load fluctuation index BD with the load fluctuation threshold BDYZ, and evaluating the power grid load fluctuation state;

[0142] Step 4: When the grid load fluctuation state is evaluated as an abnormal state, the grid operation data is collected, and the objective function J(x) is constructed, and the optimal scheduling plan of the grid resources is obtained by combining the particle swarm algorithm;

[0143] Step 5: According to the optimal dispatching plan, adjust the output of each energy point and optimize the allocation of power grid resources.

[0144] In the embodiment, by collecting the key operating data of the wind power station and the power grid monitoring system of the virtual power plant in real time, the power grid data set is constructed and merged for preprocessing, which solves the problems of incomplete and insufficient real-time data collection of traditional power grids, and by analyzing and extracting features of the collected data, the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL are calculated, and the comprehensive power grid load score ZH is further calculated, which can comprehensively and accurately evaluate the power grid operation status. At the same time, based on the load score prediction model, the comprehensive power grid load score ZH at the future time point t+Δt is predicted pred (t+Δt), and combined with the comprehensive grid load score ZH at the current time point t real(t) Calculate the grid load fluctuation index BD and accurately determine the grid fluctuation state, which makes up for the low load prediction accuracy and slow response to abnormal fluctuations of traditional dispatching methods. For load fluctuation states assessed as abnormal, this method quickly generates the optimal dispatching plan for grid resources by constructing objective functions and intelligent optimization algorithms, dynamically adjusts the output of each energy point, and realizes efficient management and dispatch of distributed energy. This method avoids traditional load cutting or emergency shutdown strategies, realizes the "self-healing" adjustment of the grid through fine-tuning, ensures the stability and reliability of grid operation, and improves the economy, intelligence level and new energy utilization efficiency of system dispatch.

[0145] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A virtual power plant collaborative control intelligent dispatching system, characterized by: It includes data collection module, data analysis module, load score prediction and evaluation module, self-healing scheduling and adjustment module and resource allocation module; The data acquisition module is used to collect power grid related data in real time through the wind power station and power grid monitoring system of the virtual power plant, construct a power grid data set S, and pre-process the power grid data set S; The data analysis module is used to perform summary calculation based on the power grid data set S to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, and perform comprehensive calculation based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain the comprehensive power grid load score ZH; The load score prediction and evaluation module is used to construct a load score prediction model, combine historical power grid data, and predict the comprehensive power grid load score ZH at the future time point t+Δt. pred (t+Δt), and combined with the comprehensive grid load score ZH at the current time point t real (t), obtaining a power grid load fluctuation index BD, and presetting a load fluctuation threshold BDYZ, comparing and analyzing the power grid load fluctuation index BD with the load fluctuation threshold BDYZ, and evaluating the power grid load fluctuation state; The self-healing scheduling and adjustment module is used to collect power grid operation data and construct an objective function J(x) when the power grid load fluctuation state is evaluated as an abnormal state, and to obtain the optimal scheduling plan for power grid resources in combination with a particle swarm algorithm; The resource allocation module is used to adjust the output of each energy point according to the optimal scheduling plan and optimize the allocation of power grid resources.

2. A virtual power plant collaborative control intelligent dispatching system according to claim 1, characterized in that: The data acquisition module is used to deploy intelligent sensor groups at different nodes of the power grid and integrate the intelligent sensor groups into the virtual power plant system to collect power grid dispatch related data in real time, including wind speed V wind , power generation P gen , power grid load P load and the grid frequency f grad , wherein the smart sensor group includes an ultrasonic anemometer, a wind power station inverter, a regional smart meter, a grid load monitor, and a power quality analyzer; The wind speed V wind The wind speed is obtained by real-time monitoring using an ultrasonic anemometer; The power generation P gen Obtained through wind power station inverters; The grid load P load Obtained through regional smart meters and grid load monitors; The grid frequency f grad Real-time monitoring of the grid acquisition through the use of power quality analyzers; According to the wind speed V wind , power generation P gen , power grid load P load and the grid frequency f grad , construct a power grid data set S, and preprocess the power grid data set S, where the preprocessing includes data cleaning, denoising and data normalization.

3. A virtual power plant collaborative control intelligent dispatching system according to claim 2, characterized in that: The data analysis module includes a feature extraction unit and a comprehensive power grid load score calculation unit; The feature extraction unit is used for the power grid data set S to perform summary calculations to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL; The wind speed fluctuation coefficient FS is obtained in the following way: Among them, V wind (i) represents the wind speed at time point i, represents the mean value of wind speed, n represents the total number of time points of data collection, i = {1, 2, 3, ..., n}; The grid balance coefficient DW is obtained in the following way: Where P gen (i) represents the actual power generation at time point i, P load (i) represents the grid load at time point i.

4. A virtual power plant collaborative control intelligent dispatching system according to claim 3, characterized in that: The grid frequency stability coefficient PL is obtained in the following manner: In the formula, f grad (i) represents the grid frequency at time point i, f nominal represents the rated frequency of the power grid, and n represents the total number of time points for data collection.

5. A virtual power plant collaborative control intelligent dispatching system according to claim 4, characterized in that: The comprehensive power grid load score calculation unit is used to perform summary calculation using a weighted algorithm based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain a comprehensive power grid load score ZH. The comprehensive power grid load score ZH is obtained in the following manner: ZH=α1·FS+α2·DW+α3·PL+C; Wherein, α1, α2 and α3 represent the weight coefficients of the wind speed fluctuation coefficient FS, the grid balance coefficient DW and the grid frequency stability coefficient PL respectively, and C represents the first correction constant.

6. A virtual power plant collaborative control intelligent dispatching system according to claim 5, characterized in that: The load scoring prediction and evaluation module includes a comprehensive power grid load scoring prediction unit and a load fluctuation evaluation unit; The comprehensive power grid load scoring prediction unit is used to collect historical power grid dispatch related data, construct a time series set D, and construct a load scoring prediction model based on a linear regression algorithm, and input the time series set D into the load scoring prediction model to train the load scoring prediction model, wherein the load scoring prediction model is specifically expressed in the form of: ZH pred (t+Δt)=β0+β1·ZH(t-1)+β2·ZH(t-2)+…+β k ·ZH(tk)+∈; In the formula, β0 represents the intercept, β1, β2 and β k represents the regression coefficient of the historical comprehensive grid load score ZH, ∈ represents the random error term, ZH pred (t+Δt) represents the comprehensive grid load score at the future time point t+Δt; The load score prediction model parameters are fitted by minimizing the mean square error (MSE), and the trained load score prediction model is used to obtain the comprehensive grid load score ZH at the future time point t+Δt. pred (t+Δt); The load fluctuation evaluation unit is used to evaluate the comprehensive grid load score ZH at the future time point t+Δt. pred (t+Δt) and the comprehensive grid load score ZH at the current time point t real (t), obtaining the grid load fluctuation index BD, wherein the grid load fluctuation index BD is obtained in the following manner: In the formula, ZH pred (t+Δt) represents the comprehensive grid load score at the future time point t+Δt, ZH real (t) represents the comprehensive grid load score at the current time point t; The load fluctuation threshold BDYZ is preset, and the load fluctuation threshold BDYZ is compared and analyzed with the grid load fluctuation index BD to evaluate the grid load fluctuation state. The specific evaluation contents are as follows: If the grid load fluctuation index BD ≥ load fluctuation threshold BDYZ, the grid load fluctuation is determined to be abnormal, triggering the intelligent alarm mechanism, generating alarm information, and sending the alarm information to the virtual power plant monitoring personnel to perform self-healing and dispatching adjustments of the grid; If the grid load fluctuation index BD is less than the load fluctuation threshold BDYZ, the grid load fluctuation is determined to be normal. At this time, no adjustment or scheduling is required, and the grid fluctuation situation should be continuously monitored.

7. A virtual power plant collaborative control intelligent dispatching system according to claim 6, characterized in that: The self-healing scheduling and adjustment module includes an objective function construction unit and a scheduling algorithm execution unit; The objective function construction unit is used to divide the power grid into several areas when the power grid load fluctuates abnormally, and obtain a power grid area set R, wherein the power grid area set R includes real-time collection of power grid operation data, wherein the power grid operation data includes the real-time load L of each area, the maximum load L max , the output power P of the power generation equipment f , Output power range Unit dispatch cost C, output power of energy storage equipment P E And the output power range According to the power grid operation data, an objective function J(x) is constructed and constraints are set. The objective function J(x) is: Where R represents the total number of power grid regions, r = {1, 2, 3, ..., R}, L r represents the actual load of region r, represents the maximum load of region r, G represents the total number of power generation equipment, j = {1, 2, 3, ..., G}, C j represents the unit dispatch cost of the jth power generation equipment, P f，j represents the output power of the jth power generation equipment; The constraints include: in, and They represent the minimum output power and maximum output power of the j-th power generation equipment, and represent the minimum output power and maximum output power of the sth energy storage device, respectively. represents the maximum load of region r, U represents the total number of energy storage devices, and s = {1, 2, 3, ..., U}.

8. A virtual power plant collaborative control intelligent dispatching system according to claim 7, characterized in that: The scheduling algorithm execution unit is used to obtain the optimal scheduling scheme of power grid resources using a particle swarm optimization algorithm according to the objective function J(x), wherein the optimal scheduling scheme acquisition process is as follows: N scheduling schemes are randomly generated to form a scheduling scheme group, and the power allocation value and power change step size of each scheduling scheme are randomly generated, wherein the scheduling scheme is specifically expressed as follows: X k =[P f，1 ,P f，2 ,…,P f，j ,P E，1 ,P E，2 ,…,P E，s ]; Where P f，j represents the output power of the jth power generation equipment, P E，s represents the output power of the sth energy storage device, X k represents the kth scheduling scheme, k = {1, 2, 3, ..., N}; According to the N randomly generated scheduling plans, the objective function value J(x) of each scheduling plan is obtained, and the objective function value of each scheduling plan is used as the personal historical optimal solution. The objective function values ​​J(x) of all scheduling plans are compared, and the scheduling plan with the lowest objective function value J(x) is selected as the global optimal solution. According to the power allocation value and power change step of each scheduling scheme, the scheduling scheme is adjusted, and the objective function value of the adjusted scheduling scheme is calculated. If the objective function value of the adjusted scheduling scheme If the value is less than the personal historical optimal solution, the adjusted scheduling scheme is updated to be the personal historical optimal solution, and the objective function value of each adjusted scheduling scheme is analyzed. Select the objective function value The lowest scheduling plan is updated as the global optimal solution, and the iterative update of the global optimal solution is repeated until the global optimal solution no longer changes. At this time, the scheduling plan represented by the global optimal solution is recorded as the optimal scheduling plan, where the optimal scheduling plan includes the power output values ​​of each power generation equipment and energy storage equipment.

9. A virtual power plant collaborative control intelligent dispatching system according to claim 8, characterized in that: The resource allocation module is used to obtain the optimal power output value of each power generation equipment and energy storage equipment according to the optimal scheduling plan, and generate control instructions based on the acquired optimal power output value of each power generation equipment and energy storage equipment, and send the control instructions to each power generation equipment and energy storage equipment to adjust the output power of the equipment.

10. A virtual power plant collaborative control intelligent dispatching method, used to implement a virtual power plant collaborative control intelligent dispatching system as claimed in any one of claims 1 to 9, characterized in that: The following steps are included: Step 1: Through the wind power station and power grid detection system of the virtual power plant, collect power grid related data in real time, build a power grid data set S, and pre-process the power grid data set S; Step 2: Based on the power grid data set S, perform summary calculation to obtain the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL, and perform comprehensive calculation based on the wind speed fluctuation coefficient FS, the power grid balance coefficient DW and the power grid frequency stability coefficient PL to obtain the comprehensive power grid load score ZH; Step 3: Construct a load score prediction model, combine historical grid data, and predict the comprehensive grid load score ZH at the future time point t+Δt pred (t+Δt), and combined with the comprehensive grid load score ZH at the current time point t real (t), obtaining a power grid load fluctuation index BD, and presetting a load fluctuation threshold BDYZ, comparing and analyzing the power grid load fluctuation index BD with the load fluctuation threshold BDYZ, and evaluating the power grid load fluctuation state; Step 4: When the grid load fluctuation state is evaluated as an abnormal state, the grid operation data is collected, and the objective function J(x) is constructed, and the optimal scheduling plan of the grid resources is obtained by combining the particle swarm algorithm; Step 5: According to the optimal dispatching plan, adjust the output of each energy point and optimize the allocation of power grid resources.

Images