Intelligent regulation and control method and system for full-quantity resource collaborative power grid power flow

Abstract

The invention discloses an intelligent regulation and control method and system for full-amount resource collaborative power grid power flow, and relates to the technical field of resource collaborative power grid power flow, and the method comprises the steps: obtaining a plurality of partition groups, and combining the partition groups with a power grid topology; obtaining a plurality of voltage grades, and randomly combining the partition groups with the voltage grades to obtain a plurality of efficiency optimization groups; voltage fluctuation data and power flow burden data in the power flow regulation and control process are obtained, and a power flow evaluation model is constructed based on a long-short-term neural network; and matching and constructing a point cloud grid model according to the output of the power flow evaluation model and the efficiency optimization group, performing surface analysis on the point cloud grid model to obtain the selected efficiency optimization group, and applying the selected efficiency optimization group to power grid power flow regulation and control, thereby solving the problems of contradiction between power grid power flow calculation precision and regulation and control efficiency and insufficient dynamic adaptability.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of power grid currents, and more specifically, to a method and system for intelligent control of power grid currents in coordination with all resources. Background Art

[0002] Grid flow calculation is one of the core contents in power system analysis. It helps power engineers understand the operation status of the grid by simulating the current, voltage, power and other states between nodes and branches in the grid. In the full-resource coordinated grid control, flow calculation is not limited to the flow balance of traditional generators and loads, but also considers the participation of flexible resources such as distributed energy and energy storage systems.

[0003] In order to achieve full resource coordination, the power grid needs to conduct comprehensive monitoring and data collection of various resources. Through intelligent sensors, data collection equipment and real-time monitoring platforms, data from all aspects of the power grid can be obtained, including current, voltage, power, frequency, etc. At the same time, the Internet of Things technology is used to connect distributed resources (such as distributed photovoltaics, wind power, energy storage equipment, etc.) to the power grid to form a complete resource data collection system.

[0004] The energy storage system can adjust charging and discharging according to changes in the power grid flow to balance the contradiction between supply and demand of the power grid; through machine learning algorithms (such as deep learning, reinforcement learning, etc.), it can extract patterns from a large amount of power grid operation data and predict the operation trend of the power grid.

[0005] Distributed energy and microgrid technologies significantly improve the flexibility and stability of power grid flow control by providing flexible energy generation and regulation capabilities. Distributed energy (such as solar energy, wind energy, etc.) can provide local power supply, reduce load fluctuations in centralized power grids, and balance supply and demand differences through intelligent scheduling and energy storage systems. Microgrid technology allows microgrids to operate independently or interact with the main grid to achieve load regulation and power supply optimization when the grid is under great pressure. The combination of these technologies effectively reduces load fluctuations and power imbalances in the grid, and improves the adaptability and anti-interference ability of the grid, especially in an environment where there are more new energy access and large fluctuations in load demand.

[0006] For example, the invention patent with announcement number: CN105137756B discloses a coordinated control method and system for the power grid of a steel enterprise, including an advanced optimization dispatching layer for integrating the power grid model, planning data, and load forecasting results to form a future mode section; the coordinated optimization control layer analyzes the power flow changes of the operating mode section set in different time periods through power flow calculation, combines the gateway demand prediction and the power factor calculation results to form dispatching decision information, introduces sensitivity factors in the coordinated optimization control link to correct the control strategies of the advanced AGC and PDC, and improves the response speed of active power regulation; at the same time, a heuristic cross-progressive method is used to check and correct the control instructions of AGC and AVC to obtain comprehensive coordinated control instructions, thereby minimizing the mutual influence of the two types of instructions and realizing comprehensive optimized coordinated control of the economy and safety of the power grid of the steel enterprise.

[0007] For example, the invention patent with announcement number CN111413890B announces a simulation method for a primary platform of an intelligent substation based on digital-physical hybrid simulation, including: constructing a simulation model for power system power element flow calculation; constructing a simulation model for power system grid short-circuit current calculation; coupling and connecting the power element flow calculation simulation model with the grid short-circuit current calculation simulation model; firstly performing flow simulation calculation on the power network, and then performing short-circuit current simulation calculation, and graphically displaying the electrical quantity information of each branch and component; using a server to calculate the electrical quantity information of each device on the primary platform of the substation, and using a communication card to convert the electrical quantity information into SV and GOOSE messages and send them to the actual relay protection device to complete real-time simulation. The method of the present invention realizes the combination of flow calculation and short-circuit current, graphical display of electrical quantity information, physical interface connection to the actual relay protection device, and dynamic simulation of the intelligent substation, which greatly improves the accuracy and speed of digital-physical hybrid simulation.

[0008] The above disclosed technical solutions have at least the following technical problems: In power grid power flow control, granularity and parallelism play a key role in the accuracy of power flow calculation and the efficiency of regulation. However, the existing technology lacks systematic optimization of these two aspects, resulting in limitations in power flow analysis and regulation.

[0009] The step size (granularity) of voltage level division determines the accuracy of the grid model in distinguishing different voltage levels, which directly affects the accuracy of power flow calculation and the sophistication of the control strategy. Reasonable voltage level division can ensure that the voltage distribution of the power grid is accurately characterized, so that the dispatching strategy can not only ensure voltage stability, but also reduce unnecessary calculation burden. However, if the step size of voltage level division is too large, the power flow model of the power grid will be oversimplified, making it impossible to accurately reflect the voltage deviation of some key nodes, affecting voltage stability control; if the step size is too small, it will lead to a surge in the amount of calculation, increase the complexity of calculation and control, and may affect real-time performance.

[0010] The size (parallelism) of the classification of all resources affects the execution efficiency and control response speed of power flow calculation. Appropriate classification strategies can make full use of computing resources, improve the parallelism of power flow calculation, and enable different types of power resources to be analyzed independently and optimized collaboratively, achieving precise control and rapid adjustment. However, if the classification of all resources is too large, the accuracy of power flow calculation will be reduced, making it difficult to accurately control different types of resources; if the classification is too small, the computing burden will increase, affecting the overall computing efficiency and control speed of large-scale power grids.

[0011] The existing technology lacks standardized control in setting the voltage level division step size and the total resource classification size, and often adopts fixed rules or empirical parameters, which are difficult to dynamically adjust according to the grid operation status, resulting in difficulty in optimizing the trade-off between computing efficiency and control accuracy. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0012] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method and system for intelligent control of power grid flow in coordination with all resources, which automatically evaluates and analyzes the classification size and voltage level division step of all resources and controls them in real time to solve the contradiction between the calculation accuracy of power grid flow and the control efficiency and the problem of insufficient dynamic adaptability.

[0013] To achieve the above object, the present invention provides the following technical solutions: A method for intelligent control of power grid flow in coordination with all resources comprises the following steps: classifying all resources to obtain basic partitions, randomly integrating the basic partitions to obtain several partition groups, and combining them with power grid topology respectively; obtaining initial voltage level divisions, and perturbing the voltage level divisions several times based on computer random numbers to obtain several voltage levels, and randomly combining the partition groups with the voltage levels to obtain several efficiency optimization groups; constructing a basic power grid flow model according to the voltage levels, and updating the basic power grid flow model according to the power grid topology to obtain a power grid flow model; performing power grid flow control on the power grid according to the power grid flow model, obtaining voltage fluctuation data and power flow burden data in the power flow control process, and constructing a power flow evaluation model based on a long-term and short-term neural network; matching the output of the power flow evaluation model with the efficiency optimization group and constructing a point cloud mesh model, and performing surface analysis on the point cloud mesh model to obtain the selected efficiency optimization group, and applying it to power grid flow control.

[0014] In a preferred embodiment, the full amount of resources is classified to obtain basic partitions, the basic partitions are randomly integrated to obtain several partition groups, and they are respectively combined with the power grid topology, specifically: the full amount of resources of the power grid operation is obtained, and the full amount of resources are basically classified according to a preset classification rule library to obtain basic partitions; the partitions are randomly integrated based on a random perturbation method, and the power flow coupling degree, voltage support capability and power regulation margin of each partition group are calculated, and the partition combinations that do not meet the preset threshold are screened out to obtain several partition groups; the partition groups are mapped to the actual power grid node and branch relationship based on the power grid topology structure; the impedance relationship within each partition and between partitions is calculated through the node admittance matrix, and the power flow path is determined, thereby combining with the power grid topology.

[0015] In a preferred embodiment, the initial voltage level division is obtained, and the voltage level division is disturbed several times based on computer random numbers to obtain several voltage levels, and the partition groups and voltage levels are randomly combined to obtain several performance optimization groups, specifically: historical power grid operation data is obtained, and the historical power grid operation data includes voltage fluctuations, load demands and power generation resources; the voltage fluctuation range of each area in the power grid is determined according to the historical power grid operation data, so as to perform preliminary voltage level division; the performance optimization group includes a voltage level division length and a partition group.

[0016] In a preferred embodiment, the basic power flow model of the power grid is constructed according to the voltage level, and the basic power flow model of the power grid is updated according to the power grid topology to obtain the power grid power flow model, specifically: all nodes in the power grid are divided into different node sets according to the voltage level; each voltage level corresponds to a node set; for each node in the power grid, an admittance matrix is ​​constructed according to the power grid topology and line parameters; based on the admittance matrix, a mathematical model for power flow calculation of each node is established in combination with the node power equation; the voltage amplitude, phase angle and branch power flow of each node are solved by the Newton-Raphson method; according to the power flow calculation results, the voltage and power flow information of the power grid are integrated to form a basic power flow model; first prediction data is obtained according to the power grid topology, and the basic power flow model of the power grid is updated according to the first prediction data to obtain the power grid power flow model, wherein the first prediction data includes power grid load prediction data and full resource power generation prediction data.

[0017] In a preferred embodiment, the voltage fluctuation data includes a voltage fluctuation coefficient; the specific method for obtaining the voltage fluctuation coefficient is as follows: obtaining system voltage measurement data for several time periods during the power grid flow control process; calculating the standard deviation and average value of the system voltage measurement data within the same time interval; calculating the coefficient of variation of the voltage fluctuation based on the standard deviation and average value of the system voltage measurement data; calculating the voltage fluctuation coefficient based on the voltage fluctuation coefficient of variation based on a preset voltage fluctuation coefficient calculation formula.

[0018] In a preferred embodiment, the power flow burden data includes a power flow burden coefficient; the method for obtaining the power flow burden coefficient is as follows: obtain the topological structure, node distribution, and power flow burden of the power grid, and draw a three-dimensional power flow distribution diagram through a computer; filter the power flow distribution diagram; extract features from the processed power flow data, and calculate the power flow burden evaluation coefficient based on power conservation and voltage stability conditions, combined with the load distribution coefficient and boundary constraints; simulate the power flow transmission path within the power grid through power flow simulation, and establish a power flow evolution equation to obtain the power flow distribution effect coefficient; calculate the power flow burden influence coefficient in combination with the power flow burden evaluation coefficient.

[0019] In a preferred embodiment, the output of the power flow assessment model is matched with the performance optimization group and a point cloud mesh model is constructed, specifically: a three-dimensional coordinate system is constructed, in which the three dimensions respectively represent the voltage level division length, the partition group and the power flow assessment value; for each performance optimization group, the corresponding power flow assessment value is calculated based on the power flow assessment model, and point cloud data is obtained in combination with the performance optimization group and stored as a point in the point cloud; the point cloud data corresponding to all performance optimization groups are mapped in the three-dimensional coordinate system to form an initial point cloud distribution and construct a point cloud mesh; based on the Delaunay triangulation method, the point cloud data are interconnected to form a mesh with a topological structure; the point cloud mesh is rendered to obtain a point cloud mesh model in the form of color gradient and height mapping.

[0020] In a preferred embodiment, the point cloud mesh model is subjected to surface analysis to obtain a selected efficiency optimization group, and applied to power grid flow control, specifically: the constraints of the point cloud mesh model are preset according to actual needs; several peak values ​​of the point cloud mesh model are obtained while satisfying the constraints; the efficiency optimization group corresponding to the most gentle peak value is obtained as the selected efficiency optimization group, and applied to power grid flow control.

[0021] A full-resource coordinated power grid flow intelligent control system comprises a resource partitioning module, a voltage classification module, a power grid flow module, a flow evaluation module and an efficiency optimization group selection module; the resource partitioning module is used to classify the full resources to obtain basic partitions, randomly integrate the basic partitions to obtain several partition groups, and respectively combine them with the power grid topology; the voltage classification module is used to obtain the initial voltage level division, and perform several perturbations on the voltage level division based on computer random numbers to obtain several voltage levels, and randomly combine the partition groups with the voltage levels to obtain several efficiency optimization groups; the power grid flow module is used to construct a basic power grid flow model according to the voltage level, update the basic power grid flow model according to the power grid topology, and obtain the power grid flow model; the flow evaluation module is used to perform flow control on the power grid according to the power grid flow model, obtain voltage fluctuation data and flow burden data in the flow control process, and construct a flow evaluation model based on a long-term and short-term neural network; the efficiency optimization group selection module is used to match the output of the flow evaluation model with the efficiency optimization group and construct a point cloud mesh model, and perform surface analysis on the point cloud mesh model to obtain the selected efficiency optimization group, and apply it to power grid flow control.

[0022] Technical effects and advantages of a method and system for intelligently controlling power flow in coordinated full-resource grids of the present invention: 1. The present invention improves the operating efficiency and stability of the power grid by optimizing the allocation of power grid resources. The method forms a variety of partition groups by classifying all resources and randomly integrating them, and accurately models the power grid in combination with the power grid topology. By calculating the power flow coupling degree, voltage support capability and power regulation margin, the voltage fluctuation and power flow of the power grid are dynamically regulated. Random number perturbation and multi-round iterative adjustment mechanism are used to optimize the partitions and voltage levels, thereby achieving more sophisticated power grid power flow control. At the same time, the introduction of an evaluation method based on voltage fluctuation coefficient and power flow burden coefficient improves the response speed and control accuracy of the power grid when load fluctuates and resources change, effectively solves the balance problem between the efficiency and computational complexity of power grid control, and enhances the dynamic adaptability and security of the power grid.

[0023] 2. The present invention combines the output of the power flow assessment model with the efficiency optimization group, constructs a point cloud mesh model and performs surface analysis, thereby achieving precise optimization of power grid power flow control. Through mapping in a three-dimensional coordinate system and the Delaunay triangulation method, a point cloud mesh with a topological structure is formed, providing an intuitive and efficient means of power grid status analysis. Surface analysis further ensures that the selected efficiency optimization group has the most gentle peak value, ensuring the stability and reliability of the control process. Through precise mathematical modeling and derivative quantification, this technology optimizes the effect of power flow control, reduces the unstable factors caused by power grid load fluctuations, and improves the response speed and adjustment accuracy of power grid dispatch, thereby achieving efficient and safe operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The present invention is a flowchart of a method for intelligently controlling power grid currents in coordination with all resources.

[0025] Figure 2 It is a structural schematic diagram of a full-resource coordinated power grid current intelligent control system of the present invention. DETAILED DESCRIPTION

[0026] 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.

[0027] Embodiment 1, Figure 1 The present invention provides a method for intelligently controlling power flow in a coordinated manner using all resources, including: S1, classify all resources to obtain basic partitions, randomly integrate the basic partitions to obtain several partition groups, and combine them with the power grid topology respectively.

[0028] In this embodiment, all resources are classified to obtain basic partitions, specifically: Obtain all resources of the power grid operation, and perform basic classification on all resources according to the preset classification rule library to obtain basic partitions; In this embodiment, the basic partitions are randomly integrated to obtain several partition groups, which are respectively combined with the power grid topology, specifically: After obtaining the basic partitions, the partitions are randomly integrated based on the random perturbation method, and the power flow coupling, voltage support capability and power regulation margin of each partition group are calculated. The partition combinations that do not meet the preset threshold are screened out to obtain several partition groups. Mapping the partition groups to actual grid node and branch relationships based on a grid topology; The impedance relationship within each partition and between partitions is calculated through the node admittance matrix to determine the power flow path; Specifically, firstly, based on the random number generation method, the basic partitions are randomly combined, and connection constraints are imposed to ensure that the integrated partition groups maintain electrical connectivity in the topological structure. Secondly, the power flow coupling, voltage support capacity and power regulation margin of each partition group are calculated to screen out the partition combinations that do not meet the set thresholds. Subsequently, based on the partition groups that meet the constraints, a multi-round iterative adjustment mechanism is introduced to further adjust the partition boundaries to optimize the resource coordination and power flow balance within the partitions. Finally, several partition groups that meet different operating requirements are generated.

[0029] It should be noted that by constructing a node admittance matrix, the impedance relationship between each node and other nodes is calculated. This matrix reflects the electrical characteristics of each node in the power grid and the conductance and admittance values ​​between nodes. On this basis, the impedance relationship within and between each partition is calculated to clarify the path and intensity of power flow. According to the impedance relationship, the power distribution and flow direction of each part of the power grid can be accurately determined, providing basic data for the precise control of power grid currents. The main purpose is to more efficiently reflect the power coupling relationship between different partitions in the power grid, so there is no need to elaborate on the basic operations in the existing technology.

[0030] It should be noted that the full resource data includes but is not limited to: power grid equipment data (substations, transmission lines, switchgear, etc.); geographic information data (line direction, equipment location, regional environmental information, etc.); historical flow data (voltage, flow distribution, node load, etc.); load characteristic data (industrial, commercial, residential electricity consumption types and peak and valley characteristics); power grid topology information (node ​​connection relationship, subnet structure, etc.).

[0031] It should be noted that the power flow coupling is used to measure the degree of power interaction between different partitions. When calculating, the power transfer ratio between partitions can be solved based on the node admittance matrix to evaluate the electrical coupling strength between partitions. A higher power flow coupling means that the power flows between partitions have a greater mutual influence. Adjusting the power distribution in a certain area may have a significant impact on other areas, so a more sophisticated control strategy is needed; the voltage support capacity reflects the support role of each partition for voltage stability. When calculating, the voltage regulation capacity of the partition can be evaluated based on the capacity of the reactive compensation equipment in the partition, the response range of the voltage regulation equipment, and the voltage sensitivity matrix of the key nodes. A higher voltage support capacity means that the partition can better maintain voltage stability when the load fluctuates or the topology is adjusted; the power regulation margin measures the regulation capacity of the partition under load changes and renewable energy fluctuations. It is usually determined by calculating the remaining regulation capacity of the dispatchable units in the partition, the discharge capacity of the available energy storage resources, and the adjustable load proportion of the demand side response. A larger power regulation margin indicates that the partition has stronger dynamic regulation capability and can adapt more flexibly to changes in the grid operation status. The above-mentioned power flow coupling, voltage support capability and power regulation margin have been evaluated by mature technologies. The main purpose here is to constrain the partitions, so they will not be elaborated here.

[0032] S2, obtaining an initial voltage level division, and performing several disturbances on the voltage level division based on computer random numbers to obtain several voltage levels, and randomly combining the partition groups and the voltage levels to obtain several performance optimization groups.

[0033] In this embodiment, the initial voltage level division is obtained as follows: Acquiring historical power grid operation data, wherein the historical power grid operation data includes voltage fluctuations, load demands, and power generation resources; Based on the historical grid operation data, the voltage fluctuation range and stability requirements of each area in the grid are determined, so as to make a preliminary voltage level division.

[0034] The feasible method of dividing the initial voltage level is as follows: Extract voltage data of each node from historical power grid operation data Where i represents the node, t represents the time point, and the voltage data is the instantaneous voltage value of each node in the power grid; Perform statistical analysis on the voltage of each grid node and calculate its fluctuation range. The commonly used fluctuation calculation method is to measure the fluctuation amplitude by the maximum and minimum values ​​of the node voltage; the voltage fluctuation range can be calculated by the following formula: , where is the voltage fluctuation range.

[0035] In order to meet the stability requirements of the power grid, a voltage fluctuation tolerance threshold is set, which can be set according to the load demand of the power grid and the coordination of power generation resources. If the voltage fluctuation range exceeds the threshold, it means that the voltage fluctuation of the node does not meet the stability requirements of the power grid and its voltage level needs to be adjusted.

[0036] The performance optimization group includes a voltage level partition length and a partition group.

[0037] S3, constructing a basic power flow model of the power grid according to the voltage level, updating the basic power flow model of the power grid according to the power grid topology, and obtaining a power grid power flow model.

[0038] The basic power flow model of the power grid is constructed according to the voltage level, and the basic power flow model of the power grid is updated according to the power grid topology to obtain the power grid power flow model, specifically: Divide all nodes in the power grid into different node sets according to voltage levels; each voltage level corresponds to a node set; For each node in the power grid, an admittance matrix is ​​constructed based on the power grid topology and line parameters; Based on the admittance matrix and combined with the node power equation, a mathematical model for power flow calculation of each node is established; The Newton-Raphson method is used to solve the voltage amplitude, phase angle and branch power flow at each node; According to the power flow calculation results, the voltage and power flow information of the power grid are integrated to form a basic power flow model; Acquire first prediction data according to the power grid topology, and update the basic power grid flow model according to the first prediction data to acquire the power grid flow model, wherein the first prediction data includes power grid load prediction data and total resource power generation prediction data; It should be noted that the node set reflects the voltage levels of different areas in the power grid, and the division of the node set provides the basis for subsequent power flow calculations; It should be noted that line parameters include conductance, reactance, resistance and admittance. Their application in the admittance matrix is ​​mainly reflected in the calculation of the admittance relationship between nodes. The impedance of each power line can be converted into admittance and filled into the corresponding non-diagonal elements in the admittance matrix. The admittance value between nodes reflects the electrical characteristics of the line, which in turn affects the power flow calculation results of the power grid, such as voltage distribution and power flow path. Therefore, the conductance and reactance of the line are crucial to the power flow analysis of the power grid, which determines the efficiency and stability of power transmission in the power grid.

[0039] It should be noted that the admittance matrix describes the electrical relationship between nodes, that is, the relationship between the current and voltage between nodes, and is defined as:

[0040] In the formula, is the value of the i-th row and j-th column of the admittance matrix.

[0041] It should be noted that some of the proprietary names involved in this embodiment, such as admittance matrix, power grid flow model, power flow evaluation model, etc., are all existing technologies in the field of power system analysis. Among them, there are mature calculation methods and algorithms for the power flow calculation process, such as the Newton-Raphson method, DC power flow calculation, etc., which can efficiently solve the voltage distribution and power flow of the power system. Therefore, based on the power flow calculation, this embodiment focuses on how to optimize the power grid control strategy based on the power flow evaluation model, and does not repeat the specific process of the power flow calculation.

[0042] S4, performs power flow control on the power grid according to the power grid power flow model, obtains voltage fluctuation data and power flow burden data during the power flow control process, and builds a power flow evaluation model based on long-term and short-term neural networks.

[0043] The voltage fluctuation data includes a voltage fluctuation coefficient, and the power flow burden data includes a power flow burden coefficient; The voltage fluctuation coefficient is used to measure the sensitivity of the voltage change of the power grid under specific load changes or resource fluctuations. This coefficient reflects the degree of response of the voltage fluctuation in the power grid to load changes or power generation fluctuations, and is usually calculated by constructing a sensitivity matrix (Jacobian matrix). Through this matrix, it is possible to analyze how small changes in voltage affect other parameters in the system, thereby evaluating the regulation accuracy and voltage stability of the power grid. In the calculation process, it is usually necessary to solve a multidimensional nonlinear system of equations, which may involve advanced mathematical methods such as the Newton-Raphson method and optimization algorithms. In order to improve the accuracy of the analysis, it is also necessary to extract voltage fluctuation characteristics from historical data and perform statistical modeling so that the voltage fluctuation characteristics under different load or resource configurations can be predicted and analyzed. In power grid power flow control, the voltage fluctuation coefficient is used to evaluate the impact of different full resource classifications and voltage level division steps on the power flow control effect, helping to optimize resource scheduling and voltage stability.

[0044] Analyzing the voltage fluctuation coefficient is helpful for evaluating the effect of overall power flow control under different classification sizes of total resources and voltage level division steps, solving the contradiction between power flow calculation accuracy and control efficiency and the problem of insufficient dynamic adaptability. It has the following advantages: Optimize the balance between calculation accuracy and efficiency: By accurately evaluating the voltage fluctuation coefficient, the resource classification and voltage level division step can be dynamically adjusted while ensuring the calculation accuracy, thereby avoiding excessive or insufficient consumption of computing resources and achieving the best balance between accuracy and efficiency in power flow calculation.

[0045] Improve the response speed and flexibility of power grid regulation: The voltage fluctuation coefficient can provide real-time feedback for power flow regulation, enabling the system to quickly respond to load changes and resource fluctuations, improve the dynamic adaptability of power grid regulation, and solve the problem of insufficient response to emergencies or fluctuations in traditional methods.

[0046] Reduce computational complexity and improve parallel computing efficiency: Through reasonable resource classification and voltage level division, the amount of computation can be reduced, the computation path can be optimized, and the efficiency of parallel computing can be improved, the bottleneck phenomenon in the computation process can be reduced, and the overall computing speed can be improved.

[0047] Enhance the stability and security of the power grid: Real-time monitoring of voltage fluctuations combined with power flow burden analysis can detect potential power grid stability issues at an early stage, adjust resource allocation in a timely manner, and ensure safe and stable operation of the power grid under different load and resource conditions.

[0048] The specific method for obtaining the voltage fluctuation coefficient is as follows: Obtain system voltage measurement data for several time periods during the power grid power flow control process; Calculate the standard deviation and average value of system voltage measurement data within the same time interval; Calculate the coefficient of variation of voltage fluctuation based on the standard deviation and average value of system voltage measurement data; The voltage fluctuation coefficient of variation is calculated based on a preset voltage fluctuation coefficient calculation formula.

[0049] The specific calculation formula of the coefficient of variation of the voltage fluctuation is as follows:

[0050] The specific calculation formula of the voltage fluctuation coefficient is as follows:

[0051] In the formula, is the coefficient of variation of voltage fluctuation, is the system voltage measurement data at the jth moment, is the total time, j is the time label; is the voltage fluctuation coefficient.

[0052] The power flow burden coefficient is used to measure the load bearing capacity of each area or line in the power grid during the power flow control process. It reflects the degree of burden of each part of the power grid on power flow under a given resource configuration. In parallel computing, the power flow burden coefficient can help evaluate the allocation efficiency of computing resources, especially when using multi-threaded or distributed computing architectures (such as GPU acceleration, HPC computing). During the calculation process, it may be necessary to optimize the computing time by analyzing the load balancing problem to ensure that the computing tasks are reasonably distributed to multiple computing nodes. In addition, it is necessary to monitor the communication delay and computing bottleneck during the calculation process, and performance optimization may be performed with the help of tools such as MPI (Message Passing Interface). The power flow burden coefficient can be used to evaluate the impact of different full resource classifications and voltage level division steps on power flow control, help optimize resource scheduling and computing efficiency, and thus improve the overall control effect.

[0053] Analyzing the power flow burden coefficient has the following advantages for evaluating the effect of overall power flow control under different classification sizes of total resources and voltage level division steps, solving the contradiction between power grid power flow calculation accuracy and control efficiency and the problem of insufficient dynamic adaptability: Improve computing efficiency and control response speed: By evaluating the power flow burden coefficient, we can clearly understand the burden status of each zone and grid area, optimize the parallel processing of resources, reduce unnecessary calculations, and thus improve computing efficiency and control response speed.

[0054] Optimizing resource allocation and load balancing: The power flow burden factor helps to identify load imbalance problems in the power grid and balance the load by adjusting resource allocation to avoid overload or underload in certain areas, thereby improving the stability and security of the power grid.

[0055] Improving the dynamic adaptability of grid control: The flow burden coefficient helps to monitor the load status of the grid under different operating conditions in real time, and provides flexible control strategies. It can make fast and dynamic adjustments in situations such as grid load fluctuations and resource changes, solving the problem of insufficient response to dynamic changes in traditional methods.

[0056] Reduce computational complexity and enhance system scalability: Through quantitative analysis of the power flow burden coefficient, we can reduce excessive calculation of all resources, optimize the allocation strategy of power grid resources, improve the computing power and scalability of the power grid system under large-scale operation conditions, and avoid wasting computing resources.

[0057] The method for obtaining the power flow burden coefficient is as follows: Obtain the topological structure, node distribution, and power flow burden of the power grid, and draw a three-dimensional power flow distribution map through a computer; filter the power flow distribution map; extract features from the processed power flow data, and calculate the power flow burden assessment coefficient based on power conservation and voltage stability conditions, combined with load distribution coefficients and boundary constraints; Through power flow simulation, the power flow transmission path inside the power grid is simulated, and the power flow evolution equation is established to obtain the power flow distribution effect coefficient; combined with the power flow burden assessment coefficient, the power flow burden influence coefficient is calculated; The specific calculation formula of the power flow burden evaluation coefficient is as follows:

[0058] The specific calculation formula of the power flow distribution effect coefficient is as follows:

[0059] The specific calculation formula of the power flow burden coefficient is as follows:

[0060] In the formula, is the power flow burden assessment coefficient, is the power flow distribution effect coefficient, is the power flow burden coefficient, is the error term, is the total power transmission capacity of the power grid, is the load area of ​​the power grid, is the load density, is the load distribution influencing factor, is the Laplace operator of the power flow distribution, is the load change rate, is the average power flow, is the nonlinear variation term of power flow.

[0061] It should be noted that the power flow burden assessment coefficient is used to measure the burden degree of each load area in the power grid, taking into account the load density, load distribution influencing factors and the diffusion characteristics of power flow transmission. This coefficient reflects the power transmission characteristics of different areas of the power grid and the impact of load distribution on the operation of the power grid.

[0062] Power flow distribution effect coefficient: It mainly measures the propagation of power flow in the power grid, including the instantaneous power change rate and the nonlinear change term of power flow. This coefficient evaluates the dynamic response capability of the load by calculating the evolution of power flow in the power grid topology.

[0063] Flow burden impact coefficient: The flow burden assessment coefficient and flow distribution effect coefficient are combined, and the error term is added to finally obtain the quantitative index of the overall power flow burden of the power grid. This coefficient can be used to evaluate the operating status of the power grid under different load levels and provide a basis for the flow control strategy.

[0064] Load density: indicates the distribution of power demand in a unit area, which can be fitted and corrected based on historical load data or grid planning data.

[0065] Load distribution influencing factor: It is used to describe the distribution pattern of load in different areas of the power grid. It may be affected by factors such as the power grid topology, power supply capacity and power demand, and is usually determined based on statistical data or empirical formulas.

[0066] Laplace operator of power flow distribution: It is used to describe the diffusion characteristics of power flow in the power grid, which is similar to the diffusion process in the physical field and reflects the spatial distribution and gradient changes of voltage and power in the power grid.

[0067] Instantaneous power change rate: Indicates the rate of change of load power, which can be calculated from historical load data and is usually used to analyze the dynamic characteristics and mutations of the load.

[0068] Nonlinear change term of power flow: It is used to characterize the nonlinear change characteristics of power flow in complex power grid structures, such as non-uniform power flow distribution caused by factors such as voltage fluctuations, line impedance and load characteristics.

[0069] Error term: used to correct possible errors in model calculations, including measurement errors, model assumption errors, and the impact of external factors on tidal flow distribution that are not considered.

[0070] Total power transmission capacity of the power grid: It refers to the total power load that the power grid can carry within a specific time range, which can be calculated based on the power grid capacity, transmission line limitations and scheduling strategy.

[0071] Load area of ​​the power grid: refers to the area in the power grid that needs to be analyzed in detail, such as the power supply range of substations with high loads, areas with tight power supply, or areas where key industrial users are located.

[0072] In this implementation, a power flow assessment model is constructed based on long-term and short-term neural networks, specifically: The obtained voltage fluctuation coefficient and power flow burden coefficient are used to construct a power flow assessment model based on a long-term and short-term neural network to generate a power flow assessment value; The specific calculation formula of the power flow evaluation value is as follows:

[0073] In the formula, is the power flow evaluation value, is the preset proportional coefficient of the voltage fluctuation coefficient, is the preset proportionality factor of the power flow burden factor, is the voltage fluctuation coefficient, is the current burden coefficient.

[0074] This implementation improves the operational efficiency and stability of the power grid by optimizing the allocation of power grid resources. This method forms multiple partition groups by classifying all resources and randomly integrating them, and accurately models the power grid in combination with the power grid topology. By calculating the power flow coupling degree, voltage support capability and power regulation margin, the voltage fluctuation and power flow of the power grid are dynamically regulated. The partitions and voltage levels are optimized by using random number perturbations and multi-round iterative adjustment mechanisms, thereby achieving more refined power grid power flow control. At the same time, the introduction of an evaluation method based on the voltage fluctuation coefficient and the power flow burden coefficient improves the response speed and control accuracy of the power grid when the load fluctuates and the resources change, effectively solving the balance problem between the efficiency and computational complexity of power grid control, and enhancing the dynamic adaptability and security of the power grid.

[0075] Embodiment 2, S5, matches and constructs a point cloud mesh model according to the output of the power flow assessment model and the efficiency optimization group, performs surface analysis on the point cloud mesh model, obtains the selected efficiency optimization group, and applies it to power grid power flow control.

[0076] In this embodiment, the output of the power flow assessment model and the efficiency optimization group are matched and a point cloud mesh model is constructed, specifically: Construct a three-dimensional coordinate system, where the three dimensions represent the voltage level division length, the partition group, and the power flow assessment value; For each performance optimization group, a corresponding power flow evaluation value is calculated based on the power flow evaluation model, and point cloud data is obtained in combination with the performance optimization group and stored as a point in the point cloud; Map the point cloud data corresponding to all performance optimization groups in a three-dimensional coordinate system to form an initial point cloud distribution and construct a point cloud grid; Based on the Delaunay triangulation method, the point cloud data are connected to each other to form a mesh with a topological structure; Render the point cloud mesh and obtain the point cloud mesh model in the form of color gradient and height map.

[0077] In this implementation, a surface analysis is performed on the point cloud mesh model to obtain the selected efficiency optimization group, which is applied to power grid flow control, specifically: Preset the constraints of the point cloud mesh model according to actual needs; Obtain peak values ​​of several point cloud mesh models under the constraints; The efficiency optimization group corresponding to the most gentle peak is obtained as the selected efficiency optimization group, and is applied to power grid power flow control.

[0078] It should be noted that the degree of smoothness can be quantitatively judged by constructing a mathematical model corresponding to the surface and performing quantitative derivation.

[0079] In this implementation, by combining the output of the power flow assessment model with the efficiency optimization group, a point cloud mesh model is constructed and surface analysis is performed to achieve accurate optimization of power grid power flow control. Through mapping in a three-dimensional coordinate system and the Delaunay triangulation method, a point cloud mesh with a topological structure is formed, providing an intuitive and efficient means of power grid status analysis. Surface analysis further ensures that the selected efficiency optimization group has the most gentle peak value, ensuring the stability and reliability of the control process. Through precise mathematical modeling and derivative quantification, this technology optimizes the effect of power flow control, reduces the unstable factors caused by power grid load fluctuations, and improves the response speed and adjustment accuracy of power grid dispatch, thereby achieving efficient and safe operation of the power grid.

[0080] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0081] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0082] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0083] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0084] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0085] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligent control of power flow in coordination with all resources, characterized in that: The steps include: Classify all resources to obtain basic partitions, randomly integrate the basic partitions to obtain several partition groups, and combine them with the power grid topology respectively; Obtaining an initial voltage level division, and performing several perturbations on the voltage level division based on a computer random number to obtain several voltage levels, and randomly combining the partition groups and the voltage levels to obtain several performance optimization groups; Construct a basic power flow model of the power grid according to the voltage level, update the basic power flow model of the power grid according to the power grid topology, and obtain a power grid power flow model; According to the power flow model, the power grid is regulated, the voltage fluctuation data and power flow burden data of the power flow regulation process are obtained, and the power flow evaluation model is constructed based on the long-term and short-term neural network; According to the output of the power flow assessment model and the efficiency optimization group, a point cloud mesh model is matched and constructed, and a surface analysis is performed on the point cloud mesh model to obtain the selected efficiency optimization group and apply it to power grid power flow regulation.

2. The method for intelligent control of power flow in coordinated power grid with full resources according to claim 1 is characterized in that: The whole amount of resources is classified to obtain basic partitions, and the basic partitions are randomly integrated to obtain several partition groups, which are respectively combined with the power grid topology, specifically: Obtain all resources of the power grid operation, and perform basic classification on all resources according to the preset classification rule library to obtain basic partitions; Based on the random perturbation method, the partitions are randomly integrated, and the power flow coupling, voltage support capability and power regulation margin of each partition group are calculated, and the partition combinations that do not meet the preset threshold are screened out to obtain several partition groups; Mapping the partition groups to actual grid node and branch relationships based on a grid topology; The impedance relationship within each partition and between partitions is calculated through the node admittance matrix to determine the power flow path, which is then combined with the grid topology.

3. The method for intelligent control of power flow in coordinated power grid with full resources according to claim 2 is characterized in that: The initial voltage level division is obtained, and the voltage level division is disturbed several times based on computer random numbers to obtain several voltage levels, and the partition groups and voltage levels are randomly combined to obtain several performance optimization groups, specifically: Acquiring historical power grid operation data, wherein the historical power grid operation data includes voltage fluctuations, load demands, and power generation resources; Determine the voltage fluctuation range of each area in the power grid based on historical power grid operation data, and make a preliminary voltage level division; The performance optimization group includes a voltage level partition length and a partition group.

4. The method for intelligent control of power flow in coordinated power grid with full resources according to claim 3 is characterized in that: The basic power flow model of the power grid is constructed according to the voltage level, and the basic power flow model of the power grid is updated according to the power grid topology to obtain the power grid power flow model, specifically: Divide all nodes in the power grid into different node sets according to voltage levels; each voltage level corresponds to a node set; For each node in the power grid, an admittance matrix is ​​constructed based on the power grid topology and line parameters; Based on the admittance matrix and combined with the node power equation, a mathematical model for power flow calculation of each node is established; The Newton-Raphson method is used to solve the voltage amplitude, phase angle and branch power flow at each node; According to the power flow calculation results, the voltage and power flow information of the power grid are integrated to form a basic power flow model; First prediction data is obtained according to the power grid topology, and the basic power grid flow model is updated according to the first prediction data to obtain a power grid flow model, wherein the first prediction data includes power grid load prediction data and total resource power generation prediction data.

5. The method for intelligent control of power flow in coordinated power grid with full resources according to claim 4 is characterized in that: The voltage fluctuation data includes a voltage fluctuation coefficient; The specific method for obtaining the voltage fluctuation coefficient is as follows: Obtain system voltage measurement data for several time periods during the power grid flow control process; Calculate the standard deviation and average value of system voltage measurement data within the same time interval; Calculate the coefficient of variation of voltage fluctuation based on the standard deviation and average value of system voltage measurement data; The voltage fluctuation coefficient of variation is calculated based on a preset voltage fluctuation coefficient calculation formula.

6. The method for intelligent control of power flow in coordinated power grid with full resources according to claim 5 is characterized in that: The power flow burden data includes a power flow burden coefficient; The method for obtaining the power flow burden coefficient is as follows: Obtain the topological structure, node distribution, and power flow burden of the power grid, and draw a three-dimensional power flow distribution diagram through a computer; Filter the power flow distribution diagram; Extract features from the processed power flow data, and calculate the power flow burden assessment coefficient based on power conservation and voltage stability conditions, combined with load distribution coefficients and boundary constraints; Through power flow simulation, the power flow transmission path inside the power grid is simulated, and the power flow evolution equation is established to obtain the power flow distribution effect coefficient; The flow burden impact coefficient is calculated in combination with the flow burden assessment coefficient.

7. The method for intelligent control of power flow in coordinated power grid with full resources according to claim 6 is characterized in that: The output of the power flow assessment model is matched with the efficiency optimization group and a point cloud mesh model is constructed, specifically: Construct a three-dimensional coordinate system, where the three dimensions represent the voltage level division length, the partition group, and the power flow assessment value; For each performance optimization group, a corresponding power flow evaluation value is calculated based on the power flow evaluation model, and point cloud data is obtained in combination with the performance optimization group and stored as a point in the point cloud; Map the point cloud data corresponding to all performance optimization groups in a three-dimensional coordinate system to form an initial point cloud distribution and construct a point cloud grid; Based on the Delaunay triangulation method, the point cloud data are connected to each other to form a mesh with a topological structure; Render the point cloud mesh and obtain the point cloud mesh model in the form of color gradient and height map.

8. The method for intelligent control of power flow in coordinated power grid with full resources according to claim 7 is characterized in that: The point cloud mesh model is subjected to surface analysis to obtain the selected efficiency optimization group, and applied to power grid flow control, specifically: Preset the constraints of the point cloud mesh model according to actual needs; Obtain peak values ​​of several point cloud mesh models under the constraints; The efficiency optimization group corresponding to the most gentle peak value is obtained as the selected efficiency optimization group, and is applied to power grid power flow control.

9. A system using the method for intelligent control of power flow in coordinated full resource grid according to any one of claims 1 to 8, characterized in that: It includes resource partitioning module, voltage classification module, power grid flow module, flow evaluation module and efficiency optimization group selection module; Resource partitioning module, used to classify all resources to obtain basic partitions, randomly integrate the basic partitions to obtain several partition groups, and combine them with the power grid topology respectively; A voltage grading module is used to obtain an initial voltage grading division, and to perform several perturbations on the voltage grading division based on a computer random number to obtain several voltage gradings, and to randomly combine the partition groups with the voltage gradings to obtain several performance optimization groups; The power grid flow module is used to construct a basic power grid flow model according to the voltage level, update the basic power grid flow model according to the power grid topology, and obtain the power grid flow model; The power flow assessment module is used to perform power flow control on the power grid according to the power grid power flow model, obtain the voltage fluctuation data and power flow burden data during the power flow control process, and build a power flow assessment model based on the long-term and short-term neural network; The efficiency optimization group selection module is used to match and construct a point cloud mesh model based on the output of the power flow evaluation model and the efficiency optimization group, and perform surface analysis on the point cloud mesh model to obtain the selected efficiency optimization group and apply it to power grid power flow control.

Images