Dynamic scheduling method for micro-wind power generation access to smart grid

Abstract

The application relates to the technical field of power grid dispatching, and provides a dynamic dispatching method for micro-wind power access to a smart power grid. Wind power generation data of a wind power plant is acquired and processed to obtain wind power generation, and the operation state of the wind power plant is monitored in real time to determine whether the wind power plant is abnormal. Load data of the power grid is acquired and processed to obtain power grid load, and the operation state of the power grid is monitored in real time to determine whether the power grid is abnormal through an abnormality feedback mechanism. Particle parameters of each particle are generated based on the wind power generation and the power grid load, a dispatching strategy is generated by comprehensively considering the abnormal state of the wind power plant and the abnormal type of the power grid, and the wind power access proportion of each monitoring point, the energy storage dispatching sequence and the load fluctuation threshold are output, so that the wind energy resources can be used to the maximum limit, and energy waste is reduced. The actual execution of monitoring and dispatching and the operation states of the wind power plant and the power grid are monitored to obtain feedback results, and the dispatching strategy is adjusted according to the feedback results.

Description

Technical Field

[0001] The present application relates to the field of power grid dispatching technology, and in particular to a dynamic dispatching method for connecting breeze power generation to a smart grid. Background Art

[0002] To address the uncertainty and intermittency of wind power generation, researchers have proposed multi-objective optimization scheduling strategies that can reduce microgrid operating costs and improve economic efficiency. With the development of machine learning technology, deep reinforcement learning is increasingly being used in microgrid optimization scheduling, enabling real-time responses to wind speed changes and load demands, ensuring grid stability and reliability. Furthermore, scheduling algorithms are adaptable to varying grid structures and operating conditions, demonstrating strong flexibility and adaptability. However, most approaches have not addressed how to determine the scheduling status of each monitoring point based on the type of grid anomaly and the abnormal state of the wind farm.

[0003] For example, a Chinese patent application with publication number CN117424238A discloses a method, system and storage medium for optimizing the energy scheduling of a power grid. The method includes: collecting energy data of each node in the power grid to generate an energy data set; preprocessing and cleaning the energy data set, and constructing a topological map of the power grid based on the preprocessed and cleaned energy data set; using a graph computing algorithm to analyze and calculate the power grid topological map to obtain calculation results; generating an optimized scheduling decision based on the calculation results, so as to optimize the scheduling of the power grid energy of the power grid according to the optimized scheduling decision, so as to achieve optimized scheduling of the power grid energy and improve the operating efficiency and stability of the power grid.

[0004] For example, the Chinese patent with authorization announcement number CN111260248B discloses a method for self-healing scheduling of distribution network faults. (1) The distribution terminal uploads monitoring information of the distribution network and power equipment to a monitoring platform; (2) The monitoring platform processes and analyzes the information to determine whether a fault has occurred; (3) If no fault has occurred, the optimization scheduling is performed with the goal of minimizing the economic cost of operating the distribution network; if a fault has occurred, the optimization scheduling is performed with the goal of minimizing the operating cost of the distribution network under the fault state. This method fully utilizes the dispatchable characteristics of distributed power sources and the demand response characteristics of distribution network users, while considering the self-healing scheduling of the distribution network under different fault scenarios. It can restore power to as many users as possible in the event of a fault, reduce load shedding, and improve the self-healing capability of the distribution network.

[0005] The above existing technologies have the problem raised by this background technology: the above existing technologies do not solve the problem of how to determine the dispatching status of each monitoring point according to the abnormal type of the power grid and the abnormal state of the wind farm. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, this application provides a dynamic scheduling method for connecting micro-wind power generation to a smart grid. The method includes: acquiring and processing wind power data of a wind farm to obtain wind power generation, and monitoring the operating status of the wind farm in real time to determine whether there is any abnormality in the wind farm;

[0007] Acquire and process the load data of the power grid to obtain the power grid load, monitor the operation status of the power grid in real time, and determine whether there is any abnormality in the power grid and the type of abnormality through the abnormality feedback mechanism;

[0008] Each particle in the particle swarm represents a monitoring point in the power grid. The particle parameters of each particle are generated based on the wind power generation and grid load. The abnormal state of the wind farm and the abnormal type of the power grid are combined to generate a scheduling strategy, outputting the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point.

[0009] Monitor the actual implementation of wind power access ratio, energy storage dispatch sequence and load fluctuation threshold at each monitoring point, as well as the operating status of wind farms and power grids, obtain feedback results, and adjust the dispatch strategy based on the feedback results.

[0010] As an optional implementation, the generation logic of the scheduling policy includes:

[0011] A set of particle swarms is randomly generated, each particle represents a monitoring point of the power grid;

[0012] Generate particle parameters for each particle based on wind power generation and grid load;

[0013] According to the abnormal state of the wind farm and the abnormal type of the power grid, dynamic constraints are imposed on the particle parameters;

[0014] Calculate the fitness value of each particle and update the speed and position of each particle;

[0015] Until the average fitness value of the particle swarm reaches the fitness threshold, the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point are output.

[0016] As an optional implementation manner, the particle parameters of each particle include the wind power output ratio of each monitoring point, the energy storage response weight and the load fluctuation tolerance threshold.

[0017] As an optional implementation, the logic for applying dynamic constraints to particle parameters includes:

[0018] If the wind turbine is in a fault shutdown or maintenance shutdown, the wind power output ratio of the associated monitoring point is set to 0;

[0019] If there is a global anomaly in the power grid, the load fluctuation tolerance threshold of all monitoring points is set as the safety threshold.

[0020] As an optional implementation manner, the fitness value of each particle is composed of a weighted combination of wind power output deviation, voltage deviation and frequency deviation.

[0021] As an optional implementation, the updating logic of the velocity of each particle includes:

[0022] Dynamically adjust the inertia weight according to the number of abnormal monitoring points;

[0023] When there is a global anomaly in the power grid, the historical optimal solutions of neighboring particles are tracked to update the speed of each particle;

[0024] When there is a local anomaly in the power grid, the speed of each particle is updated through a hybrid guidance of individual historical optimal and global optimal.

[0025] As an optional implementation manner, the logic for obtaining wind power generation includes:

[0026] Determine the number of wind turbines in the wind farm and obtain the wind speed of each wind turbine;

[0027] Determine the power of each wind turbine generator set based on its wind speed;

[0028] The power of each wind turbine is accumulated to obtain the power of the wind farm;

[0029] The wind power generation is obtained by time-integrating the power of the wind farm;

[0030] And the wind power generation is corrected according to the operating status of the wind farm.

[0031] As an optional implementation manner, the power judgment logic of each wind turbine generator set includes:

[0032] Configure the starting wind speed and rated wind speed, compare the wind speed of each wind turbine with the starting wind speed and rated wind speed to obtain the operating mode of each wind turbine, and calculate the power of each wind turbine according to the operating mode of each wind turbine.

[0033] As an optional implementation, the abnormality feedback mechanism includes:

[0034] Layout monitoring points of the power grid, obtain sensor data from each monitoring point and determine whether there is any abnormality at each monitoring point. If there is no abnormality at each monitoring point, it means that there is no abnormality in the power grid;

[0035] If there is an abnormality at a monitoring point, compare the adjacent monitoring points to see if there is an abnormality.

[0036] If at most one adjacent monitoring point has an anomaly, it indicates that there is a local anomaly in the power grid;

[0037] If there are abnormalities at least at two adjacent monitoring points, verify the abnormal fluctuations of the monitoring point and the adjacent monitoring points;

[0038] If the abnormal fluctuation of the monitoring point is consistent with the abnormal fluctuation of the adjacent monitoring points, it means that there is a global abnormality in the power grid;

[0039] If the abnormal fluctuation of the monitoring point is inconsistent with the abnormal fluctuation of the adjacent monitoring points, it means that there is a local abnormality in the power grid.

[0040] Compared with the prior art, the present application has the following beneficial effects: by acquiring and processing wind power data of a wind farm, wind power generation is obtained, and the operating status of the wind farm is monitored in real time to determine whether there is any abnormality in the wind farm; by acquiring and processing load data of the power grid, grid load is obtained, and the operating status of the power grid is monitored in real time, and whether there is any abnormality in the power grid and the type of abnormality in the power grid are determined through an abnormality feedback mechanism; each particle in a particle swarm represents a monitoring point of the power grid, particle parameters of each particle are generated based on wind power generation and grid load, and a scheduling strategy is generated based on the abnormal status of the wind farm and the abnormal type of the power grid, and the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point are output, which can maximize the utilization of wind energy resources and reduce energy waste; the actual implementation of the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point, as well as the operating status of the wind farm and the power grid, is monitored to obtain feedback results, and the scheduling strategy is adjusted according to the feedback results, which helps to reduce the operating cost of the microgrid, improve the operating efficiency and economy of the microgrid, and increase revenue; and timely measures can be taken to avoid power grid failures and ensure the stability of power supply. The dynamic scheduling method can provide a more flexible and reliable power supply according to the changes in load demand to meet the needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be derived from these drawings without inventive work. Among them:

[0042] Figure 1 A flow chart of a method for dynamically dispatching breeze power generation to a smart grid provided in an embodiment of the present application;

[0043] Figure 2 A logic diagram for obtaining wind power generation in a dynamic dispatching method for connecting breeze power generation to a smart grid provided in an embodiment of the present application;

[0044] Figure 3A flowchart of the abnormal feedback mechanism of the dynamic scheduling method for connecting breeze power generation to the smart grid provided in an embodiment of the present application;

[0045] Figure 4 A scheduling strategy generation logic diagram for the dynamic scheduling method for connecting breeze power generation to a smart grid provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the objectives, technical solutions and advantages of the embodiments of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application are clearly and completely described below in conjunction with the drawings in the specification. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0047] Example

[0048] like Figure 1 As shown, a flow chart of a method for dynamically scheduling a breeze power generation system connected to a smart grid is provided in an embodiment of the present application. The method includes:

[0049] S1. Acquire and process wind power data of the wind farm to obtain wind power generation, monitor the operating status of the wind farm in real time, and determine whether there is any abnormality in the wind farm.

[0050] The logic for obtaining wind power generation is as follows: Figure 2 As shown, specifically including:

[0051] Determine the number of wind turbines in the wind farm, obtain the wind speed of each wind turbine, determine the power of each wind turbine based on the wind speed of each wind turbine, accumulate the power of each wind turbine to obtain the power of the wind farm, perform time integration on the power of the wind farm to obtain the wind power generation, and correct the wind power generation according to the operating status of the wind farm.

[0052] In reality, a wind farm usually contains multiple wind turbines. It is necessary to obtain the wind speed of each wind turbine in the wind farm area. Usually, the wind speed can be obtained through on-site wind speed measurement. The wind speed of each wind turbine is usually sampled in time units, such as once per hour or once per minute.

[0053] When the wind speed is known, the wind power generation is calculated according to the time step, where the time step is, for example, per hour or per minute. The wind power generation is the total power generation of the wind farm, for example, per time period The power of the wind farm is , then the wind power generation can be approximately calculated by the following integral: ,in is the wind power generation, and Represent different time points, and time periods .

[0054] The power judgment logic of each wind turbine includes:

[0055] Configure the starting wind speed and rated wind speed, compare the wind speed of each wind turbine with the starting wind speed and rated wind speed to obtain the operating mode of each wind turbine, and calculate the power of each wind turbine according to the operating mode of each wind turbine.

[0056] The function expression for calculating the power of each wind turbine is as follows:

[0057] ;

[0058] Where, Indicates the The power of a wind turbine, Indicates the The wind speed of each wind turbine, Indicates the starting wind speed. represents the air density, Indicates the wind swept area of ​​the wind turbine. represents the power factor, Indicates the rated wind speed, Indicates rated power.

[0059] It should be noted that: starting wind speed It is used when the wind speed of the wind turbine is less than the starting wind speed value and the wind turbine does not generate electricity; air density The usual value is 1.225kg / m 3 ;Swept area of ​​wind turbine It is through the formula A=πr 2 Calculated, where is the blade length of the wind turbine; power coefficient Usually refers to a constant within the wind speed range, with a theoretical maximum value of 0.59; rated wind speed It is used to ensure that the power output of the wind turbine is maintained at the rated power level when the wind speed is greater than the rated wind speed.

[0060] When the wind speed of the wind turbine is between the starting wind speed and the rated wind speed (including the starting wind speed and the rated wind speed), the power of the wind turbine is in a cubic relationship with the wind speed (i.e. ) reflects the nonlinear relationship between wind speed and power.

[0061] When the wind speed of the wind turbine is lower than the starting wind speed, the operation mode of the wind turbine is the stop mode. The stop mode means that when the wind speed is lower than the starting wind speed, the wind turbine cannot provide enough power to start power generation, so it will stop running. When the wind speed of the wind turbine is between the starting wind speed and the rated wind speed (including the starting wind speed and the rated wind speed), the operation mode of the wind turbine is the start mode. The start mode usually refers to the state where the wind turbine has just started to generate electricity. At this stage, the power of the wind turbine gradually increases with the increase of wind speed, which is usually calculated using the above calculation formula. When the wind turbine's When the wind speed is greater than the rated wind speed, the operating mode of the wind turbine is the limiting mode. The limiting mode means that after the wind speed reaches the rated wind speed and exceeds the rated wind speed, the power output of the wind turbine enters a plateau period and no longer increases with the increase in wind speed. At this time, the power output of the wind turbine is limited to the rated power to protect the wind turbine from overload operation; at the same time, a safety limit wind speed for the wind turbine will be set in real life. When the wind speed is greater than the safety limit wind speed, the wind turbine will stop running to avoid damage and protect the safety of the wind turbine. This is used to protect wind turbines under extreme weather conditions.

[0062] It should be understood that in the calculation of wind farm power generation, the operating status of the wind turbines has a significant impact on the actual wind power generation. If this factor is taken into account, the revised wind power generation can more accurately reflect the actual wind power generation.

[0063] Real-time monitoring of the operating status of the wind farm to determine whether there are any abnormalities in the wind farm. For example, the operating status of the wind farm can be divided into the following situations, including normal operation, fault shutdown, maintenance shutdown and inefficient operation. Normal operation means that the wind turbine is in good working condition and can output the corresponding power according to the design parameters. Fault shutdown means that the wind turbine is shut down due to a fault and cannot generate electricity. The power output needs to be set to zero. Maintenance shutdown means that the wind turbine is shut down due to regular maintenance. There is usually a scheduled downtime, and the wind power generation needs to be adjusted during this period. Inefficient operation means that the wind turbine has a fault but can still operate, but the power output of the wind turbine is low at this time. The correction factor needs to be adjusted according to the actual power loss. Except for normal operation, others (fault shutdown, maintenance shutdown and inefficient operation) all indicate that there is an abnormality in the wind farm. When correcting the wind power generation, the wind power generation of each wind turbine needs to be adjusted according to the operating status of the wind turbine. If some wind turbines are shut down due to fault or maintenance, the power of these wind turbines should be considered to be 0 when correcting the wind power generation, such as the two operating statuses of wind turbines: fault shutdown and maintenance shutdown. Alternatively, the wind power generation can be corrected by the correction factor, such as in the operating status of inefficient operation. The correction factor is determined according to the fault and maintenance status of the wind turbine.

[0064] S2. Obtain and process the load data of the power grid to obtain the power grid load, and monitor the operating status of the power grid in real time. Through the abnormal feedback mechanism, determine whether there is any abnormality in the power grid and the type of abnormality of the power grid.

[0065] It should be understood that the load data of the power grid refers to the sum of the electricity demand of all users in the power grid (including industry, commerce and households). The load data of the power grid is usually obtained by various substations, distribution networks or power metering equipment. At the same time, the power grid load refers to the total power required by the power grid in a certain period of time, that is, the sum of the loads of all users.

[0066] Abnormal feedback mechanism such as Figure 3 As shown, specifically including:

[0067] Layout monitoring points of the power grid, obtain sensor data from each monitoring point and determine whether there is any abnormality at each monitoring point. If there is no abnormality at each monitoring point, it means that there is no abnormality in the power grid;

[0068] If there is an abnormality at a certain monitoring point, compare the adjacent monitoring points to see if there is an abnormality;

[0069] If at most one adjacent monitoring point has an anomaly, it indicates that there is a local anomaly in the power grid;

[0070] If there are abnormalities at least at two adjacent monitoring points, verify the abnormal fluctuations of the monitoring point and the adjacent monitoring points;

[0071] If the abnormal fluctuation of the monitoring point is consistent with the abnormal fluctuation of the adjacent monitoring points, it means that there is a global abnormality in the power grid;

[0072] If the abnormal fluctuation of the monitoring point is inconsistent with the abnormal fluctuation of the adjacent monitoring points, it means that there is a local abnormality in the power grid.

[0073] Key areas of the power grid (such as substations, transmission lines, distribution networks, and user terminals) are selected as monitoring points for monitoring. Each monitoring point is equipped with sensors to cover the entire power grid. After the monitoring points are laid out, the sensor data from each monitoring point needs to be collected to determine whether any anomalies exist at each point. This is achieved through simple threshold judgment. For example, a threshold range for normal sensor data operation is configured for each monitoring point. This threshold range includes the normal fluctuation range of current, voltage, power, and temperature data, or whether these fluctuations exceed a preset tolerance range. If the sensor data of a monitoring point exceeds the threshold range, it indicates that there is an anomaly in the power grid. The monitoring point where the sensor data exceeds the threshold is then compared with its neighboring monitoring points to ensure consistency of sensor data from multiple monitoring points. If two or more adjacent monitoring points show abnormal sensor data, including abnormal current, voltage, power, or temperature data, a preliminary judgment can be made whether there is a systemic fault. This reduces the probability of false alarms from a single monitoring point. Otherwise, the abnormality of the monitoring point is considered separately and a local anomaly is determined in the power grid. Global anomaly and local anomaly refer to the types of power grid anomalies.

[0074] The abnormal fluctuations of the monitoring points where abnormal sensor data appear are compared, including whether the amplitudes of abnormal fluctuations in current, voltage, power and temperature data are similar, or whether the time points of abnormal occurrence are similar, or whether the types of abnormalities are consistent, such as excessive voltage and excessive current. When the abnormal fluctuations are consistent, it will indicate that there is a global abnormality in the power grid. For example, if the voltage of this monitoring point fluctuates abnormally, and the voltage of the adjacent monitoring point also fluctuates abnormally, or the time of abnormal occurrence is consistent, the global abnormality means that a certain area of ​​the power grid has a widespread fault or a precursor to a fault. When the abnormal fluctuations are inconsistent, it will indicate that there is a local abnormality in the power grid. For example, if the voltage of this monitoring point fluctuates abnormally, but the voltage of the adjacent monitoring point does not fluctuate abnormally, but the temperature fluctuates abnormally, or the time deviation of the abnormal occurrence is large, the local abnormality means that the problem is limited to a local problem such as a transformer failure, poor contact of the transmission line and load mutation.

[0075] When there is an abnormality in the power grid, an alarm signal will be issued. Even more, when it indicates that there is a global abnormality in the power grid, the power grid protection mechanism will be activated, such as triggering automatic circuit breakers, cutting off the fault area, adjusting the power grid load, allocating other power grid resources to ensure power supply, and starting the backup power supply or enabling energy storage equipment; when it indicates that there is a local abnormality in the power grid, the grid load will be automatically adjusted or the power generation power will be adjusted to reduce the impact on the power grid, and the maintenance or detection process will be started to inspect and repair the fault monitoring point.

[0076] The above-mentioned abnormal feedback mechanism can effectively reduce misjudgments or missed judgments caused by a single monitoring point, achieve efficient and accurate grid abnormality monitoring, effectively improve the detection and response speed of grid faults, and ensure the stability and reliability of the grid.

[0077] S3. Each particle in the particle swarm represents a monitoring point of the power grid. The particle parameters of each particle are generated based on the wind power generation and the power grid load. The abnormal state of the wind farm and the abnormal type of the power grid are combined to generate a scheduling strategy, and the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point are output.

[0078] like Figure 4 As shown, the generation logic of the scheduling policy includes:

[0079] A set of particle swarms is randomly generated, each particle represents a monitoring point of the power grid;

[0080] Generate particle parameters for each particle based on wind power generation and grid load;

[0081] According to the abnormal state of the wind farm and the abnormal type of the power grid, dynamic constraints are imposed on the particle parameters;

[0082] Calculate the fitness value of each particle and update the speed and position of each particle;

[0083] Until the average fitness value of the particle swarm reaches the fitness threshold, the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point are output.

[0084] The particle parameters of each particle include the wind power output ratio of each monitoring point, the energy storage response weight and the load fluctuation tolerance threshold.

[0085] The monitoring points are abstracted as particles in order to use the particle swarm algorithm to uniformly dispatch and optimize the key locations of the power grid. Based on the setting of the power grid monitoring points in the above step S2, the optimal scheduling scheme is found through iterative search of the particle swarm to achieve stable operation of the power grid and efficient utilization of wind power. In the power grid, based on the relevant information of the layout of the power grid monitoring points in step S2, multiple key locations are selected as monitoring points, such as substations, transmission lines and user terminals. These locations are the key to obtaining power grid operation status data. The random number generation function of the programming language is used to give each particle an initial random identifier and position information. The position of each particle can be initially set to an initial state value related to the monitoring point, such as the initial voltage or current value, and the speed represents the state change trend of the monitoring point.

[0086] Taking the initial voltage of the monitoring point as an example, the position of the particle is randomly set between 0.95 and 1.05 times the rated voltage to simulate the uncertainty of the initial voltage state of the monitoring point, while the speed of the particle is randomly taken in the range of 0 to 1 to represent the initial uncertainty of the state change trend of the monitoring point; by randomly generating a particle swarm, it is possible to search for scheduling solutions extensively in the solution space to avoid falling into local optimal solutions. Each particle represents a monitoring point, which facilitates targeted scheduling and control of various parts of the power grid, echoing the layout of the power grid monitoring points in step S2, and laying the foundation for the subsequent generation of scheduling strategies based on the status of the monitoring points.

[0087] The initialization logic of the particle parameters includes: if there is no abnormality in the wind farm, the initial value of the wind power output ratio of all monitoring points is the theoretical maximum output; if there is a local abnormality in the power grid, the initial value of the energy storage response weight of the abnormal monitoring point is set to 0, and the load fluctuation tolerance threshold is automatically tightened according to the data of adjacent monitoring points.

[0088] Wind power generation and grid load are important factors affecting grid dispatch. By generating particle parameters based on these two factors, the dispatch plan can better adapt to the actual operation of the grid, improve the utilization rate of wind power, and balance the grid load. The key data in steps S1 and S2 are combined with the dispatch parameters of the grid monitoring points. Based on the wind power generation obtained and processed in step S1 and the grid load obtained and processed in step S2, the particle parameters are calculated for each particle (monitoring point). The particle parameters include the wind power output ratio, energy storage response weight, and load fluctuation tolerance threshold of each monitoring point.

[0089] The theoretical maximum output refers to the theoretical wind power generation obtained by allocating the corrected wind power generation according to the ratio of the rated power of the wind turbine corresponding to each monitoring point to the total rated power of the wind farm; the wind power output ratio refers to the proportion of wind power output based on the theoretical maximum output at a certain monitoring point after taking into account various factors, including the wind power resources near the monitoring point, the grid load, and abnormal conditions of the wind farm and the grid; the energy storage response weight refers to the relative importance and participation of the energy storage system in responding to the power regulation needs of the grid. The energy storage response weight at different monitoring points reflects the The response strength of the energy storage equipment at the measuring point to regulating grid power and balancing electricity supply and demand. For example, when the energy storage response weight of a monitoring point is set high (such as 0.8), it means that the energy storage system in the area where the monitoring point is located is given greater regulation capability in the scheduling strategy. When the power grid fluctuates or wind power output is unstable, the energy storage system will more actively participate in the regulation process. For example, it will store electricity when wind power generation is in excess and release electricity when the power supply is insufficient to maintain the stable operation of the grid. Conversely, when the energy storage response weight of a monitoring point is low (such as 0.2), the regulation effect of the energy storage system in the area where the monitoring point is located is relatively weak.

[0090] The load fluctuation tolerance threshold (LFTT) is a numerical standard that measures the degree of load fluctuation that the power grid can withstand at a monitoring point. Different LFTTs are set for different monitoring points, depending on the power consumption characteristics and power supply stability requirements of their areas. For example, for important user-end monitoring points with extremely high power supply stability requirements, the LFTT can be set relatively low (e.g., 0.05, indicating a load fluctuation range of ±5%). Once the load fluctuation falls outside this range, the dispatching strategy needs to be adjusted accordingly, such as by adjusting power generation capacity or deploying energy storage resources to stabilize the load. For monitoring points in areas with relatively low power supply stability requirements, the LFTT can be appropriately relaxed (e.g., 0.1, indicating a load fluctuation range of ±10%). In actual dispatching, if the load fluctuation at a monitoring point exceeds its LFTT, the dispatching strategy will be triggered to rebalance power supply and demand in the area where the monitoring point is located to ensure safe and stable operation of the power grid.

[0091] Among them, when there is no abnormality in the wind farm, the initial value of the wind power output ratio of all monitoring points is the theoretical output maximum value, and will subsequently change around the theoretical output maximum value according to various factors. For example, if the wind power resources near a monitoring point are abundant and the grid load is relatively low, the wind power output ratio of the monitoring point can be set higher. The abundant wind power resources can be judged based on the power distribution and wind power generation of the wind farm in step S1 and the wind power generation threshold, and the relatively low grid load can be obtained by comparing the grid load and the grid load threshold in step S2; if the load fluctuation in the area where the monitoring point is located is large, the load fluctuation tolerance threshold can be appropriately relaxed. The large load fluctuation can be combined with the monitoring of the grid operation status in step S2 to judge the load fluctuation, such as by comparing with the set fluctuation threshold.

[0092] Specifically, based on the proportional relationship between wind power generation and grid load, the theoretical maximum output value of each monitoring point is calculated, and the wind power output ratio is determined based on historical data. For example, when the initial state value of the monitoring point (taking voltage as an example) is greater than 1.02 times the rated voltage, combined with the wind farm power distribution and wind power generation in step S1, it is judged that the wind power resources at the monitoring point are abundant, and the grid load obtained in step S2 is relatively low, then the wind power output ratio of the monitoring point is set to a higher value, such as 0.8; the energy storage response weight is first uniformly set to an empirical value, such as 0.5; and the load fluctuation tolerance threshold needs to refer to the historical data of the load fluctuation of the monitoring point in step S2. If the historical load fluctuation is greater than the fluctuation threshold, the load fluctuation tolerance threshold of the monitoring point is appropriately relaxed; thereby making the scheduling plan more in line with the actual operation of the power grid, improving the utilization rate of wind power, and reducing the load fluctuation of the power grid. By reasonably setting the particle parameters, the relationship between wind power, energy storage and load can be better coordinated to achieve stable operation of the power grid. The data in steps S1 and S2 are effectively applied to the generation of scheduling parameters, providing reasonable initial values ​​for subsequent adjustment and optimization of the scheduling strategy.

[0093] The logic of imposing dynamic constraints on particle parameters includes: if the wind turbine is in a fault shutdown or maintenance shutdown, the wind power output ratio of the associated monitoring point is set to 0; if there is a global abnormality in the power grid, the load fluctuation tolerance threshold of all monitoring points is set to a safety threshold.

[0094] The logic of imposing dynamic constraints on particle parameters also includes: when there is a global anomaly in the power grid, the speed parameters of the particle swarm are triggered to be reset. The range of the speed parameter reset includes the inertia weight and acceleration constant; when there is a local anomaly in the power grid, the maximum speed of the particles at the abnormal monitoring point is limited.

[0095] The abnormal state of wind farms and the abnormal type of power grid will have a serious impact on the safe and stable operation of the power grid. By imposing dynamic constraints on particle parameters, the scheduling plan can be adjusted in time to avoid further deterioration of abnormal conditions and ensure the safe operation of the power grid. Real-time scheduling strategy adjustments are made based on the abnormal monitoring results in steps S1 and S2.

[0096] If the wind turbine is in a fault shutdown or maintenance shutdown, the wind power output ratio of the associated monitoring point is set to 0. The fault shutdown and maintenance shutdown are determined based on the real-time monitoring of the wind farm operating status in step S1. For example, when a wind turbine in a wind farm fails and shuts down, the wind power output ratio of the monitoring point near the wind farm is immediately adjusted to 0 through the association between the wind turbine and the monitoring point established in step S1. If there is a global anomaly in the power grid, the load fluctuation tolerance threshold of all monitoring points is set to a safety threshold. The global anomaly in the power grid is determined based on the anomaly feedback mechanism in step S2. For example, when a global anomaly such as a large-scale power outage occurs in the power grid, the load fluctuation tolerance threshold of all monitoring points is adjusted to a pre-set safety threshold.

[0097] At the same time, when there is a global abnormality in the power grid, the speed parameter reset of the particle swarm is triggered. The range of speed parameter reset includes inertia weight and acceleration constant. The inertia weight is set to the default value (for example, 0.7) and the acceleration constant is restored to the initial setting value (for example, 1.4). When there is a local abnormality in the power grid, the abnormal monitoring points are determined and the speed of the particles at these abnormal monitoring points is limited to a certain range, such as setting the maximum speed to 50% of the normal situation. This enhances the robustness and adaptability of the scheduling scheme. When abnormal conditions occur in wind farms and power grids, the scheduling strategy can be adjusted in time to ensure the safe and stable operation of the power grid. Through dynamic constraints, it can avoid abnormal conditions from causing greater damage to the power grid. It is closely integrated with the abnormal monitoring in steps S1 and S2 to achieve real-time response and adjustment. The particle parameters after applying dynamic constraints will be used to calculate the fitness value. The constraint conditions will affect the fitness evaluation of the particles, thereby guiding the particle swarm to search for a better scheduling scheme. The parameters after dynamic constraints make the subsequent fitness calculation more in line with the scheduling needs under the actual abnormal state of the power grid.

[0098] The fitness value of each particle is composed of the weighted wind power output deviation, voltage deviation and frequency deviation.

[0099] The calculation formula of the fitness value of each particle is as follows:

[0100] ;

[0101] Where, Indicates the The fitness value of a particle, Represents the weight coefficient of wind power output deviation, Indicates the The actual wind power output of particles, Indicates the The wind power output ratio of particles, Indicates the The theoretical maximum output of a particle is Represents the weight coefficient of voltage deviation, Indicates the The voltage deviation of each particle, represents the weight coefficient of frequency deviation, Indicates the The frequency deviation of each particle.

[0102] It should be noted that: wind power output deviation refers to the difference between the actual wind power output and the theoretical maximum output in the particle (monitoring point); voltage deviation refers to the voltage deviation of the particle (monitoring point); frequency deviation refers to the grid frequency deviation in the particle (monitoring point); 、 and The value range of is between 0 and 1, and .

[0103] The update logic for each particle's velocity includes:

[0104] Dynamically adjust the inertia weight according to the number of abnormal monitoring points;

[0105] When there is a global anomaly in the power grid, the historical optimal solutions of neighboring particles are tracked to update the speed of each particle;

[0106] When there are local anomalies in the power grid, the speed of each particle is updated through a hybrid guidance of individual historical optimal and global optimal.

[0107] The fitness value is used to evaluate the quality of the scheduling scheme represented by each particle. By calculating the fitness value and updating the speed and position of the particle, the particle swarm can continuously converge to a better scheduling scheme, and finally find the optimal scheduling scheme. The quality of the scheduling scheme is evaluated by comprehensively considering various data in steps S1 and S2. According to the fitness value calculation formula, the fitness value of each particle is calculated. Finally, the fitness value of each particle is composed of the weighted wind power output deviation, voltage deviation and frequency deviation. In the calculation process, the actual wind power output refers to the wind power generation of the wind turbine corresponding to the monitoring point calculated in step S1. The voltage deviation of the monitoring point can be combined with the voltage sensor data of the monitoring point in step S2. The grid frequency deviation is also obtained based on the monitoring of the grid operation status in step S2. The wind power output deviation, voltage deviation and frequency deviation are multiplied by their respective weight coefficients and added together to obtain the fitness value of the particle swarm.

[0108] Then the speed and position of each particle are updated according to the fitness value of the particle. When the speed of the particle is updated, the inertia weight is dynamically adjusted according to the number of abnormal monitoring points obtained in the S2 step abnormal feedback mechanism. For each additional abnormal monitoring point, the inertia weight is reduced by 0.1. When there is a global anomaly in the power grid, each particle tracks the historical optimal solution of the adjacent particle to update the speed. The relationship between adjacent particles is set in advance according to the physical location of the monitoring point. When there is a local anomaly in the power grid, the particle updates the speed by mixing the individual historical optimal solution and the global optimal solution. The individual historical optimal solution is the optimal position reached by the particle itself during the iteration process. The global optimal solution is the optimal position found by the entire particle swarm at present. Finally, the current position of the particle is added to the speed value to obtain a new position according to the updated speed. By continuously updating the speed and position of the particle, the particle swarm can gradually converge to the optimal scheduling scheme, improving the quality and efficiency of the scheduling scheme. The calculation of the fitness value takes into account multiple factors, which can comprehensively evaluate the pros and cons of the scheduling scheme. Combined with the data in S1 and S2 steps, the scheduling scheme is more in line with the actual operation.

[0109] The fitness threshold is used to determine whether the particle swarm has converged to a relatively satisfactory scheduling scheme. When the average fitness value of the particle swarm reaches the fitness threshold, it means that the particle swarm has found a relatively optimal scheduling scheme, and the result can be output for actual power grid scheduling, providing initial scheduling scheme data for S4 step. After each iteration calculation, check whether the average fitness value of the particle swarm is less than or equal to the pre-set fitness threshold. The average fitness value of the particle swarm is obtained by averaging the sum of the fitness values of all particles. If the average fitness value of the particle swarm is less than or equal to the fitness threshold, the wind power integration ratio of each monitoring point, the energy storage scheduling sequence and the load fluctuation threshold are output as the final scheduling result. These results will be used to monitor the actual execution of the scheduling result in S4 step, which is closely related to the monitoring content in S4 step. Thus, the optimal scheduling scheme can be output in time to provide guidance for the actual operation of the power grid. By setting the fitness threshold, unnecessary iterations can be avoided, improving the scheduling efficiency. The effective initial scheduling scheme provided for monitoring and adjustment in S4 step ensures the efficiency and accuracy of power grid scheduling.

[0110] The wind power access ratio refers to the share or proportion of wind power in the entire power grid, which is usually expressed as the ratio of wind power access capacity to the total power grid capacity. It can also be measured by the ratio of actual wind power output power to the total load demand of the power grid. This ratio reflects the importance and impact of wind power in the power grid. A higher wind power access ratio means that the power grid is more dependent on wind power. The volatility and uncertainty of wind power will have a greater impact on the operation of the power grid, which will bring challenges to the stability, reliability and power quality of the power grid. Corresponding measures need to be taken in power grid planning, operation and scheduling to deal with it. The energy storage scheduling sequence refers to the order and operation arrangement of scheduling and control of the energy storage system during the operation of the power grid. It stipulates the order at different time nodes and different power grid operation states. , what operations should the energy storage system perform, such as charging, discharging or maintaining standby status, as well as the corresponding power size, etc. A reasonable energy storage scheduling sequence can effectively utilize the regulation capability of the energy storage system, play a role in smoothing wind power fluctuations, optimizing grid power balance, and improving grid operation economy. By formulating a scientific energy storage scheduling sequence, the energy storage system can be better matched with wind power, other power sources and load demands, thereby improving the overall operation efficiency and stability of the grid; the load fluctuation threshold refers to the boundary value of the load fluctuation range allowed in the power system. It is a set value used to judge whether the load fluctuation exceeds the range that the grid can bear or allow. When the load fluctuation is greater than this load fluctuation threshold, the grid will take corresponding control measures to adjust to ensure the stable operation of the grid.

[0111] S4. Monitor the wind power access ratio, energy storage dispatch sequence, and actual implementation of load fluctuation thresholds at each monitoring point, as well as the operating status of the wind farm and power grid, obtain feedback results, and adjust the dispatch strategy based on the feedback results.

[0112] During the execution of the dispatching strategy, real-time monitoring of wind power generation, grid load, grid frequency, grid voltage, and the operating status of wind farms and grids can determine the actual effect of the dispatching results in real time and provide data support for subsequent adjustments.

[0113] The feedback results include the matching degree between wind power generation and grid load, grid stability and operating status results. Specifically, the matching degree between wind power generation and grid load includes: if wind power generation is lower than expected, it is necessary to call on traditional energy generation or mobilize energy storage equipment to meet the grid load demand; if wind power generation is too high, the grid load will not be able to fully absorb the wind power output, resulting in waste or grid instability, and the excess wind power needs to be directed to the energy storage system or the wind power output needs to be adjusted; and grid stability specifically includes: if the grid frequency fluctuates too much (such as exceeding the preset range), it means that the load regulation of the grid has failed to effectively match the grid load. To cope with wind power fluctuations, the load response of the power grid needs to be adjusted. If the voltage fluctuation is too large and affects the stability of the power grid, it is necessary to restore the stability of the power grid voltage by adjusting the power output of the generator set, adjusting the output of wind power, or using regulating equipment. The operating status results specifically include: if a wind turbine fails, resulting in reduced or complete cessation of power generation capacity, it is necessary to adjust the scheduling strategy, reduce reliance on the wind turbine, and increase other power generation resources (such as traditional generators or other wind turbines). If the power grid fails or some monitoring points have abnormalities, it will cause problems in load scheduling, and it is necessary to adjust the output of wind power or strengthen the regulation of the power grid load.

[0114] By adjusting the dispatching strategy according to the feedback results and through iterative optimization, the error can be gradually reduced and the coordinated efficiency and stability of wind power and the power grid can be improved.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application, and all of these should be included in the scope of the claims of the present application.

Claims

1. A dynamic dispatching method for connecting breeze power generation to a smart grid, characterized in that: include: Acquire and process wind power data from wind farms to obtain wind power generation, monitor the operating status of wind farms in real time, and determine whether there are any abnormalities in the wind farms; Acquire and process the load data of the power grid to obtain the power grid load, monitor the operation status of the power grid in real time, and determine whether there is any abnormality in the power grid and the type of abnormality through the abnormality feedback mechanism; Each particle in the particle swarm represents a monitoring point in the power grid. The particle parameters of each particle are generated based on the wind power generation and grid load. The abnormal state of the wind farm and the abnormal type of the power grid are combined to generate a scheduling strategy, outputting the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point. The generation logic of the scheduling strategy includes: A set of particle swarms is randomly generated, each particle represents a monitoring point of the power grid; Generate particle parameters for each particle based on wind power generation and grid load; According to the abnormal state of the wind farm and the abnormal type of the power grid, dynamic constraints are imposed on the particle parameters; Calculate the fitness value of each particle and update the speed and position of each particle; Until the average fitness value of the particle swarm reaches the fitness threshold, the wind power access ratio, energy storage scheduling sequence and load fluctuation threshold of each monitoring point are output; The particle parameters of each particle include the wind power output ratio, energy storage response weight and load fluctuation tolerance threshold of each monitoring point; The load fluctuation threshold refers to the limit of the load fluctuation range allowed in the power system; the energy storage response weight refers to the relative importance and participation of the energy storage system in responding to the power regulation needs of the power grid; the load fluctuation tolerance threshold is a numerical standard that measures the degree of load fluctuation that the power grid can withstand at the monitoring point; Monitor the actual implementation of wind power access ratio, energy storage dispatch sequence and load fluctuation threshold at each monitoring point, as well as the operating status of wind farms and power grids, obtain feedback results, and adjust the dispatch strategy based on the feedback results.

2. The dynamic dispatching method for connecting breeze power generation to a smart grid according to claim 1, characterized in that: The logic for applying dynamic constraints to particle parameters includes: If the wind turbine is in a fault shutdown or maintenance shutdown, the wind power output ratio of the associated monitoring point is set to 0; If there is a global anomaly in the power grid, the load fluctuation tolerance threshold of all monitoring points is set as the safety threshold.

3. The dynamic dispatching method for connecting breeze power generation to a smart grid according to claim 2, characterized in that: The fitness value of each particle is composed of a weighted combination of wind power output deviation, voltage deviation and frequency deviation.

4. The method for dynamically dispatching breeze power generation connected to a smart grid according to claim 3, characterized in that: The update logic of the speed of each particle includes: Dynamically adjust the inertia weight according to the number of abnormal monitoring points; When there is a global anomaly in the power grid, the historical optimal solutions of neighboring particles are tracked to update the speed of each particle; When there are local anomalies in the power grid, the speed of each particle is updated through a hybrid guidance of individual historical optimal and global optimal.

5. The dynamic dispatching method for connecting breeze power generation to a smart grid according to claim 4, characterized in that: The logic for obtaining wind power generation includes: Determine the number of wind turbines in the wind farm and obtain the wind speed of each wind turbine; Determine the power of each wind turbine generator set based on its wind speed; The power of each wind turbine is accumulated to obtain the power of the wind farm; The wind power generation is obtained by time-integrating the power of the wind farm; And the wind power generation is corrected according to the operating status of the wind farm.

6. The dynamic dispatching method for connecting breeze power generation to a smart grid according to claim 5, characterized in that: The power judgment logic of each wind turbine generator system includes: Configure the starting wind speed and the rated wind speed, compare the wind speed of each wind turbine with the starting wind speed and the rated wind speed to obtain the operating mode of each wind turbine, and calculate the power of each wind turbine according to the operating mode of each wind turbine.

7. The dynamic dispatching method for connecting breeze power generation to a smart grid according to claim 6, characterized in that: The abnormal feedback mechanism includes: Layout monitoring points of the power grid, obtain sensor data from each monitoring point and determine whether there is any abnormality at each monitoring point. If there is no abnormality at each monitoring point, it means that there is no abnormality in the power grid; If there is an abnormality at a monitoring point, compare the adjacent monitoring points to see if there is an abnormality. If at most one adjacent monitoring point has an anomaly, it indicates that there is a local anomaly in the power grid; If there are abnormalities at least at two adjacent monitoring points, verify the abnormal fluctuations of the monitoring point and the adjacent monitoring points; If the abnormal fluctuation of the monitoring point is consistent with the abnormal fluctuation of the adjacent monitoring points, it means that there is a global abnormality in the power grid; If the abnormal fluctuation of the monitoring point is inconsistent with the abnormal fluctuation of the adjacent monitoring points, it means that there is a local abnormality in the power grid.

Images