Primary frequency modulation control method and system based on collaborative optimization strategy

Abstract

The invention discloses a primary frequency modulation control method and system based on a collaborative optimization strategy, and the method comprises the steps: obtaining a system frequency deviation value of a power system, and triggering frequency modulation control when the system frequency deviation value exceeds a preset frequency deviation threshold value; obtaining and classifying various frequency modulation resources, and determining frequency modulation priorities of the various frequency modulation resources; based on the classified frequency modulation resources, a multi-target optimization model is constructed and solved, and the optimal output adjustment amount of each type of frequency modulation resources is obtained; converting the optimal output adjustment amount of each frequency modulation resource into a corresponding frequency modulation control instruction, and sending the frequency modulation control instruction to a corresponding control device; the system frequency is monitored in real time, and when the system frequency deviation value obtained in real time is within a preset frequency deviation threshold value, frequency modulation control is completed. The frequency of the power system is intelligently and accurately regulated and controlled, the frequency modulation response speed and the control effect are improved, the frequency modulation cost is reduced, and the stability and the reliability of the system are enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of frequency modulation control, and in particular relates to a primary frequency modulation control method and system based on a collaborative optimization strategy. Background Art

[0002] When the frequency deviation of the power system exceeds the set threshold, a frequency regulation control needs to be quickly started to restore the system frequency stability. However, different types of frequency regulation resources, such as photovoltaic power stations, thermal power units and wind farms, have different frequency regulation response speeds and frequency regulation capabilities. How to reasonably configure and coordinate the output of various frequency regulation resources according to the size and change trend of the frequency deviation to achieve fast, smooth and economical frequency regulation control is a technical problem that needs to be solved urgently.

[0003] At present, the output of photovoltaic power stations and wind farms is greatly affected by weather conditions, and there is great uncertainty in their available frequency regulation capacity and response speed; while the frequency regulation response of thermal power units is relatively stable, but their frequency regulation rate and cost are relatively high.

[0004] Therefore, it is urgent to propose a primary frequency modulation control method and system based on collaborative optimization strategy, which can reduce the frequency modulation cost and enhance the system stability. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a primary frequency modulation control method and system based on a collaborative optimization strategy to solve the problems existing in the above prior art.

[0006] To achieve the above object, the present invention provides a primary frequency modulation control method based on a collaborative optimization strategy, comprising the following steps:

[0007] Acquire a system frequency deviation value of the power system, and trigger frequency modulation control when the system frequency deviation value exceeds a preset frequency deviation threshold;

[0008] Acquire various frequency modulation resources, classify the frequency modulation resources, and determine the frequency modulation priorities of various frequency modulation resources;

[0009] Based on the classified frequency regulation resources, a multi-objective optimization model is constructed, with frequency regulation response speed, frequency regulation cost and output stability as optimization targets, and system frequency recovery time and upper and lower limits of frequency regulation resource output as constraints, to solve the multi-objective optimization model and obtain the optimal output adjustment of various frequency regulation resources;

[0010] Converting the optimal output adjustment amounts of the various frequency modulation resources into corresponding frequency modulation control instructions and sending them to corresponding control devices;

[0011] The system frequency is monitored in real time, and when the real-time system frequency deviation value is within the preset frequency deviation threshold, the frequency modulation control is completed.

[0012] Optionally, the system frequency of the power system is obtained, and when the system frequency exceeds a preset frequency threshold, the process of triggering frequency modulation control includes:

[0013] The real-time operating status data of the power system is obtained through sensors, and based on the obtained real-time operating status data, the real-time system frequency value is calculated using a Fourier transform algorithm, and the system frequency value is compared with the rated frequency value of the power system to obtain the system frequency deviation value at the current moment; a preset frequency deviation threshold is obtained from a pre-established frequency deviation threshold database, and the system frequency deviation value at the current moment is compared with the preset frequency deviation threshold. If the system frequency deviation value exceeds the frequency deviation threshold, the frequency modulation control process is triggered.

[0014] Optionally, the process of acquiring various types of frequency modulation resources, classifying the frequency modulation resources, and determining the frequency modulation priorities of the various types of frequency modulation resources includes:

[0015] Obtain frequency modulation resource parameters of various frequency modulation resources; construct a load forecasting model, predict the load demand of the system based on the load forecasting model, and obtain load forecasting data; obtain the power structure information of the system, integrate the frequency modulation resource parameters, load forecasting data and power structure information, input them into the fuzzy clustering algorithm, use the fuzzy clustering algorithm to reclassify various frequency modulation resources, and determine the frequency modulation priority of various frequency modulation resources;

[0016] Among them, the frequency modulation resource parameters include available output capacity and frequency modulation response speed; the power supply structure information includes the capacity and response speed of various power supplies.

[0017] Optionally, the process of reclassifying various frequency modulation resources by using a fuzzy clustering algorithm and determining the frequency modulation priorities of various frequency modulation resources includes:

[0018] The number of clustering categories is set, and the cluster centers of each category are randomly initialized. The Euclidean distance from each frequency modulation resource sample point to each cluster center is calculated, and the membership matrix is ​​updated based on the Euclidean distance. The cluster centers of each category are recalculated based on the updated membership matrix. When the cluster centers of each category meet the convergence conditions, the clustering is terminated to obtain the classification results of the frequency modulation resources, and the frequency modulation priority of each type of frequency modulation resource is determined based on the available output capacity and response speed of the cluster centers of each category.

[0019] Optionally, the process of solving the multi-objective optimization model to obtain the optimal output adjustment amount of various frequency regulation resources includes:

[0020] A particle swarm optimization algorithm is used to solve the multi-objective optimization model. During the algorithm iteration process, the output adjustment amount of the frequency modulation resources is updated based on the position information of each particle. When the particle swarm algorithm meets the convergence condition, the position information of the current optimal particle is output as the optimal output adjustment amount of each frequency modulation resource.

[0021] Optionally, the process of converting the optimal output adjustment amount of each type of frequency modulation resource into a corresponding frequency modulation control instruction and sending the instruction to a corresponding control device includes:

[0022] With the goal of optimizing the grid frequency recovery quality and frequency regulation cost, a game matrix is ​​constructed, and the Nash equilibrium point is solved based on the game matrix to obtain a game equilibrium result; the game equilibrium result is compared with the initial frequency regulation control instruction, the frequency regulation instruction correction amount is calculated, and the optimized frequency regulation control instruction is generated; the optimized frequency regulation control instruction is continuously iteratively optimized until the convergence condition of the frequency regulation control instruction is reached, and the finally obtained optimized frequency regulation control instruction is issued to the corresponding control device.

[0023] Optionally, after the frequency modulation control is completed, the following steps are further included:

[0024] According to the power system operation data and frequency regulation control records, the dynamic response characteristics of the frequency regulation resources are analyzed by using data mining algorithms to obtain the frequency regulation resource performance evaluation results; the frequency deviation and control time under different operating conditions are analyzed by using clustering algorithms, typical control effect modes are identified, and compared with the expected control targets to obtain the frequency regulation control effect evaluation results; based on the frequency regulation resource performance evaluation results and the frequency regulation control effect evaluation results, combined with expert experience and knowledge, a frequency regulation control strategy knowledge base is constructed;

[0025] Construct a simulation model of the power system and frequency control, design simulation experiments for different frequency control strategies in the frequency control strategy knowledge base, simulate various typical operating conditions, and obtain simulation results; based on the simulation results, adopt a multi-objective optimization algorithm, comprehensively consider the frequency regulation performance and frequency regulation control effect, optimize the frequency regulation control strategy, update and improve the frequency regulation control strategy knowledge base, and use the updated frequency regulation control strategy knowledge base to guide the frequency regulation control decision of the actual power system, so as to achieve continuous optimization of frequency regulation control.

[0026] The present invention also provides a primary frequency modulation control system based on a collaborative optimization strategy, which is used to implement a primary frequency modulation control method based on a collaborative optimization strategy, including: a frequency modulation control trigger module, a frequency modulation resource classification module, a resource output adjustment module and an equipment frequency modulation control module;

[0027] The frequency modulation control trigger module is used to obtain a system frequency deviation value of the power system, and when the system frequency deviation value exceeds a preset frequency deviation threshold, the frequency modulation control is triggered;

[0028] The frequency modulation resource classification module is used to obtain various types of frequency modulation resources, classify the frequency modulation resources, and determine the frequency modulation priority of various types of frequency modulation resources;

[0029] The resource output adjustment module is used to construct a multi-objective optimization model based on the classified frequency modulation resources, taking the frequency modulation response speed, frequency modulation cost and output stability as optimization targets, taking the system frequency recovery time and the upper and lower limits of the frequency modulation resource output as constraints, solving the multi-objective optimization model, and obtaining the optimal output adjustment amount of each type of frequency modulation resource;

[0030] The device frequency modulation control module is used to convert the optimal output adjustment amount of the various frequency modulation resources into corresponding frequency modulation control instructions and send them to the corresponding control device; and to monitor the system frequency in real time. When the real-time acquired system frequency deviation value is within the preset frequency deviation threshold, the frequency control is completed.

[0031] The present invention also provides an electronic device, comprising: a memory and a processor; the memory is used to store a program; the processor is used to execute the program to implement each step of the primary frequency modulation control method based on the collaborative optimization strategy.

[0032] The present invention also provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, each step of the primary frequency modulation control method based on the collaborative optimization strategy is implemented.

[0033] Compared with the prior art, the present invention has the following advantages and technical effects:

[0034] The frequency modulation control method proposed in the present invention first obtains the system frequency deviation value of the power system, and when the system frequency deviation value exceeds the preset frequency deviation threshold, the frequency modulation control is triggered; various types of frequency modulation resources are obtained for classification, and the frequency modulation priorities of various types of frequency modulation resources are determined; based on the classified frequency modulation resources, a multi-objective optimization model is constructed and solved to obtain the optimal output adjustment of various types of frequency modulation resources; the optimal output adjustment of various types of frequency modulation resources is converted into corresponding frequency modulation control instructions, and sent to the corresponding control device; the system frequency is monitored in real time, and when the system frequency deviation value obtained in real time is within the preset frequency deviation threshold, the frequency modulation control is completed; finally, the effectiveness of the strategy is verified through simulation, and used to guide actual control decisions. The present invention realizes the intelligent and precise control of the power system frequency, improves the frequency modulation response speed and control effect, reduces the frequency modulation cost, enhances the stability and reliability of the system, and provides effective guarantee for the safe and economic operation of the power grid under large-scale new energy access. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0036] Figure 1 It is a schematic diagram of the overall method flow of an embodiment of the present invention;

[0037] Figure 2 A schematic diagram of an optimization process of the frequency modulation resource output adjustment amount according to an embodiment of the present invention;

[0038] Figure 3 The figure is a flow chart of frequency modulation control according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0040] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0041] Embodiment 1

[0042] like Figure 1 As shown, this embodiment provides a primary frequency modulation control method based on a collaborative optimization strategy, comprising the following steps:

[0043] Obtaining the system frequency deviation value of the power system, and triggering frequency modulation control when the system frequency deviation value exceeds a preset frequency deviation threshold;

[0044] Obtain various frequency modulation resources, classify the frequency modulation resources, and determine the frequency modulation priority of various frequency modulation resources;

[0045] Based on the classified frequency regulation resources, a multi-objective optimization model is constructed, with frequency regulation response speed, frequency regulation cost and output stability as optimization targets, and system frequency recovery time and upper and lower limits of frequency regulation resource output as constraints, to solve the multi-objective optimization model and obtain the optimal output adjustment of various frequency regulation resources;

[0046] Convert the optimal output adjustment of various frequency modulation resources into corresponding frequency modulation control instructions and send them to the corresponding control devices;

[0047] The system frequency is monitored in real time, and when the real-time system frequency deviation value is within the preset frequency deviation threshold, the frequency modulation control is completed.

[0048] As a specific implementation method, the following steps are specifically included:

[0049] Step S101, acquiring the real-time operating status of the power system, obtaining a system frequency deviation value, comparing the frequency deviation value with a preset frequency deviation threshold, and triggering the frequency modulation control process if the frequency deviation value exceeds the threshold.

[0050] It is feasible to obtain the real-time operating status data of the power system through sensors, including parameters such as voltage, current, active power, reactive power, etc. Based on the real-time operating status data obtained, the real-time frequency value of the system is calculated using the Fourier transform algorithm. The calculated real-time frequency value is compared with the rated frequency value of the power system to obtain the system frequency deviation value at the current moment. The preset frequency deviation threshold is obtained from the pre-established frequency deviation threshold database, and the system frequency deviation value at the current moment is compared with the preset threshold. If the system frequency deviation value exceeds the preset threshold, it is determined that the frequency modulation control process needs to be triggered. If it does not exceed the preset threshold, the real-time operating status data continues to be obtained.

[0051] Specifically, the frequency control of the power system is the key to ensure the stable operation of the power grid. First, the system operation status data is obtained through sensors. For example, in a 500kV substation, the voltage sensor measures the bus voltage as 525kV, the current transformer measures the line current as 1000A, and the power calculation obtains the active power of 900MW and the reactive power of 300Mvar. After obtaining the data, the real-time frequency is calculated using the Fourier transform algorithm. The Fourier transform can convert the time domain signal into the frequency domain, thereby extracting the fundamental frequency. Assume that the system frequency calculated at a certain moment is 49.95Hz, and the frequency deviation is -0.05Hz compared with the rated frequency of 50Hz. Different levels of thresholds are pre-set in the frequency deviation threshold database. For example, the first-level warning threshold is ±0.1Hz, and the second-level warning threshold is ±0.2Hz. The current deviation of -0.05Hz does not exceed the first-level warning threshold, and the system continues to monitor without triggering frequency modulation. If the frequency deviation reaches -0.15Hz, exceeding the first-level warning threshold, the frequency modulation control process will be triggered. Real-time monitoring and rapid response can effectively prevent large-scale power outages caused by frequency out of control.

[0052] Step S102, obtaining the available output capacity and frequency regulation response speed parameters of various frequency regulation resources at the current moment, classifying the frequency regulation resources using a fuzzy clustering algorithm according to load forecast data and power structure information, and determining the frequency regulation priority of each type of resources.

[0053] It is feasible to obtain parameter information such as the available output capacity and frequency modulation response speed of various frequency modulation resources at the current moment, and use it as the input data of the fuzzy clustering algorithm. Through the load forecasting model, the load demand in the future period is predicted to obtain load forecast data. The power structure information of the current power grid is obtained, including parameters such as the capacity and response speed of various power sources. The frequency modulation resource parameters, load forecasting data and power structure information are integrated to construct the feature vector of the fuzzy clustering algorithm. The fuzzy clustering algorithm is used to classify the frequency modulation resources, and the resources are divided into different priority levels according to the available output capacity, response speed and other parameters of the resources.

[0054] Specifically, frequency regulation resource management is a key link in the frequency regulation of the power system. First, obtain real-time parameter information of various frequency regulation resources, such as the adjustable capacity of thermal power units, the water level and output of hydropower stations, and the charge state of energy storage power stations. These parameters reflect the frequency regulation capabilities and constraints of different resources. For example, the current output of a 600MW unit in a thermal power plant is 500MW, and there is still 100MW of upward adjustment space; a pumped storage power station is in power generation mode, and the remaining water can generate electricity for 2 hours. Load forecasting is the basis for the reasonable allocation of frequency regulation resources. Through historical data analysis, weather factors and other methods, the load curve for the next 24 hours is predicted. For example, it is predicted that there will be a peak in electricity consumption from 14:00 to 16:00 the next day, and the peak load can reach 95% of the maximum load of the system. This helps to arrange frequency regulation resources in advance to deal with possible frequency fluctuations. The power structure information reflects the overall frequency regulation capability of the system. For example, in a regional power grid, thermal power accounts for 60%, hydropower accounts for 25%, and new energy accounts for 15%. Thermal power units have good frequency regulation performance, but the response is slow; hydropower regulation is flexible, but limited by water conditions; new energy output fluctuates greatly and requires cooperation from other power sources. The above information is integrated to construct feature vectors, such as adjustable capacity, response speed, duration, economy and other dimensions. The fuzzy clustering algorithm is used to divide frequency regulation resources into categories such as fast response, continuous regulation, and economy. For example, energy storage systems and frequency regulation thermal power units are classified as fast response; conventional thermal power units and hydropower stations are classified as continuous regulation; and some interruptible loads are classified as economy. According to the clustering results and load forecasts, the frequency regulation power allocation ratio of each type of resource is determined. For example, in the stage of rapid load increase, fast response resources are preferentially called, and 60% of the frequency regulation tasks can be allocated; in the stage of stable load, the proportion of continuous regulation resources is increased to 50%. This allocation method can not only ensure the timeliness of frequency regulation, but also realize the economic utilization of resources. When the actual load exceeds the forecast range, it is necessary to dynamically adjust the resource output. For example, if the actual load in a certain period is 5% higher than the forecast, the output of fast-response resources can be increased first, such as increasing the discharge power of the energy storage system; at the same time, the participation of continuous regulation resources can be gradually increased, such as increasing the output of some thermal power units. This dynamic adjustment strategy can avoid excessive use of resources while ensuring the stability of the power grid frequency. Through the above method, the intelligent management and optimal configuration of frequency regulation resources can be realized, and the frequency regulation capability and operation economy of the power system can be improved. This resource allocation method based on fuzzy clustering can adapt to the complex and changeable power system operating environment and provide strong support for ensuring the safe and stable operation of the power grid.

[0055] Furthermore, the fuzzy clustering algorithm is used to classify the frequency modulation resources. According to the available output capacity, response speed and other parameters of the resources, the process of dividing the resources into different priority levels includes:

[0056] Obtain parameter index data such as available output capacity and response speed of frequency modulation resources as input of fuzzy clustering algorithm. Set the number of cluster categories and randomly initialize the cluster centers of each category. Calculate the Euclidean distance from each frequency modulation resource sample point to each cluster center, and update the membership matrix based on the distance. Recalculate the cluster centers of each category based on the membership matrix. Determine whether the cluster center meets the convergence conditions. If not, continue iterative optimization until the convergence conditions are met. Based on the final clustering results, divide the frequency modulation resources into different priority levels, and the priority level is determined by parameters such as the available output capacity and response speed of the cluster center. For newly added frequency modulation resources, determine their priority level based on the similarity between their parameter indicators and those of each cluster center to achieve incremental resource classification.

[0057] Specifically, the fuzzy clustering algorithm is a commonly used unsupervised learning method for classifying frequency regulation resources into different priority levels. First, obtain the key parameters of the frequency regulation resources, such as the available output capacity and response speed. For example, there are 10 frequency regulation resources in a power grid system, and the available output capacity of each resource is 50MW, 80MW, 30MW, etc., and the response speed is 10s, 15s, 8s, etc. These parameters are used as input data for the clustering algorithm. The number of cluster categories is usually determined according to system requirements, such as dividing resources into three priority levels of high, medium, and low. When initializing the cluster center, the parameters of three resources can be randomly selected as the initial center. Next, the Euclidean distance from each resource to each cluster center is calculated, and the membership matrix is ​​updated. For example, if the distances from a resource to the three cluster centers are 5, 8, and 12, respectively, then the corresponding memberships may be 0.6, 0.3, and 0.1. According to the updated membership matrix, the cluster centers of each category are recalculated. This step is performed by weighted average, and the contribution of each resource to the cluster center is proportional to its membership. Determine whether the change between the new and old cluster centers meets the convergence condition, such as whether the center change between two iterations is less than the preset threshold of 0.001. If not, return to the distance calculation step to continue the iteration. The final clustering result divides the frequency regulation resources into different priority levels. Cluster centers with high priority levels usually have larger available output capacity and faster response speed. For example, the category with cluster center parameters (90MW, 5s) may be defined as the highest priority. This classification method can effectively balance the frequency regulation capability and response speed of the system. For newly added frequency regulation resources, their similarity with each cluster center can be calculated and divided into the most similar category. For example, a new resource with a parameter of (70MW, 12s) may be classified into a medium priority level by calculating its distance from each cluster center. This incremental classification method can dynamically adapt to system changes and improve the flexibility of frequency regulation resource management.

[0058] The application of fuzzy clustering algorithm in frequency regulation resource classification has many advantages. First, it can handle the uncertainty and fuzziness of resource parameters, which is more in line with the complexity of the actual system. Second, through the iterative optimization process, the algorithm can find more reasonable clustering results and improve the accuracy of classification. Finally, the incremental classification method enables the system to quickly adapt to new resources, improving the real-time and effectiveness of frequency regulation control. This clustering-based priority division method provides a scientific decision-making basis for power grid frequency regulation control, which helps to improve the frequency stability and operation efficiency of the system.

[0059] Step S103, construct a multi-objective optimization model for the classified frequency modulation resources, take frequency modulation response speed, frequency modulation cost and output stability as optimization targets, take system frequency recovery time and upper and lower limits of frequency modulation resource output as constraints, and use particle swarm optimization algorithm to solve the model to obtain the optimal output adjustment of each frequency modulation resource.

[0060] Implementable, such as Figure 2 As shown in the figure, according to the classification results of frequency modulation resources, the dynamic parameters of various frequency modulation resources are obtained, including frequency modulation response speed, frequency modulation cost function and output stability index; a multi-objective optimization model is established with frequency modulation response speed, frequency modulation cost and output stability as objective functions, and with system frequency recovery time and upper and lower limits of frequency modulation resource output as constraints; the established multi-objective optimization model is solved by particle swarm optimization algorithm, and the output adjustment amount of frequency modulation resources is updated according to the position information of each particle in the algorithm iteration process; it is judged whether the convergence condition of particle swarm algorithm is met, and if so, the position information of the current optimal particle is output as the optimal output adjustment amount of each frequency modulation resource; if not, the particle speed and position update formula is used to update the frequency modulation resource output adjustment amount. The speed and position of the particles are updated, and the optimization is continued iteratively; according to the optimal output adjustment of each frequency modulation resource, combined with the dynamic parameters of the frequency modulation resources, the fuzzy control algorithm is used to adjust the output of each frequency modulation resource in real time to ensure that the system frequency runs smoothly within the target range; in the frequency modulation process, by monitoring the system frequency recovery and the actual output of the frequency modulation resources, the reinforcement learning algorithm is used to adaptively optimize the output adjustment strategy of the frequency modulation resources to improve the system frequency recovery speed and frequency modulation effect; the optimized frequency modulation resource output adjustment strategy is applied to the frequency modulation control of the actual power system, and according to the feedback of the system operation and the frequency modulation effect, the frequency modulation resources are regularly optimized again to form a closed-loop frequency modulation optimization control.

[0061] Specifically, the acquisition of dynamic parameters of frequency regulation resources is the basis of optimization control. The frequency regulation response speed reflects the ability of resources to respond to frequency changes. For example, the response speed of thermal power units is usually in seconds, while that of energy storage systems can reach milliseconds. The frequency regulation cost function describes the economic feasibility of resource participation in frequency regulation. For example, for thermal power units, the relationship between their frequency regulation cost and output adjustment can be expressed by a quadratic function. The output stability index measures the stability of resource output. For example, the output fluctuation of wind farms is large and the stability is poor. The establishment of the multi-objective optimization model takes into account the frequency regulation effect, economy and stability. The system frequency recovery time is used as a constraint to ensure that the frequency is restored to the normal range within the specified time. The upper and lower limits of the frequency regulation resource output guarantee the actual adjustable capacity of the resource. The particle swarm optimization algorithm solves the model by simulating the intelligent behavior of the group. Each particle represents the output adjustment plan of a group of frequency regulation resources, and the position of the particle is the output adjustment of each resource. During the algorithm iteration process, the particle continuously updates its position and speed, gradually approaching the optimal solution. The convergence condition can be set to the number of iterations reaching the upper limit or the optimal solution has no obvious improvement for multiple consecutive times. When the conditions are met, the current optimal particle position is output as the final solution. If it does not converge, iterative optimization continues. For example, a system has three frequency modulation resources. After 100 iterations, the optimal output adjustment plan is obtained: thermal power increases by 20MW, pumped storage decreases by 15MW, and energy storage discharges by 10MW. The fuzzy control algorithm adjusts the output in real time according to the optimal adjustment amount and resource dynamic parameters. For example, the output adjustment of thermal power units can adopt fuzzy rules: if the frequency deviation is large and the rate of change is large, the output is adjusted significantly; if the frequency deviation is small and the rate of change is small, the output is adjusted slightly. This method can effectively handle the nonlinear characteristics and uncertainties of the system. The reinforcement learning algorithm optimizes the frequency modulation strategy by continuously interacting with the environment. For example, the Q learning algorithm is used to take the system frequency deviation and the frequency modulation resource output as the state and the frequency modulation instruction as the action. By repeatedly trying different frequency modulation strategies, the optimal decision is gradually learned. This method can adapt to changes in system parameters and improve the frequency modulation effect. Closed-loop optimization control reflects the dynamic and adaptive nature of the frequency modulation process. Frequency regulation resources can be re-optimized by regularly evaluating the frequency regulation effect, such as frequency recovery speed, frequency regulation cost and other indicators. For example, if the actual response speed of a certain energy storage power station is found to be lower than expected, its weight can be adjusted in the next round of optimization to reduce its proportion of frequency regulation. This continuous optimization method can adapt to the dynamic changes of the power system and maintain the effectiveness of frequency regulation control.

[0062] Step S104, converting the calculated output adjustment of each frequency modulation resource into a specific frequency modulation control instruction, sending it to the corresponding control device through the power communication network, constructing a game matrix using the Nash equilibrium theory in game theory, calculating the optimal strategy of each frequency modulation resource, and coordinating and correcting the frequency modulation instruction through iterative optimization.

[0063] It is feasible to use an optimization algorithm to calculate the output adjustment of each frequency modulation resource according to the dynamic characteristics of the power grid frequency deviation and frequency modulation resources to obtain the initial frequency modulation instruction. The initial frequency modulation instruction is converted into a standardized control instruction format, and the instruction is distributed to the corresponding frequency modulation resource control device through the power-specific communication network. For each frequency modulation resource, its key attribute parameters, such as adjustment rate, adjustment accuracy, adjustment cost, etc., are extracted to construct a multi-party game model. With the goal of optimizing the power grid frequency recovery quality and frequency modulation cost, a game matrix is ​​constructed, the Nash equilibrium point is solved, and the optimal response strategy of each frequency modulation resource is obtained. The game equilibrium result is compared with the initial frequency modulation instruction, the frequency modulation instruction correction amount is calculated, and the optimized frequency modulation instruction is generated. The iterative optimization method is used to send the corrected frequency modulation instruction to the control device again for iterative optimization until the convergence condition of the frequency modulation instruction is reached. The final optimized frequency modulation instruction is continuously sent to each control device, and the output of each frequency modulation resource is dynamically optimized and adjusted according to the real-time frequency deviation of the power grid, so as to achieve rapid recovery of the power grid frequency and economic optimization regulation.

[0064] Specifically, according to the dynamic characteristics of the grid frequency deviation and frequency regulation resources, the optimization algorithm is used to calculate the output adjustment of each frequency regulation resource to obtain the initial frequency regulation instruction. For example, when the grid frequency deviates by 0.2Hz, the system first detects this deviation and calculates the output adjustment of each frequency regulation resource such as gas turbine, hydropower unit and energy storage system through the optimization algorithm. Assuming that the regulation rate of the gas turbine is 5MW / min, the hydropower unit is 10MW / min, and the energy storage system is 20MW / min, the algorithm determines the initial output adjustment instruction of each resource based on these parameters and the current grid status. The initial frequency regulation instruction is converted into a standardized control instruction format, and the instruction is distributed to the corresponding frequency regulation resource control device through the power-specific communication network. Specifically, the initial frequency regulation instruction such as "gas turbine increases output by 10MW" will be converted into a control instruction that complies with the IEC61850 standard and is quickly transmitted to the control system of the gas turbine through the optical fiber communication network to ensure the accurate execution of the instruction. For each frequency regulation resource, its key attribute parameters, such as regulation rate, regulation accuracy, regulation cost, etc., are extracted to construct a multi-party game model. Taking the gas turbine as an example, its regulation rate is fast but the cost is high, and the regulation accuracy is ±1MW; the hydropower unit has a medium regulation rate, low cost, and regulation accuracy of ±0.5MW; the energy storage system has the fastest regulation rate, moderate cost, and regulation accuracy of ±0.2MW. Based on these parameters, a game matrix is ​​constructed, and the regulation effect and cost of each resource are considered to solve the Nash equilibrium point and obtain the optimal response strategy of each frequency regulation resource. The game equilibrium result is compared with the initial frequency regulation instruction, the frequency regulation instruction correction amount is calculated, and the optimized frequency regulation instruction is generated. Assume that the initial instruction requires the gas turbine to increase its output by 10MW, but the game result shows that considering the cost and regulation effect, the optimal strategy is that the gas turbine increases its output by 8MW and the hydropower unit increases its output by 2MW. Therefore, the revised frequency regulation instruction is that the gas turbine increases its output by 8MW and the hydropower unit increases its output by 2MW. The iterative optimization method is adopted to send the revised frequency regulation instruction to the control device again, and the above steps are repeated until the convergence condition of the frequency regulation instruction is met. For example, after the first iteration, the system frequency deviation dropped from 0.2Hz to 0.1Hz. The optimization calculation was performed again to adjust the output of each resource. After several iterations, the frequency deviation was stabilized within the range of ±0.02Hz, reaching the convergence condition. The final optimized frequency regulation instructions are continuously sent to each control device. According to the real-time frequency deviation of the power grid, the output of each frequency regulation resource is dynamically optimized and adjusted to achieve rapid recovery of the power grid frequency and economic optimization and regulation. For example, when the power grid load suddenly increases and the frequency drops, the system responds immediately and adjusts the output of each resource according to the optimized instructions. The gas turbine quickly increases the output, the hydropower unit cooperates with the adjustment, and the energy storage system provides a fast response to ensure that the frequency is restored to the target range in a short time, while minimizing the frequency regulation cost.In this way, not only can the grid frequency be quickly restored, but the utilization of frequency regulation resources can also be optimized and the frequency regulation cost can be reduced while ensuring the stability of the grid. The introduction of the multi-party game model enables each frequency regulation resource to reach the optimal balance point in coordinated regulation, improving the overall frequency regulation effect. Dynamic optimization and adjustment ensures that the grid can maintain efficient and stable frequency control under different operating conditions, improving the operating efficiency and reliability of the grid.

[0065] Step S105, in the process of executing the frequency modulation control instruction, continuously monitor the changing trend of the system frequency, adopt fuzzy control rules, dynamically adjust the output allocation ratio of each frequency modulation resource according to the frequency deviation and change rate, and when the system frequency returns to a stable range, gradually exit the frequency modulation control mode and restore the output of the frequency modulation resources to a normal level.

[0066] Implementable, such as Figure 3 As shown, according to the preset frequency modulation control instruction, the time period and target frequency range for frequency modulation control are determined, and the frequency modulation control process is started. The system frequency value at the current moment is obtained, and its deviation value from the target frequency is calculated, and the rate of change of the frequency in unit time is calculated. The frequency deviation value and the rate of change are used as inputs, and fuzzy reasoning is performed through the pre-established fuzzy control rule base to obtain the output allocation ratio adjustment value of each frequency modulation resource. According to the output allocation ratio adjustment value, the actual output value of each frequency modulation resource is dynamically adjusted to make the system frequency gradually approach the target frequency range. The system frequency change is continuously monitored. If the frequency deviation value and the rate of change are both maintained within the preset stable range for a certain period of time, it is judged that the system frequency has returned to stability. When the system frequency returns to stability, the output adjustment amplitude of the frequency modulation resources is gradually reduced until the output of each frequency modulation resource returns to the normal level. The frequency modulation control process is stopped, the frequency modulation resources are released, and the normal operation mode is entered. At the same time, the relevant data of this frequency modulation control process is recorded for optimizing the fuzzy control rule base.

[0067] Specifically, according to the preset frequency control instruction, the time period and target frequency range for frequency control are first determined. For example, the frequency of the power grid fluctuates greatly during peak load periods (such as 18:00-20:00 every day), and the preset target frequency range is 49.95Hz to 50.05Hz. Start the frequency control process, first obtain the system frequency value at the current moment, assume that the current frequency is 49.8Hz, calculate its deviation value from the target frequency of 50Hz as -0.2Hz, and calculate the rate of change of frequency in unit time, such as a decrease of 0.1Hz per minute. Take the frequency deviation value and the rate of change as input, and perform fuzzy reasoning through the pre-established fuzzy control rule base. The fuzzy control rule base may contain the following rules: "If the frequency deviation is large and the rate of change is fast, then the frequency resource output is greatly increased", "If the frequency deviation is small and the rate of change is slow, then the frequency resource output is slightly adjusted". According to these rules, the output allocation ratio adjustment value of each frequency resource is inferred. Assume that the fuzzy reasoning result is to increase the frequency resource output by 30%. According to the output allocation ratio adjustment value, the actual output value of each frequency modulation resource is dynamically adjusted. Assume that the original output of a frequency modulation resource is 100MW, and the output after adjustment is 130MW. Through this dynamic adjustment, the system frequency gradually approaches the target frequency range. Continuously monitor the system frequency change. If the frequency deviation value and the change rate are both maintained within the preset stable range for a certain period of time, such as the frequency is stable between 49.95Hz and 50.05Hz and the change rate is less than 0.01Hz / min, it is judged that the system frequency has recovered stability. When the system frequency recovers stability, gradually reduce the output adjustment amplitude of the frequency modulation resource until the output of each frequency modulation resource returns to the normal level. This process is similar to the progressive braking of a car, ensuring that the system smoothly transitions to normal operation. Stop the frequency modulation control process, release the frequency modulation resources, and enter the normal operation mode. At the same time, record the relevant data of this frequency modulation control process, such as the execution time of the frequency modulation instruction, the frequency change curve, the frequency modulation resource output adjustment, etc., for optimizing the fuzzy control rule base. In this way, not only can the rapid recovery and economic optimization of the grid frequency be achieved, but also the accuracy and efficiency of frequency control can be continuously improved. For example, after multiple accumulation and analysis of frequency control data, the optimal output adjustment strategy of certain frequency resources under specific frequency deviations can be found, and then the fuzzy control rule base can be optimized to make it more in line with the actual operation situation. In addition, the application of fuzzy control enables the system to maintain good control effects in an environment with high uncertainty and complexity. For example, under sudden changes in grid load or extreme weather conditions, fuzzy control can dynamically adjust the output of frequency resources according to real-time data to ensure the stability of system frequency. In general, the application of this series of steps and technical means not only improves the intelligence level of grid frequency control, but also significantly improves the operating efficiency and reliability of the system, providing a strong guarantee for the safe and stable operation of the grid.

[0068] Furthermore, the frequency deviation value and the change rate are used as inputs, and fuzzy reasoning is performed through a pre-established fuzzy control rule base to obtain the output allocation ratio adjustment value of each frequency modulation resource, including:

[0069] Obtain the grid frequency deviation value and frequency deviation change rate, and use them as the input of the fuzzy controller. Perform fuzzy reasoning calculations on the frequency deviation value and change rate according to the pre-established fuzzy control rule base. Through fuzzy reasoning, obtain the fuzzy adjustment value of the output allocation ratio of each frequency regulation resource. Defuzzify the fuzzy adjustment value to obtain an accurate output allocation ratio adjustment value. Determine whether the current output of each frequency regulation resource reaches the adjusted target output allocation ratio. If not, continue to adjust. According to the dynamic characteristic parameters of the frequency regulation resources, use the model predictive control algorithm to optimize the output adjustment process of each resource. Send the optimized output adjustment instructions to each frequency regulation resource to achieve precise control of the grid frequency.

[0070] Specifically, precise control of grid frequency is the key to ensuring stable operation of the power system. First, the grid frequency deviation value and change rate are obtained as the input of the fuzzy controller. For example, if the current frequency is 49.95Hz and the target frequency is 50Hz, the frequency deviation is -0.05Hz; if the frequency was 49.93Hz 1 minute ago, the change rate is 0.02Hz / min. These two parameters reflect the current state and change trend of the system frequency. Next, fuzzy reasoning is performed according to the pre-established fuzzy control rule base. The rule base may contain rules such as "if the frequency deviation is negative and the change rate is positive, increase the output of frequency regulation resources." Through fuzzy reasoning, the fuzzy adjustment value of the output allocation ratio of each frequency regulation resource is obtained, such as "slightly increase the output of hydropower units" and "moderately reduce the output of gas units." The fuzzy adjustment value is defuzzified to obtain the precise output allocation ratio adjustment value. For example, "slightly increase" may correspond to a 5% increase, and "moderately reduce" may correspond to a 10% decrease. In this way, the qualitative fuzzy description is converted into a quantitative adjustment instruction. Determine whether the current output of each frequency regulation resource reaches the adjusted target output allocation ratio. Assuming that the current output of a hydropower unit is 80MW and the target output is 84MW, it is necessary to continue to increase the output by 4MW. This process ensures the consistency between the actual output of the frequency regulation resources and the control instructions. Considering the dynamic characteristics of different frequency regulation resources, the model predictive control algorithm is used to optimize the output adjustment process of each resource. For example, for thermal power units with slow response speed, it may be necessary to issue adjustment instructions in advance; while for energy storage systems with fast response, it can intervene quickly when the frequency deviation is large. This optimization takes into account the characteristics of various resources and improves the efficiency and accuracy of frequency regulation. Finally, the optimized output adjustment instructions are issued to each frequency regulation resource. For example, the hydropower unit is instructed to gradually increase the output from 80MW to 84MW within 5 minutes, and the energy storage system is required to immediately increase the discharge power by 2MW. These precise instructions ensure the precise control of the power grid frequency. Through this series of steps, power grid dispatchers can flexibly allocate various frequency regulation resources according to the real-time frequency conditions to achieve rapid and precise adjustment of the power grid frequency. This method not only improves the stability of the system, but also can effectively respond to the challenges of intensified frequency fluctuations when new energy sources are connected on a large scale, providing strong support for the safe and economical operation of the power grid.

[0071] Step S106, evaluate the completed frequency regulation control process, use data mining technology to analyze the dynamic response characteristics and control effects of frequency regulation resources, combine expert experience and knowledge to form an optimized frequency regulation control strategy knowledge base, conduct comparative analysis of different strategies through the simulation platform, verify the feasibility and effectiveness of the strategies, and use the updated knowledge base to guide the frequency regulation control decisions of the actual power system to achieve continuous optimization and improvement of frequency regulation control.

[0072] It is feasible to use data mining algorithms to analyze the dynamic response characteristics of frequency regulation resources, including response speed, regulation accuracy and other indicators, based on the power system operation data and frequency regulation control records, and obtain the performance evaluation results of frequency regulation resources. Evaluate the frequency regulation control effect, use clustering algorithms to analyze indicators such as frequency deviation and control time under different operating conditions, identify typical control effect patterns, and compare them with expected control targets to determine the pros and cons of the control effect. Collect and organize the experience and knowledge summarized by experts in frequency regulation control practice, including control strategy selection and control parameter setting under different operating modes, formalize them through knowledge representation methods, and build a frequency regulation control strategy knowledge base. Construct a simulation model of power system and frequency regulation control in the simulation platform, design simulation experiments for different control strategies in the knowledge base, simulate various typical operating conditions, and compare the control effects under different strategies. According to the simulation results, use a multi-objective optimization algorithm to comprehensively consider factors such as frequency regulation performance, control effect, and economy, select feasible and effective frequency regulation control strategies, and update and improve the knowledge base. The optimized frequency control strategy knowledge base is connected to the power dispatching control system. When the actual operating conditions match the typical conditions, the corresponding control strategy is automatically recommended to guide the dispatcher to make frequency control decisions. In actual applications, frequency control data is continuously collected, data mining analysis and simulation verification are regularly carried out, and the frequency control strategy knowledge base is continuously revised and improved to form a closed-loop optimization mechanism for frequency control and achieve continuous improvement of frequency control level.

[0073] Specifically, frequency regulation resource performance evaluation is the basis for optimizing frequency regulation control. Taking a hydropower station as an example, by analyzing its historical operation data, it can be concluded that the frequency regulation response speed of the power station is to reach 90% of the command value within 30 seconds, and the regulation accuracy is ±0.02Hz. These indicators reflect the good frequency regulation performance of the hydropower station, which is suitable as the main frequency regulation resource. Cluster analysis helps to identify typical frequency regulation control effect patterns. For example, K-means clustering of the frequency regulation control data of a power grid within one year can divide the control effect into three categories: "fast recovery", "slow recovery" and "oscillation". Among them, the "fast recovery" mode accounts for 60%, the average control time is 5 minutes, and the frequency deviation is controlled within ±0.1Hz, which meets the expected control target. The construction of the frequency regulation control strategy knowledge base incorporates expert experience. For example, in the case of large power grid decoupling to form an island operation, experts recommend the use of a two-stage control strategy of "fast and large adjustment + fine tuning". This strategy is formally represented and stored in the knowledge base to provide guidance for frequency regulation control in similar scenarios. Simulation verification is an important means to optimize frequency regulation strategies. Taking a regional power grid as an example, a simulation model including thermal power, hydropower, new energy and other power sources is constructed. For the scenario of a sudden increase of 10% in load, two strategies, "unified frequency regulation of the whole network" and "partition-coordinated frequency regulation", are simulated respectively. The results show that the latter can shorten the frequency recovery time by 20%, and the regulation amplitude of various power sources is more balanced. Multi-objective optimization algorithm is used to comprehensively evaluate the frequency regulation strategy. The particle swarm algorithm is used to score different strategies with frequency recovery time, root mean square of frequency deviation and frequency regulation cost as optimization targets. The optimization results of a certain time show that the strategy of "partition coordination + new energy participation" is better than the traditional strategy in all three targets and is recommended to be updated to the knowledge base. The combination of knowledge base and dispatch control system realizes intelligent frequency regulation decision support. When the system detects that the frequency drops rapidly due to the tripping of a large unit, it automatically matches the emergency frequency regulation strategy in the knowledge base and recommends the combined measures of "starting the standby unit + removing part of the load" to the dispatcher to assist in quick decision-making. The closed-loop optimization mechanism ensures the continuous improvement of the frequency regulation control level. A power grid company mines and analyzes frequency control data every quarter and finds that the frequency regulation effect is poor during the summer when air conditioning load is concentrated. Through simulation verification, the frequency regulation resource allocation strategy for this period was optimized, which increased the frequency qualification rate by 5 percentage points. This improvement was updated to the knowledge base in a timely manner to guide the frequency regulation control for the next year.

[0074] This embodiment also provides a primary frequency modulation control system based on a collaborative optimization strategy, which is used to implement a primary frequency modulation control method based on a collaborative optimization strategy, including: a frequency modulation control trigger module, a frequency modulation resource classification module, a resource output adjustment module and an equipment frequency modulation control module;

[0075] The frequency modulation control trigger module is used to obtain the system frequency deviation value of the power system, and when the system frequency deviation value exceeds the preset frequency deviation threshold, the frequency modulation control is triggered;

[0076] The frequency modulation resource classification module is used to obtain various types of frequency modulation resources, classify the frequency modulation resources, and determine the frequency modulation priority of various types of frequency modulation resources;

[0077] The resource output adjustment module is used to construct a multi-objective optimization model based on the classified frequency modulation resources, taking the frequency modulation response speed, frequency modulation cost and output stability as optimization targets, taking the system frequency recovery time and the upper and lower limits of the frequency modulation resource output as constraints, solving the multi-objective optimization model, and obtaining the optimal output adjustment amount of various frequency modulation resources;

[0078] The equipment frequency modulation control module is used to convert the optimal output adjustment of the various frequency modulation resources into corresponding frequency modulation control instructions and send them to the corresponding control equipment; and to monitor the system frequency in real time. When the real-time system frequency deviation value is within the preset frequency deviation threshold, the frequency control is completed.

[0079] Embodiment 2

[0080] This embodiment also provides an electronic device, including: a memory and a processor; the memory is used to store a program; the processor is used to execute the program to implement each step of the primary frequency modulation control method based on the collaborative optimization strategy.

[0081] Embodiment 3

[0082] This embodiment also provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, each step of the primary frequency modulation control method based on the collaborative optimization strategy is implemented.

[0083] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application 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.

Claims

1. A primary frequency modulation control method based on collaborative optimization strategy, characterized in that: The following steps are involved: Acquire a system frequency deviation value of the power system, and trigger frequency modulation control when the system frequency deviation value exceeds a preset frequency deviation threshold; Acquire various frequency modulation resources, classify the frequency modulation resources, and determine the frequency modulation priorities of various frequency modulation resources; Based on the classified frequency regulation resources, a multi-objective optimization model is constructed, with frequency regulation response speed, frequency regulation cost and output stability as optimization targets, and system frequency recovery time and upper and lower limits of frequency regulation resource output as constraints, to solve the multi-objective optimization model and obtain the optimal output adjustment of various frequency regulation resources; Converting the optimal output adjustment amounts of the various frequency modulation resources into corresponding frequency modulation control instructions and sending them to corresponding control devices; The system frequency is monitored in real time, and when the real-time system frequency deviation value is within the preset frequency deviation threshold, the frequency modulation control is completed.

2. The method according to claim 1, characterized in that The system frequency of the power system is obtained. When the system frequency exceeds a preset frequency threshold, the process of triggering frequency regulation control includes: The real-time operating status data of the power system is obtained through sensors, and based on the obtained real-time operating status data, the real-time system frequency value is calculated using a Fourier transform algorithm, and the system frequency value is compared with the rated frequency value of the power system to obtain the system frequency deviation value at the current moment; a preset frequency deviation threshold is obtained from a pre-established frequency deviation threshold database, and the system frequency deviation value at the current moment is compared with the preset frequency deviation threshold. If the system frequency deviation value exceeds the frequency deviation threshold, the frequency modulation control process is triggered.

3. The method according to claim 1, characterized in that The process of acquiring various frequency modulation resources, classifying the frequency modulation resources, and determining the frequency modulation priorities of the various frequency modulation resources includes: Obtain frequency modulation resource parameters of various frequency modulation resources; construct a load forecasting model, predict the load demand of the system based on the load forecasting model, and obtain load forecasting data; obtain the power structure information of the system, integrate the frequency modulation resource parameters, load forecasting data and power structure information, input them into the fuzzy clustering algorithm, use the fuzzy clustering algorithm to reclassify various frequency modulation resources, and determine the frequency modulation priority of various frequency modulation resources; Among them, the frequency modulation resource parameters include available output capacity and frequency modulation response speed; the power supply structure information includes the capacity and response speed of various power supplies.

4. The method according to claim 3, characterized in that The process of reclassifying various frequency modulation resources by using fuzzy clustering algorithm and determining the frequency modulation priority of various frequency modulation resources includes: The number of clustering categories is set, and the cluster centers of each category are randomly initialized. The Euclidean distance from each frequency modulation resource sample point to each cluster center is calculated, and the membership matrix is ​​updated based on the Euclidean distance. The cluster centers of each category are recalculated based on the updated membership matrix. When the cluster centers of each category meet the convergence conditions, the clustering is terminated to obtain the classification results of the frequency modulation resources, and the frequency modulation priority of each type of frequency modulation resource is determined based on the available output capacity and response speed of the cluster centers of each category.

5. The method according to claim 1, characterized in that The process of solving the multi-objective optimization model and obtaining the optimal output adjustment amount of various frequency regulation resources includes: A particle swarm optimization algorithm is used to solve the multi-objective optimization model. During the algorithm iteration process, the output adjustment amount of the frequency modulation resources is updated based on the position information of each particle. When the particle swarm algorithm meets the convergence condition, the position information of the current optimal particle is output as the optimal output adjustment amount of each frequency modulation resource.

6. The method according to claim 1, characterized in that The process of converting the optimal output adjustment amount of each type of frequency modulation resource into a corresponding frequency modulation control instruction and sending the instruction to the corresponding control device includes: With the goal of optimizing the grid frequency recovery quality and frequency regulation cost, a game matrix is ​​constructed, and the Nash equilibrium point is solved based on the game matrix to obtain a game equilibrium result; the game equilibrium result is compared with the initial frequency regulation control instruction, the frequency regulation instruction correction amount is calculated, and the optimized frequency regulation control instruction is generated; the optimized frequency regulation control instruction is continuously iteratively optimized until the convergence condition of the frequency regulation control instruction is reached, and the finally obtained optimized frequency regulation control instruction is issued to the corresponding control device.

7. The method according to claim 1, characterized in that After completing the frequency modulation control, it also includes: According to the power system operation data and frequency regulation control records, the dynamic response characteristics of the frequency regulation resources are analyzed by using data mining algorithms to obtain the frequency regulation resource performance evaluation results; the frequency deviation and control time under different operating conditions are analyzed by using clustering algorithms, typical control effect modes are identified, and compared with the expected control targets to obtain the frequency regulation control effect evaluation results; based on the frequency regulation resource performance evaluation results and the frequency regulation control effect evaluation results, combined with expert experience and knowledge, a frequency regulation control strategy knowledge base is constructed; Construct a simulation model of the power system and frequency control, design simulation experiments for different frequency control strategies in the frequency control strategy knowledge base, simulate various typical operating conditions, and obtain simulation results; based on the simulation results, adopt a multi-objective optimization algorithm, comprehensively consider the frequency regulation performance and frequency regulation control effect, optimize the frequency regulation control strategy, update and improve the frequency regulation control strategy knowledge base, and use the updated frequency regulation control strategy knowledge base to guide the frequency regulation control decision of the actual power system, so as to achieve continuous optimization of frequency regulation control.

8. A primary frequency modulation control system based on collaborative optimization strategy, characterized in that: A primary frequency modulation control method based on a collaborative optimization strategy for implementing any one of claims 1 to 7, comprising: a frequency modulation control trigger module, a frequency modulation resource classification module, a resource output adjustment module, and an equipment frequency modulation control module; The frequency modulation control trigger module is used to obtain a system frequency deviation value of the power system, and when the system frequency deviation value exceeds a preset frequency deviation threshold, the frequency modulation control is triggered; The frequency modulation resource classification module is used to obtain various types of frequency modulation resources, classify the frequency modulation resources, and determine the frequency modulation priority of various types of frequency modulation resources; The resource output adjustment module is used to construct a multi-objective optimization model based on the classified frequency modulation resources, taking the frequency modulation response speed, frequency modulation cost and output stability as optimization targets, taking the system frequency recovery time and the upper and lower limits of the frequency modulation resource output as constraints, solving the multi-objective optimization model, and obtaining the optimal output adjustment amount of each type of frequency modulation resource; The device frequency modulation control module is used to convert the optimal output adjustment amount of the various frequency modulation resources into corresponding frequency modulation control instructions and send them to the corresponding control device; and to monitor the system frequency in real time. When the real-time acquired system frequency deviation value is within the preset frequency deviation threshold, the frequency control is completed.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store programs; the processor is used to execute the programs to implement the various steps of the primary frequency modulation control method based on the collaborative optimization strategy as described in any one of claims 1 to 7.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the primary frequency modulation control method based on a collaborative optimization strategy as described in any one of claims 1 to 7 is implemented.

Images