102 results about "Artificial bee colony algorithm" patented technology

The invention provides an unmanned aerial vehicle path planning method and system and a storage medium, and the method comprises the steps: constructing a three-dimensional path planning model of an unmanned aerial vehicle in a target flight region; solving the three-dimensional path planning model by using an improved artificial bee colony algorithm, obtaining the optimal flight path of the unmanned aerial vehicle under each target function, and summarizing the optimal flight path into a Pareto optimal solution set; constructing a plurality of decision intelligent agents in one-to-one correspondence with the plurality of objective functions, performing multi-dimensional scoring under different objective functions on each flight path in the Pareto optimal solution set by using the plurality of decision intelligent agents, obtaining a comprehensive score of each flight path, and determining a global optimal unmanned aerial vehicle flight path based on the comprehensive score; according to the method, through refined multi-dimensional constraint modeling, an improved multi-target artificial bee colony algorithm and multi-agent collaborative decision based on a near-end strategy optimization algorithm, full-process optimization from path generation to intelligent decision is realized.

The invention discloses a high-pressure jet grouting pile construction optimization control method and system, and belongs to the technical field of pile foundation construction. According to the method, the pile body strength is predicted based on the XGBoost regression model, the HSIO hybrid swarm intelligent optimization algorithm is combined with the artificial bee colony algorithm and the seathead optimization algorithm, the construction parameter combination is dynamically adjusted, and optimal matching of the target strength and the real-time construction state is achieved. The method solves the problems of parameter setting deviation and non-uniform pile forming quality caused by the fact that traditional construction depends on artificial experience, is particularly suitable for high-precision construction requirements under complex geological conditions (such as karst areas and deep and thick sludge layers), remarkably improves the pile body strength stability, construction efficiency and quality controllability, and reduces the construction cost. And meanwhile, material waste and personal error risks are reduced.

The invention discloses a multi-rotating-speed flexible job shop scheduling method applied to double-resource constraint, and the method comprises the steps: taking the maximum completion time and minimum total energy consumption of a minimum machine as target functions, and constructing a flexible job shop scheduling model considering the rotating speed energy consumption of the machine and the production demands of a fine process; a machine speed gear constraint, a fine process constraint, a process sequence constraint, a completion time constraint, a machine processing constraint and a worker operation constraint are established as constraint conditions of the model; the flexible job shop scheduling problem is solved by adopting an improved artificial bee colony algorithm, bee colony search guided by excellent genes is adopted in bee learning operation in the improved artificial bee colony algorithm, and nectar source optimization is carried out based on the searched excellent genes; the following bee operation adopts a neighborhood structure which considers machine speed change and balances the working time of workers to carry out dynamic neighborhood search so as to optimize a nectar source. The effectiveness of the improved strategy is verified through experiments, and the superiority is verified through comparison of different algorithms on expansion standard examples.

The invention discloses a cascade reservoir scheduling method and system based on an artificial bee colony algorithm, and the method comprises the following steps: the system collects multi-source hydrological data in real time through a hydrological perception and preprocessing module, and introduces a large language model to carry out the semantic judgment and anomaly labeling of an abnormal hydrological time sequence; inputting the processed high-quality data into a reservoir model construction module, establishing a cascade reservoir optimal scheduling system, and setting corresponding boundary conditions and operation constraints in combination with reservoir scheduling regulations; the scheduling optimization module receives model input, adopts a variable structure taking a water level as a core to construct an optimization individual, and completes population initialization, disturbance generation and fitness evaluation based on a potential solution guide mechanism in an improved artificial bee colony algorithm; the system transmits the scheduling sequence optimized and output by the scheduling optimization module into an LLM intelligent auxiliary module; and the intelligent text interpretation generated by the LLM and the scheduling optimization solution enter a result evaluation and visualization module together. According to the scheduling method and system, a high-quality and physically feasible scheduling scheme can be output within reasonable calculation time.

The invention relates to the field of parameter identification, and provides a photovoltaic model parameter identification method and system based on an improved artificial bee colony algorithm, and the method specifically comprises the steps: obtaining a to-be-identified photovoltaic model, and constructing an optimization objective function which is based on a root-mean-square error; initializing an artificial bee colony algorithm and performing terrain complexity evaluation to obtain a terrain type, the terrain complexity evaluation being based on the population fitness difference matrix; and according to the terrain type, adaptive search is carried out to obtain a final identification result, and the adaptive search is based on a smooth terrain processing mechanism and a rugged terrain processing mechanism. According to the method, the defect of a traditional artificial bee colony algorithm facing a complex terrain area is avoided, and the accuracy and efficiency of photovoltaic model parameter identification are improved.

The invention provides a robot path planning method and system based on an artificial bee colony and reinforcement learning, and relates to the field of path planning, and the method specifically comprises the following steps: initializing an artificial bee colony algorithm population according to known environment information, a starting point and a target point; and performing iterative optimization on the population based on a search mechanism of an artificial bee colony algorithm. And when a preset optimization process termination condition is met, selecting a batch of high-quality solutions from the final population to form an elite solution set. And encoding the elite solution set into priori knowledge of a reinforcement learning agent to initialize a value function of reinforcement learning, and setting a reward function fusing a path length and a smoothness target. And based on the initialized value function, running a reinforcement learning algorithm to carry out strategy search and optimization so as to carry out fine adjustment on a path strategy. And after the reinforcement learning algorithm converges, outputting an optimal path. According to the technical scheme, the problem that a short and smooth path cannot be quickly generated in a complex environment in the prior art is solved.

The invention discloses a green robust independent parallel locomotive scheduling method with uncertain processing time. The method comprises the following steps: acquiring a to-be-scheduled parameter set; constructing an irrelevant parallel machine scheduling model taking worst scene completion time WC and scene average energy consumption MTEC as double targets based on the parameters; a scene-driven double-population discrete artificial bee colony algorithm is adopted for solving, and the method comprises the steps of population initialization, employed bee global search based on ternary championics and two-point crossing, division into two sub-populations according to MTEC, MN local search based on a mean value scene, WN local search based on a worst scene, LN observation bee self-adaptive neighborhood search based on Q-learning and scout bee disturbance. And finally, outputting a robust scheduling solution set with both robustness and low-carbon property according to a Pareto criterion. And a plurality of scheduling schemes considering robustness and energy consumption optimization are provided for decision makers.

The invention provides an underwater vehicle measuring point layout optimization method and system, and belongs to the technical field of experimental measuring point layout optimization. Comprising the following steps: obtaining simulation data of an underwater vehicle structure through finite element simulation, and extracting strain data of a unit capable of arranging a strain gauge area according to actual engineering constraints; centroid coordinates of the units, sensitivity values of strain response values relative to working conditions and strain gradients are calculated by taking centroids of the units in the areas where the strain gauges can be arranged as nodes; mapping the strain response values of all the nodes to a unified interval to obtain a normalized strain response value; obtaining sensitivity coverage and strain gradient coverage, and calculating space coverage of all nodes; integrating the normalized strain response value, sensitivity coverage, strain gradient coverage and space coverage as optimization indexes, and performing optimization by adopting an improved discrete artificial bee colony algorithm; and a final layout scheme is decided by outputting the relationship between the number of the measuring points and the information amount.

The invention discloses a resource scheduling method and device, equipment, a storage medium and a program product, and the method comprises the steps: intercepting Pods created in a Kubernetes cluster, and obtaining a plurality of to-be-scheduled Pods; screening out a plurality of nodes to be deployed from the Kubernetes cluster according to the resource types and the resource requirements of the plurality of Pods to be scheduled and the resource information of the nodes in the Kubernetes cluster; according to a resource scheduling strategy based on an improved artificial bee colony algorithm, determining a target node of each to-be-scheduled Pod from the plurality of to-be-deployed nodes, and deploying corresponding target nodes for the plurality of to-be-scheduled Pods; balanced deployment of resources can be realized, and the resource utilization rate and the load balancing degree of the whole cluster are improved.

The present invention relates to an optimization method for integrated joint scheduling of production and logistics in a flexible job shop (FJSP) having a plurality of different types of automated guided vehicles (AGVs). The invention belongs to the field of assembly workshop production scheduling. Comprising the following steps: 1) according to a special assembly workshop machine and AGV combined scheduling process, Tent chaotic mapping is adopted to initialize a scheduling scheme and encode the scheduling scheme; 2) performing iterative optimization adjustment on the scheduling scheme through an improved multi-target artificial bee colony algorithm; and 3) carrying out production scheduling by using the optimized scheduling scheme. According to the method, the maximum completion time and the total energy consumption are optimized at the same time, the production efficiency is concerned, the requirements of green manufacturing and sustainable development are also considered, and enterprises are helped to achieve cost reduction and efficiency improvement, especially in the production process sensitive to energy consumption.

The invention relates to the technical field of production scheduling, in particular to a hybrid flow shop self-learning scheduling method considering machine deterioration, and aims to solve the problem of hybrid flow shop scheduling considering machine deterioration, a self-learning artificial bee colony algorithm is adopted to optimize the maximum completion time. In the hired bee stage, populations are divided into P1, P2 and P3 groups according to fitness to realize diversified learning, in the bee observation stage, the food source selection probability is improved based on knowledge, multi-neighborhood search is combined, the investigation bee stage is improved, and a nectar source diversity enhancement strategy is added. The SLABC is superior to PPSOGA, DWSA and other algorithms under different workpiece numbers, stage numbers and deterioration rates, the influence of machine deterioration on the construction period can be effectively handled, the scheduling performance is remarkably improved, and the actual production requirements are met.

The invention provides an inspection robot trajectory planning method suitable for a compact obstacle environment of a power distribution room, and solves the problems that keys are compact and an obstacle is close to a target key in the power distribution room environment. According to the method, a bee colony and a Q table are initialized, a fitness function is constructed, robot motion and environment constraints are introduced, and an exploration or development strategy is dynamically selected according to a population state. A particle swarm algorithm with control parameters is introduced in a scout bee stage, and the global search capability and convergence efficiency are enhanced. The optimal solution is updated through an elitist strategy and neighborhood search, and the Q value is adaptively adjusted by means of feedback, so that the strategy flexibility is improved. According to the method, the global search advantage of the particle swarm and the local development advantage of the artificial bee colony are integrated, the trajectory precision and the operation safety are guaranteed, the efficiency and the real-time performance are also improved, the success rate reaches 98.16%, the time is reduced by 10.11% compared with that of an artificial bee colony algorithm, and the method is suitable for popularization and application. The method is suitable for a patrol robot track planning task with a complex distribution room structure, a compact control panel arrangement and a closer obstacle to a target.

The invention relates to the technical field of intelligent inspection and cooperative control, and particularly discloses a robot dog and unmanned aerial vehicle cooperative inspection system and method based on an artificial bee colony algorithm, and the system comprises an unmanned aerial vehicle, a robot dog and a cooperative decision system. The robot dog comprises a navigation module, an environment sensing module, a first communication module and a charging platform; the unmanned aerial vehicle comprises a flight control module, a visual identification module, an image acquisition module, an energy management module and a second communication module; the collaborative decision module is used for operating an artificial bee colony algorithm to realize intelligent collaboration of the robot dog and the unmanned aerial vehicle; according to the invention, through dynamic conversion of roles, active exploration of an unknown area and autonomous precise charging on the mobile platform, all-weather, self-organization and full coverage of the inspection process are realized, and the inspection efficiency and the intelligent level in a complex environment are significantly improved.

The invention provides a micro-seismic positioning method and device based on an artificial bee colony algorithm, equipment and a storage medium. Relates to the technical field of dry micro-seismic data processing. The method comprises the following steps: acquiring micro-seismic data, identifying the micro-seismic data by adopting a long-short time window ratio method to obtain micro-seismic data of a micro-seismic event, and performing noise reduction processing and first arrival pickup to obtain an arrival time sequence of the micro-seismic event; taking the cosine similarity as a target function, taking the coordinates of each sensor as initial parameters, randomly selecting a plurality of initial points, starting inversion based on a set iteration number, calculating a time difference sequence from each initial point to each sensor as an arrival time sequence of an inversion waveform, and calculating the arrival time sequence of the inversion waveform; and if the cosine similarity of the arrival time sequence of the inversion waveform is higher than the value before iteration, replacing the arrival time sequence of the previous inversion waveform with the arrival time sequence of the current inversion waveform, otherwise, continuing to search until the set number of iterations is reached, and outputting the optimal seismic source point. According to the invention, the accuracy of micro-seismic positioning can be obviously improved.

The application provides a robot path planning method and system based on artificial bee colony and reinforcement learning, and relates to the field of path planning, and the method specifically comprises the following steps: initializing an artificial bee colony algorithm population according to known environment information, a starting point and a target point; iteratively optimizing the population based on a search mechanism of the artificial bee colony algorithm; when a preset optimization process termination condition is met, selecting a batch of high-quality solutions from the final population to form an elite solution set; encoding the elite solution set as prior knowledge of a reinforcement learning agent, initializing a value function of the reinforcement learning, and setting a reward function that combines path length and smoothness objectives; based on the initialized value function, running a reinforcement learning algorithm for policy search and optimization to fine-tune the path strategy; and outputting an optimal path after the reinforcement learning algorithm converges. The technical scheme of the application overcomes the problem in the prior art that a short and smooth path cannot be quickly generated in a complex environment.

This invention discloses an edge computing federated aggregation optimization method for multi-regional distribution networks, belonging to the field of distribution network collaborative optimization and federated learning technology. This method is based on a three-tiered "cloud-edge-device" architecture, integrating an improved artificial bee colony (ABC) algorithm and an optimal transport (OT) mechanism to construct an AOT aggregation optimization scheme: the edge side collects and preprocesses runtime sequence data, and edge nodes complete local model training and upload parameters; the cloud searches for optimal aggregation weights using the improved ABC algorithm, adapting to differences between node data and models, and uses the OT mechanism to achieve parameter distribution alignment and weighted fusion; efficient aggregation is ensured through dual-trigger collaborative communication scheduling. This invention solves the problems of fixed weights and insufficient parameter alignment in traditional federated aggregation, significantly improving the global model accuracy, stability, and convergence speed in scenarios where data is not independent and identically distributed, reducing communication overhead, and providing reliable support for intelligent collaborative scheduling of multi-regional distribution networks.

The invention discloses a composite material pultrusion frame data acquisition and management system based on reinforcement learning. The system comprises a real-time process parameter acquisition and filtering unit, a digital twinborn simulation unit, a reinforcement learning control unit, a hyper-parameter optimization unit, an abnormity early warning unit and a data management unit. The system obtains a process parameter data stream through a sensor array and nonlinear Kalman filtering, generates a control strategy through a digital twin model and a SoftActor-Critic algorithm, optimizes hyper-parameters through an artificial bee colony algorithm, identifies and early warns process abnormal parameters in real time, and constructs a data feature library and auxiliary decision data. According to the invention, the data acquisition precision of the pultrusion process, the intelligent level of the production process and the product quality stability are improved.

The present application relates to the technical field of astronomical satellite mission planning, in particular to a method and system for solving short-term mission planning of astronomical satellite. The method comprises: step 1) constructing a short-term mission planning mathematical model of the astronomical satellite to be planned, and abstracting the mission planning problem as a maximized optimization problem; step 2) solving by using a hybrid search strategy artificial bee colony algorithm, searching for excellent solutions by the employed bees based on an "elite solution guided search" strategy, and searching by the follower bees based on a "neighborhood optimal solution update" strategy, so as to accelerate the solution and improve the solution accuracy. Compared with the basic artificial algorithm, the method has the advantages of fast convergence speed, high solution accuracy and strong optimization ability, and has fewer control parameters than other swarm intelligence algorithms; in the short-term mission planning problem of the astronomical satellite, the method can obtain higher task completion degree and greater observation benefit.

The invention discloses a wind power-containing electric power system source load cooperative scheduling method considering a high energy load, and the method comprises the steps: analyzing the scheduling demands of a wind power-containing electric power system, building an electric arc furnace load and wind power output uncertainty model based on the schedulable characteristics of the electric arc furnace load, generating a scene through employing a data driving method, and carrying out the calculation of the scene. The method comprises the following steps: analyzing uncertainty superposition influence of electric arc furnace load and wind power output, determining reserve capacity required by an electric power system to cope with uncertainty fluctuation, proposing a smelting type high-energy load adaptive electricity price mechanism considering wind power output change and source load fluctuation, establishing an optimal scheduling model, and solving the optimal scheduling model by adopting an artificial bee colony algorithm; according to the method, powerful theoretical support is provided for the smelting type high-energy load to participate in power grid dispatching and wind power absorption, the accuracy of power grid dispatching is guaranteed, and absorption of new energy and safe operation of a power system are promoted.

The invention relates to the technical field of artificial intelligence, in particular to an extensible deep neural network channel pruning method and system. According to the method, a channel pruning problem is converted into a structure search problem, and an artificial bee colony algorithm is provided to find an optimal pruning structure. A coupled model structure is mapped into a non-coupled search structure through structure mapping, and an optimal pruning structure is searched and fine-tuned based on an artificial bee colony algorithm, so that the search space of the deep neural network is reduced, and the pruning problem of the coupled structure is solved.

The application discloses a kind of underwater defect detection robot path planning and tracking control method based on multi-data fusion, to solve the problem of poor positioning accuracy, unstable path tracking and insufficient defect approaching ability in GPS denial environment.The method uses improved artificial bee colony algorithm to optimize BP neural network assisted Kalman filter, fuses INS, DVL, electronic compass and depth gauge data, realizes high-precision integrated navigation;Secondly, combined with global "bow" shape coverage and local pure tracking approaching strategy, generate full coverage scanning path and fine approach to high priority defect points;Finally, based on pre-defined time sliding mode controller, cooperate disturbance observer and adaptive mechanism, realize the fast robust tracking of path.Experiments show that the positioning error is only 0.15m after DVL failure, the local approaching accuracy is 0.05m, the path tracking convergence time is 5s, the steady-state deviation is 0.03m, which is significantly better than the traditional method.The application is suitable for automatic defect detection of underwater structures such as bridge pier, dam, submarine pipeline, etc.

This application relates to the field of artificial bee colony algorithm technology, providing a method and system for optimizing energy storage frequency regulation strategies based on the artificial bee colony algorithm, addressing the problems of slow suppression speed and low stability of power grid frequency fluctuations. The method includes: acquiring power grid frequency deviation data under load abrupt changes, extracting frequency change characteristics through sliding time window processing, and performing extreme value normalization to form a standard fluctuation sequence characterizing the dynamic characteristics of the frequency. After converting this sequence into an initial power compensation demand signal, phase compensation preprocessing is performed to obtain an optimized power compensation signal. A parameter optimization model is established with virtual inertial control parameters as optimization variables, and the optimal parameter combination is obtained by iteratively solving using the artificial bee colony algorithm, thereby constructing a frequency-power mapping function. Based on the mapping function, an adaptive virtual inertial control strategy is formed, and after processing the power compensation signal, an energy storage output command is generated. This application improves the suppression speed and stability of power grid frequency fluctuations.

The invention provides a non-replacement flow shop scheduling and opportunity maintenance optimization method based on quality control, and the method comprises the steps: setting basic assumptions, parameter definitions and constraint conditions according to the characteristics of a non-replacement flow shop; constructing a non-replacement flow shop scheduling optimization model with the maximum completion time, the weighted advance / tardiness cost, the quality loss cost and the preventive maintenance cost being minimized as optimization objectives; an original optimization problem is decomposed into two sub-problems: a workpiece processing sequence arrangement sub-problem, and a maintenance opportunity prevention and opportunity maintenance grouping arrangement sub-problem, and an improved multi-target co-evolution artificial bee colony algorithm is used to solve a constructed non-permutation flow shop scheduling optimization model. And outputting an optimal workpiece scheduling and opportunity prevention maintenance scheme. According to the method, collaborative optimization of production efficiency, maintenance economy and product stability is realized, and a feasible solution is provided for a complex manufacturing system.

The invention discloses a training flight path optimization method and system in combination with low-altitude training and transportation in an airport and a medium. Relates to the technical field of air transportation and flight training. Improvement is carried out on the basis of the prior art, a built training flight airspace grid model finely expresses training flight airspace information, the fine degree of airspace resource management is improved, a multi-target optimization model comprehensively considering safety, economy and green targets is built, and through reasonable setting of constraint conditions, the optimization efficiency of airspace resource management is improved. The optimization result better meets the actual flight training requirement, and the risk, oil consumption and pollutant emission of training flight are effectively reduced; based on the improved artificial bee colony algorithm, the solving efficiency and quality are improved, a better solution can be quickly found in a complex multi-objective optimization problem, and reliable algorithm support is provided for training flight trajectory optimization.

The application discloses a pre-hospital care helicopter optimization deployment method fusing weighted K-means and IABC, first, patient data is acquired and the geographical position coordinates thereof are standardized pretreated, the patients are classified according to the injury degree and corresponding weight values are given, and a weighted data set is formed; subsequently, a weighted clustering square error and a silhouette coefficient are calculated, the elbow rule and the silhouette coefficient method are cooperatively decided, and the optimal helicopter deployment number k is determined; then, the weighted data set is taken as input, and the initial deployment position coordinates of the k helicopters are solved through a weighted K-means algorithm; then, a single-target fitness function integrated by three sub-functions including service coverage, rescue response time and economic cost is constructed, and a multi-target optimization problem is integrated into a single-target optimization problem; finally, an improved artificial bee colony (IABC) algorithm with a directional learning mechanism is adopted, the initial deployment position of the helicopter is taken as an initialization population for iterative updating, and the final pre-hospital care helicopter deployment coordinates are output.

The application discloses a node selection method and device, electronic equipment and a storage medium. The method comprises the following steps: determining the position of at least one candidate network node in a first space; determining at least one network node for measuring network speed from the at least one candidate network node based on the position of the at least one candidate network node based on an artificial bee colony algorithm (ABC); wherein, in the process of selecting the network node for measuring the network speed, in the observation bee stage, the position information of the first candidate network node and the first parameter are used to select at least one second candidate network node. The technical scheme provided by the application further selects the node in the observation bee stage in the process of selecting the node based on the ABC, in combination with the node information shared by the leading bee and the global parameter, so as to guarantee the diversity of the node in the iteration process, thereby improving the probability of obtaining the optimal node, that is, improving the search accuracy of the node.

The application provides a kind of TP321H steel material welding parameter determination method, medium and system, belong to steel material welding technical field, including: first, collect the material characteristic parameters of TP321H steel material, such as corrosion resistance equivalent, tensile strength and the like, and obtain small-scale orthogonal welding test data, including welding parameter and welding quality data.Based on this, the welding quality equation set considering material characteristics and welding parameters is established.After obtaining the minimum welding quality requirement of construction process, the reasonable range and step length of each parameter are determined according to the parameter meaning and experience, and the welding parameter optimization path problem is constructed.After obtaining the optimal welding parameters by using artificial bee colony algorithm optimization solution, actual welding verification is carried out.If the result does not meet the requirements, adjust the fitness function weight of the algorithm and repeat the optimization until the verification result meets the requirements, output the optimal parameters, solve the problem that the existing technology usually depends on engineering experience and is difficult to determine the welding parameters.

The application discloses a kind of for ship section painting plan scheduling method, consider VOCs exhaust treatment equipment consumption electric energy and the carbon emission caused by LNG gas, consider the constraint of human resources and establish the multi-schedule plan scheduling model of two kinds of tasks including sand washing and spraying.A kind of improved artificial bee colony algorithm is proposed to effectively obtain approximate optimal solution within reasonable time, and a three-dimensional coding solution mechanism is designed based on the model.In the algorithm, the search efficiency of the algorithm is increased by mixing greedy random adaptive search algorithm and variable neighborhood search algorithm, which can be well used for ship section painting plan scheduling problem.

The application discloses a method for cooperative jamming based on intelligent optimization algorithm, comprising: constructing a jamming decision model, the jamming decision model comprising a cooperative jamming decision matrix, a gain matrix, a jamming matrix, a jamming gain matrix, a jamming bandwidth ratio factor, a jam-to-signal ratio, and a jamming benefit; establishing an objective function and a constraint condition of the jamming decision model according to the jamming benefit; and using an artificial bee colony algorithm to take the jamming benefit as a fitness function and optimize the cooperative jamming decision matrix A. In different complex electromagnetic spectrum environments such as limited spectrum resources and the same frequency band shared by jamming devices and illegal users, the limited jamming resources are reasonably distributed under the condition that the jamming device of the own side can normally communicate, so that greater jamming benefit is achieved; the algorithm convergence speed and search ability are improved, and the method is helpful for making a decision with higher jamming benefit in a shorter time.

The invention relates to the technical field of artificial bee colony algorithms, provides an energy storage frequency modulation strategy optimization method and system based on an artificial bee colony algorithm, and solves the problems of low suppression speed and low stability of power grid frequency fluctuation. The method comprises the following steps: acquiring power grid frequency deviation data under a load sudden change working condition, extracting frequency change characteristics through sliding time window processing, and performing extreme value normalization to form a standard fluctuation sequence representing frequency dynamic characteristics; and after the sequence is converted into an initial power compensation demand signal, phase compensation preprocessing is carried out to obtain an optimized power compensation signal. Establishing a parameter optimization model taking virtual inertia control parameters as optimization variables, performing iterative solution by adopting an artificial bee colony algorithm to obtain an optimal parameter combination, and constructing a mapping function of frequency and power according to the optimal parameter combination. And forming a self-adaptive virtual inertia control strategy based on the mapping function, and generating an energy storage output instruction after processing the power compensation signal. According to the invention, the suppression speed and stability of power grid frequency fluctuation are improved.