100 results about "Local convergence" patented technology

The invention provides a charging station dynamic pricing method based on a game grey wolf algorithm, and aims to solve the problem of dynamic pricing of a single charging station. According to the method, modeling is carried out on a charging station dynamic pricing problem, a Markov decision process is combined, a double-layer gradual-reduction population reduction strategy is introduced, the search range is gradually reduced, and the global search efficiency of an algorithm is improved; meanwhile, in combination with a self-adaptive local search strategy, the local search strength is dynamically adjusted to solve the optimal solution as much as possible. In order to further improve the performance of the algorithm, the neural plasticity variation strategy is fused, the capability of the algorithm for jumping out of local optimum is enhanced, a competitive cooperation mechanism based on hierarchical game is designed, and the global search and local convergence capability among populations is balanced. The improvements effectively improve the convergence, stability and robustness of the algorithm. Experimental results show that the method has good solution precision and convergence efficiency under dynamic requirements, the income optimization capability of the charging station and the accuracy of dynamic pricing are remarkably improved, and an efficient and practical solution is provided for the dynamic pricing problem of the charging station.

A train-track-bridge coupling response prediction method based on a sparrow optimization algorithm and a long short-term memory network comprises the steps that data such as train speed, axle load, track vibration acceleration, bridge strain and environment temperature are collected in real time through a multi-source sensor, and a multivariable time series data set is constructed after wavelet denoising and standardized preprocessing; and designing an LSTM network architecture on this basis, introducing an attention mechanism to dynamically allocate feature weights of each time step so as to enhance the ability to capture key signals in the track irregularity mutation and bridge resonance interval, and adopting a sparrow optimization algorithm to globally search an optimal combination of a hidden layer neuron number, a learning rate and a time step length in order to solve the problem of LSTM hyper-parameter optimization. Through the dynamic adaptive step length strategy balance algorithm, the early-stage global exploration and later-stage local development capabilities are balanced, the local convergence defect of a traditional grid search or genetic algorithm is avoided, the calculation efficiency can be remarkably improved, errors can be reduced, and the prediction precision can be improved.

The invention belongs to the technical field of virtual power plant scheduling, and particularly relates to a virtual power plant scheduling method based on multi-objective optimization. Aiming at the defects of slow convergence, easy falling into local optimum and the like of a single optimization algorithm adopted in a multi-objective optimization processing process of an existing virtual power plant scheduling method, the invention adopts the following technical scheme: the virtual power plant scheduling method based on multi-objective optimization comprises the following steps: S1, collecting related data required by virtual power plant scheduling; s2, constructing a multi-objective optimization model, and setting related constraint conditions; s3, an improved whale optimization algorithm is combined with a sequence least square programming algorithm to carry out optimization solution; and S4, generating an equipment control instruction according to an optimization result, and performing scheduling. The virtual power plant scheduling method based on multi-objective optimization has the beneficial effects that the limitation of a single optimization algorithm is overcome by combining the global search capability of the WOA and the local convergence efficiency of the SLSQP.

The invention provides a double-target medical waste transportation method based on an improved RSA-VNS, and relates to the technical field of path optimization. The method comprises the following steps: firstly, acquiring a medical waste transportation task and vehicle scheduling resources; secondly, on the basis of medical waste transportation tasks and vehicle scheduling resources, constructing a dual-target medical waste transportation model by taking minimization of cost and storage risk as optimization targets; and finally, solving the dual-target medical waste transportation model by adopting an improved RSA-VNS algorithm, and outputting a Pareto external file so as to decode and obtain an optimal medical waste transportation scheme. The storage risk of the medical waste is effectively quantified by integrating the service starting time of the medical institution, the type and strength of the medical waste and the possibility of disease transmission, and modeling is more in line with the characteristics of recovery and transportation of the medical waste. In addition, the proposed hybrid algorithm effectively improves the convergence speed of the algorithm and the diversity of the solution, and also enhances the local convergence capability at the end of the algorithm.

The invention provides a multi-magnetic accurate positioning method based on APSO-LM fusion, and solves the technical contradictions that global search and local convergence are unbalanced and the resolving efficiency is low in the existing multi-magnetic accurate positioning technology. The method comprises the following steps: constructing a magnetic moment calibration module, selecting a sensor array calibration point, collecting magnetic field data, and calculating a precise magnetic moment amplitude based on a magnetic dipole model; an NSS-VMGT initial positioning result is input through a parameter initialization module, APSO and LM algorithm parameters are configured, and particle vectors are normalized; generating a particle swarm by using an APSO global search module, and iteratively screening a global optimal parameter as an LM initial value; calculating a residual error and a Jacobian matrix by means of an LM local refining module, and iteratively optimizing and outputting a magnetic source position and a magnetic moment; and the dynamic compensation module is used for calibrating clock skew and inhibiting environment interference. The device is realized based on STM32H743, and comprises a 7 * 7 lis2mdltr sensor array, a dual-chip data acquisition link, an APSO-LM algorithm hardware acceleration module and an upper computer interaction unit.

The present application relates to lightning monitoring and strategy optimization technical field, especially a kind of lightning detection station layout method and system based on intelligent optimization algorithm.The site coordinates are used as the initial population satisfying the site position constraint of solution generation, and the individual corresponding to solution is 3N-dimensional vector, corresponding to the three-dimensional coordinates of N sites;Then population iteration is carried out through population mutation, crossing.This application can maintain the higher diversity of individual in the process of individual mutation, and randomly selects multiple parent individuals for mutation.Although the convergence speed is slow in high-dimensional, multi-peak experimental environment, it has stronger global search ability and the advantage of avoiding local convergence.The present application solves the problem that lightning positioning site distribution cannot consider terrain and cannot guarantee detection efficiency and accuracy in the prior art.

The invention discloses a power distribution control system of an intelligent low-voltage power distribution cabinet. The power distribution control system comprises a data acquisition module, a preprocessing module, a low-voltage power distribution cabinet fault monitoring model establishment module, a parameter tuning module and a power distribution control module. The invention belongs to the field of power distribution control, and particularly relates to an intelligent low-voltage power distribution cabinet power distribution control system, which performs kernel smoothing processing on a power distribution cabinet feature vector by introducing a lag attenuation coefficient, reduces the interference of noise on subsequent analysis, shortens the lag of the system on a sudden change condition, and is beneficial to finding potential faults in time. Time memory factors are introduced to design a power distribution shield loss function, frequent control caused by small-amplitude jitter of voltage and current is avoided, and power supply reliability is improved; continuously occurring abnormal conditions are restrained, so that the control accuracy of the low-voltage power distribution cabinet is improved; by optimizing the artificial fish swarm algorithm, parameters are dynamically adjusted, and local convergence is accelerated; fish school behaviors and reverse learning are used together, so that the control efficiency of the low-voltage power distribution cabinet is improved.

The invention discloses a mechanical arm tail end posture constraint path optimization method based on constraint manifold. A discrete constraint manifold point set which is high in posture precision, robust in singular point, uniform in space distribution and constrained in kinematics is generated based on a Monte Carlo method, a weighted damping pseudo-inverse method, an attenuation type learning rate gradient descent iteration method and voxel density optimization; based on measurement of a laser tracker, the relation between static environment information and a target point relative to a mechanical arm base coordinate system is constructed; a coarse path is obtained through a sampling method, and a feasible path on a constraint manifold meeting attitude constraint and having no collision is generated based on a search strategy of Euler distance difference + maximum joint angle variation + collision detection + index reconstruction criterion; and in combination with a multi-target weighted CHOMP algorithm, the feasible path is optimized into a collision-free and smooth path which meets the requirements of attitude constraint and kinematics constraint. According to the method, the problems of insufficient consideration of attitude constraint, low efficiency, local convergence and the like of traditional path planning are solved.

The invention provides a tool geometric parameter optimization design method and system based on integrated simulation and a storage medium, and relates to the technical field of tool geometric parameter optimization design. Orthogonal experiment design is combined with parametric modeling, tool geometric parameter combination space is fully covered, and invalid exploration is reduced; establishing a nonlinear relation model by using support vector regression, and accurately mapping parameter and performance association; based on a dynamic weight distribution mechanism of an entropy method and a judgment matrix method, the rough machining cost reduction requirement and the finish machining high-quality requirement are met; an improved simulated annealing algorithm introduces a dynamic cooling strategy and global solution test, global search and local convergence capabilities are balanced, and premature convergence is avoided; and through modeling-simulation-optimization closed-loop iteration, automatic parameter updating and feedback verification are realized, and the design period is remarkably shortened. Therefore, while the machining quality is guaranteed, the tool loss and the trial and error cost are greatly reduced, and a high-stability and high-economical-efficiency solution is provided for machining of complex materials.

The invention provides a dynamic target searching method for a plurality of unmanned underwater vehicles (UUVs) based on strategy adaptive fusion, and the dynamic target searching method comprises the following steps of: selecting a plurality of UUVs (Unmanned Underwater Vehicles) from a plurality of UUVs (Unmanned Underwater Vehicles); constructing a two-dimensional grid sea area and target trajectory set, and generating a target existence probability map; the weighted cumulative detection probability is used as the unified progress, the adaptive particle swarm is driven to synchronously optimize the position, the course and the speed, and parameter self-adjustment from global exploration to local convergence is achieved. When the weighted cumulative detection probability gain is lower than a threshold value, triggering partition snakelike traversal, and completing region assignment; and when the traversal is finished and the weighted cumulative detection probability does not reach 1, the target existence probability and the historical coverage degree are fused to construct a harvesting value map, and then the group is guided to converge to a high-value area. According to the method, three strategies of searching, traversing and harvesting are comprehensively planned through a single observable index, a closed-loop self-adaptive strategy fusion mechanism is formed, the coverage range and the local precision are considered, the searching success rate is increased, and the completion time is shortened.

The invention provides a robot safety interaction path planning method based on a dynamic danger criterion and bidirectional search, and the method comprises the steps: introducing the dynamic danger criterion in a planning stage, carrying out the quantification of a potential collision risk between a robot and a person, and comprehensively considering the inertia and rigidity of the robot, and the influence of the relative distance between the robot and the person; in combination with forward global search and reverse local search, a safe path is quickly generated in a complex environment. The forward search preferentially reduces danger indexes, the reverse search reversely avoids obstacles from a target position, the path continuity is ensured through an attitude potential energy function, and the local convergence problem of a traditional planning algorithm in an obstacle dense area is solved; the system also combines a real-time monitoring and safety control module, fuses a danger criterion, an obstacle potential field and a target potential field, and supports online path optimization in a dynamic environment so as to ensure that the robot can be always in a safe state in a final interaction process.

The invention discloses a Parkinson's disease prediction method based on adaptive federal learning and related equipment. The method comprises the following steps: step 1, initializing and broadcasting a global model; 2, user local training; step 3, the user uploads the local convergence speed of the local training to the server; step 4, the server adaptively calculates and adjusts the participation rate of the current round of communication, the selected users and parameters of the next round of local training of the users according to the local convergence speed, and sends calculation results to all the users; 5, updating the global model and broadcasting again; step 6, performing iterative training and model optimization; and step 7, predicting new patient data by using the global model obtained by training to realize early prediction of Parkinson's disease. According to the method, on the premise of ensuring the global model precision, the equipment with higher convergence speed and higher contribution degree is adaptively selected to participate in training, so that unnecessary communication overhead is reduced, and the model convergence efficiency is improved.

The invention discloses a wind turbine generator structured model all-condition parameter optimization method, which comprises the steps of generating N groups of PI parameter combinations as a population, initializing the population, constructing an improved fitness function, performing global search by adopting a genetic algorithm, and determining an improved fitness optimal solution region; an optimal solution output by the genetic algorithm is used as an initial solution set of a simulated annealing algorithm, disturbance is added to the initial solution set, and a new solution set is generated; calculating a fitness difference value delta F of the new solution set; if delta Flt; 0, accepting the new solution, otherwise, accepting the new solution with an improved acceptance probability P; and simulating an annealing algorithm to optimize the solution and outputting a globally optimal solution. The genetic algorithm quickly locates a potential optimal solution area, and the global search efficiency is remarkably improved; the simulated annealing algorithm dynamically adjusts the acceptance strategy for the inferior solution in the local search, the two realize the complementary enhancement of global exploration and local convergence, and the contradiction between the convergence speed and the globality of a single algorithm is broken through.

The invention discloses a user side long-time energy storage planning method and system of an electric power system, a medium and equipment, and the method comprises the steps: constructing an energy storage planning cost-operation scheduling two-stage optimization model; a user side electricity utilization pricing framework is established based on a two-part electricity price mechanism, and modeling is carried out for operation scheduling actions; a time-of-use electricity price driven long-time power scheduling strategy and a demand management strategy are introduced in the operation scheduling stage, and the long-time power scheduling strategy establishes charging and discharging constraints including peak period discharging priorities; according to the demand management strategy, the flexible power adjusting capacity of energy storage is used for restraining load capacity exceeding, and capacity control punishment is converted into measurable economic benefits; establishing a comprehensive benefit evaluation mechanism of energy storage as an evaluation basis of an energy storage configuration result; an improved SA-PSO algorithm is adopted to realize collaborative optimization of capacity configuration and operation scheduling, the global search capability of simulated annealing and the local convergence characteristic of particle swarm optimization are fused, and an energy storage optimal planning result is solved.

The present invention discloses a positioning method for a wireless sensor network based on hybrid RSS / AOA with unknown transmission power. Based on the RSS measurement model, the AOA azimuth measurement model, and the AOA elevation angle measurement model, an original non-convex objective optimization problem is constructed according to the least squares criterion; the semi-definite relaxation technique and the second-order cone relaxation technique are used to relax the variables or constraint conditions in the equivalent description after introducing variables in the original non-convex objective optimization problem constructed according to the least squares criterion, so as to obtain the final convex objective optimization problem; the hybrid semi-definite / second-order cone programming technique is used to solve the final convex objective optimization problem, and the position and transmission power of the target node are jointly estimated; the advantage is that it can ensure that the optimization of unknown variables can obtain the global optimal solution and is not affected by local convergence, thereby effectively improving the positioning accuracy and effectively suppressing the influence of measurement noise errors.

The invention provides a dynamic scheduling system and method for logistics human resources, and the method comprises the steps: constructing a decision index system for the multi-dimensional scheduling of the logistics human resources through core data in a logistics human resource scheduling scene; performing dynamic iteration optimization on a preset scheduling scheme of the logistics human resources according to the decision index system and a preset iteration number of the logistics human resource scheduling, and determining a global adjustment scheduling scheme of the logistics human resource scheduling; determining influence characteristics of the decision index system on the logistics human resources, and generating a dynamic matching scheme of logistics human resource scheduling based on the influence characteristics and the global adjustment scheduling scheme; and dynamically updating the dynamic matching scheme, and dynamically scheduling the logistics human resources by using the dynamically updated dynamic matching scheme. By adopting the scheme of the invention, the influence of local convergence defects caused by environmental complexity and decision limitation on dynamic scheduling of logistics human resources in a human resource scheduling process can be overcome.

The application discloses a motion platform fuzzy control method based on a genetic algorithm optimization, a computer device and a storage medium, relates to the field of fuzzy control, and solves the problems that the determination of initial values of PID parameters, membership functions and fuzzy rules is complicated, a particle swarm optimization algorithm is only suitable for continuous problems, and is prone to local convergence and unable to achieve a global optimal solution. The application provides the following technical scheme: a motion platform model is established, the lengths of multiple push rods are calculated through inverse solution of a target attitude on the platform; a cascade PID control system is designed, the cascade PID control system is formed by connection of an attitude angle fuzzy controller and a push rod cascade controller; the attitude angle fuzzy controller is designed; genetic algorithm optimization is carried out; a genetic algorithm optimization fuzzy PID control system is constructed; the artificial fuzzy rule fuzzy PID control system and the genetic algorithm optimization fuzzy PID control system are compared; and simulation control results are obtained. The application is suitable for setting of PID parameters of a motion platform control system and determination of fuzzy rules.

The application discloses a steady-state detection method based on two-fluid six-equation transient calculation, which comprises the following steps: before a detection time point, data required for local convergence judgment and TASS judgment are calculated and arranged; time local convergence of a system is judged according to enthalpy change of the system; after the convergence judgment is completed, TASS judgment is started on a historical state of the system, so that the system is ensured to completely enter a final stable state. The application improves the reliability under a low-energy and small-step working condition through a mixed relative convergence-time average judgment method, and meanwhile, the original steady-state detection efficiency is maintained.

The invention belongs to the technical field of training cost optimization, and discloses an unmanned aerial vehicle cluster-oriented federated learning training cost optimization method, which comprises the following steps of: jointly optimizing a local convergence threshold value, a local iteration frequency, computing resource allocation, bandwidth allocation and transmitting power allocation; the training cost (defined as a weighted sum of training energy consumption and training time) is minimized. The framework comprises a dichotomy-based joint optimization algorithm, each sub-optimization problem is solved alternately, and balance between energy and time is ensured. The core elements of the framework are as follows: in a system modeling stage, the training cost is defined as the weighted sum in the whole federal learning process, and a problem is expressed as a non-convex mixed integer programming problem.

The application discloses a passive time difference positioning method based on linear elimination weight and population center, mainly solves the problems that the artificial honeybird algorithm has resource waste and a large number of repetitive behaviors when solving the time difference positioning problem, thereby reducing the population richness, easily falling into local convergence and affecting the positioning accuracy. The implementation scheme is as follows: firstly, on the basis of elimination and update of the artificial honeybird algorithm, the current elimination weight is calculated by using an elimination weight calculation formula, and then the candidate solution is updated by using an elimination update equation. Secondly, based on the population center, the distribution state of the current population is judged by calculating the historical step length of each individual and the distance from the population center, when the population is in a redundant state, the population center is used to replace the invalid individual, and the problem that the positioning efficiency is not high due to a large number of invalid behaviors in the late iteration of the prior art is overcome.

The invention discloses a motion platform fuzzy control method based on genetic algorithm optimization, computer equipment and a storage medium, relates to the field of fuzzy control, and solves the problems that existing PID parameter initial values, membership functions and fuzzy rules are tedious to determine; and a particle swarm optimization algorithm is only suitable for a continuous problem, is easy to fall into local convergence and cannot achieve a globally optimal solution. According to the technical scheme, a motion platform model is established, and the lengths of a plurality of push rods are calculated through an upper platform target attitude inverse solution; designing a cascade PID control system, wherein the cascade PID control system is formed by connecting an attitude angle fuzzy controller and a push rod cascade controller in series; a constructed attitude angle fuzzy controller is designed; genetic algorithm optimization is carried out; and constructing a genetic algorithm to optimize the fuzzy PID control system, comparing the fuzzy PID control system of which the fuzzy rule is artificially formulated with the fuzzy PID control system optimized by the genetic algorithm, and simulating a control result. The method is suitable for PID parameter setting and fuzzy rule determination of a motion platform control system.

The invention relates to the technical field of thunder and lightning monitoring and strategy optimization, in particular to a thunder and lightning detection station layout method and system based on an intelligent optimization algorithm. The station coordinates serve as a solution to generate an initial population meeting station position constraints, individuals corresponding to the solution are 3N-dimensional vectors, and the 3N-dimensional vectors correspond to three-dimensional coordinates of N stations; and population iteration is carried out through population variation and crossover. In the individual variation process, multiple parent individuals are randomly selected for variation, and high diversity of the individuals can be kept. In a high-dimensional and multi-peak experimental environment, although the convergence speed is low, the method has the advantages of higher global search capability and avoidance of local convergence. The problems that in the prior art, lightning positioning station distribution cannot give consideration to the terrain, and the detection efficiency and precision cannot be guaranteed are solved.

The present application relates to the technical field of photonic device design, and especially relates to a photonic device reverse design optimization method, the present application adopts sin mapping, segmented linear chaotic mapping and reverse learning combined mode to generate initial population, enhances initial population diversity, reduces initial error and accelerates algorithm convergence; adopts Levy flight and teaching mechanism alternating parallel mode to accelerate algorithm convergence process, and adopts teaching factor decreasing with iteration number to further accelerate algorithm to find optimal solution; designs dynamic self-adaptive discovery probability, discovery probability linearly decreases in search late stage, and local optimal solution is more easily discovered, thereby accelerating local convergence; differential evolution retains individuals with higher fitness, so that population continuously approaches optimal solution, thereby increasing algorithm convergence speed; in addition, the present application adopts differential evolution mechanism to generate new individuals through differential calculation on individuals in population, thereby increasing population diversity and global search capability.

The application relates to the technical field of path planning, in particular to a crowd-sensing cooperative optimization method and system based on evolutionary fusion two-stage mixing. The method comprises the following steps: constructing a multi-agent cooperative optimization model of a heterogeneous space based on a mobile crowd-sensing operation scene; adopting a stage-type evolutionary fusion strategy to deeply fuse MOPSO and NSGA-II, and constructing an EF-DH algorithm; using the EF-DH algorithm to perform unmanned aerial vehicle multi-target path planning based on the constructed multi-agent cooperative optimization model, including first-stage unmanned aerial vehicle cluster path optimization and second-stage ground operation personnel task optimization; and performing air-ground cooperative execution and dynamic re-optimization based on the path planning. The evolutionary fusion two-stage optimization algorithm fusing MOPSO and NSGA-II is constructed, and the global search capability and local convergence performance of the multi-target optimization problem are effectively improved.

The invention discloses a family life cycle optimal consumption decision-making method based on an artificial intelligence algorithm, and belongs to the crossing field of family finance and artificial intelligence technologies. The method can assist families in making long-span rational consumption investment decisions in complex and uncertain environments, and is another scientific and technical method for financial service families. The method comprises the following steps: 1, constructing a decision-making model under the diversity uncertainty of the family life cycle, wherein the decision-making model comprises the uncertainty of labor income, accidental expenditure and financial asset yield; and 2, designing an artificial intelligence decision algorithm fusing a UCT algorithm and Q-learning, and optimizing cross-period consumption and investment decisions. And 3, by improving node return value calculation and a Q value dynamic updating mechanism, rapid convergence of a high-dimensional complex model is realized, the problems of local convergence and poor expansibility of a traditional algorithm are solved, and the prediction precision of the optimal decision of the family life cycle is remarkably improved.

The invention discloses an operation state regulation and optimization method for a power distribution network containing a distributed power supply and a flexible load, and the method comprises the following steps: obtaining a topological structure and operation parameters of the power distribution network, and constructing a power distribution network operation state regulation model according to the topological structure and the operation parameters; on the basis of a power distribution network operation state regulation and control model, a traditional knowledge acquisition sharing algorithm is improved, and a spiral updating strategy is introduced to guide a search individual to approach a current candidate optimal solution along a spiral trajectory; and performing iterative solution on the operation state regulation and control model by using an improved knowledge acquisition sharing algorithm to obtain operation optimization configuration of the power distribution network. According to the method, the global search capability and the local convergence precision of the algorithm are remarkably improved, so that the operation optimization configuration can be quickly and accurately obtained, and the stability, the adaptability and the overall operation performance of the power distribution network when the power distribution network deals with renewable energy output fluctuation and load change are enhanced.

The invention relates to the technical field of detection training, in particular to a mixing station intelligent detection method and system based on data analysis. According to the method, through combination of vibration signal characteristics jointly reflected by multiple parts of a stirring machine in a mixing station, signal components are decomposed and segmented, the attention situation of convergence trend is adjusted in a segmented mode, and the convergence situation of similar-rule vibration is comprehensively analyzed; obtaining a fault factor by combining the local convergence change of the work batch and the abnormal fluctuation characteristic condition in the local idling period; the model training of the work batch is adjusted through the fault factor, and a more reliable training model is obtained. According to the method, the abnormal characteristics are analyzed according to the condition that each historical working frequency has more obvious characteristics in the idling period between the local adjacent multiple stirring batches, so that the weight of the signal participating in prediction model construction is optimized, and the detection result of the mixing station is improved.

The invention belongs to the technical field of pig feed formula optimization, and particularly relates to a live pig daily feed formula optimization method based on improved SOA. According to the method, a nutritional requirement model is constructed based on SID amino acid indexes, and an improved snake optimization algorithm EISOA is provided to cope with complex problems such as multivariable coupling and multiple constraints in feed formula prediction. By introducing Latin hypercube sampling, an interactive learning mechanism and an elite guidance mechanism, the EISOA significantly enhances the global exploration capability and the local convergence precision. Experimental results show that the algorithm is superior to a comparison algorithm in the aspects of optimization precision, operation stability and feed cost control, and an efficient and feasible technical path is provided for realizing refined and intelligent feed formula design in the field of live pig breeding.

The application discloses an improved marlin algorithm-based virtual synchronous generator linear active disturbance rejection controller control parameter setting method and relates to the technical field of electric power. The linear active disturbance rejection control is used for improvement on the basis of virtual synchronous generator control. In view of the problem of linear active disturbance rejection control parameter selection, the marlin algorithm (SFO) is introduced to set parameters of the active disturbance rejection controller of the virtual synchronous generator model. Since the marlin algorithm is a group-based algorithm, it is very suitable for optimization problems without structure modification and can play a great role in parameter setting problems. On the basis, the improved marlin optimization algorithm (KSFO) is obtained, so that the marlin optimization algorithm has the advantages of faster global and local convergence and stronger optimization ability. The optimal setting of the active disturbance rejection controller control parameters is realized, and the method has great significance for improving power quality.

The invention relates to the technical field of reservoir group flood control scheduling, and discloses a cascade reservoir group flood control scheduling method based on multi-strategy fusion, comprising the following steps: determining an initial calculation condition; setting calculation parameters; adopting an artificial experience decision or a conventional method to generate a water level initial trajectory of cascade reservoir group flood control scheduling meeting constraint conditions, and calculating to obtain a neighborhood step length of an optimization range of each stage of each reservoir; starting iteration, decomposing the optimization scheduling problem into a plurality of two-stage sub-problems, and respectively executing an adaptive t distribution learning strategy and an adaptive immune clone selection strategy on each decomposed two-stage sub-problem in sequence; and stopping calculation, and outputting a final optimal track. According to the cascade reservoir group flood control scheduling method and system based on multi-strategy fusion, the defects of dimensionality disaster, premature convergence, local convergence and the like existing in the POA processing cascade reservoir group flood control scheduling problem are overcome, and the method and the system have excellent global exploration capability and local development capability.