37 results about "Gray wolf" patented technology

Embodiments of the invention disclose a multi-target-based improved gray wolf optimization algorithm which is used to solve the technical problems that a standard gray wolf algorithm falls into a local optimal value easily and has a low convergence speed and other defects while processing a multi-target optimization problem in the prior art. The method of the embodiments comprises the following steps: S1, setting a wolf pack initialization parameter and a direction correction probability, and randomly initializing wolves' individual positions; S2, calculating an adaptability value of each wolf individual according to a solving target, and selecting the three wolf individuals ranking top; S3, optimizing the wolves' individual positions of a wolf pack, generating moderate wolves, and updating a wolf pack position; S4, executing direction correction operation on the updated wolf pack, controlling the upgraded wolf pack to participate correction of the size of dimensions according to the direction correction probability, and obtaining a corrected wolf pack position; and S5, determining whether an iteration frequency reaches a preset maximum iteration frequency, outputting the corrected wolf pack position as a final optimization result if the iteration frequency reaches the preset maximum iteration frequency, and, if the iteration frequency does not reaches the preset maximum iteration frequency, turning to the S3 so as to continue performing iteration searching.

Disclosed are a variable weight grey wolf algorithm optimization method and an application. According to the method, social classes are set and act on the whole search and predation processes of a grey wolf population, and the grey wolf population surrounds a target in the search process and surrounds the target in the center in the predation process; and in the iterative search process, the positions of an alpha grey wolf, a beta grey wolf and a delta grey wolf with high social classes in the population are firstly, secondly and thirdly close to the target all the time, and in the iterative process, the positions of the grey wolfs in the population are described by combination of variable weight functions of the alpha grey wolf, the beta grey wolf and the delta grey wolf, wherein the weight w1 of the position of the alpha grey wolf is gradually reduced to 1 / 3 from 1, the weight w2 of the beta grey wolf and the weight w3 of the delta grey wolf are gradually increased to 1 / 3 from 0, and the use of w1, w2 and w3 meets the requirements that the sum of w1, w2 and w3 is equal to 1, w1 is greater than or equal to w2, and w2 is greater than or equal to w3. The method has the advantages that the search process is remarkably accelerated and the optimization calculation can be finished more quickly.

The invention discloses a grey wolf algorithm-based rectangular part layout method and device, and belongs to the field of structural part optimization blanking. The grey wolf algorithm-based rectangular part layout method comprises the following steps: setting an algorithm target function and constraint conditions; performing wolf pack initialization, decoding and positioning based on a lowest horizontal line method of fitness, solving a material utilization rate and determining alpha wolves through grade sorting; performing all grey wolves walk, solution of the material utilization rate andalpha wolf updating; performing the grey wolf attacking preys, solution of the material utilization rate and alpha wolf updating; and performing loop iteration until a termination condition is reachedto obtain an optimal solution, namely the position code Xalpha of the alpha wolf and the prey odor concentration Yalpha felt by the alpha wolf, namely the optimal layout scheme and the correspondingmaterial utilization rate. The grey wolf algorithm-based rectangular part layout method can improve the plate utilization rate and the solving speed, and has good practicability and applicability in solving the rectangular part layout problem.

The invention belongs to the technical field of data analysis and discloses a kernel incremental transfinite learning machine and a differential multi-population gray wolf mixed optimization method. Aiming at the problem that the kernel incremental transfinite learning machine (KI-ELM) has the redundant nodes with low learning efficiency and poor accuracy; At first, that invention utilize the differential evolution algorithm and the multi-population grey wolf optimization algorithm to propose a hybrid intelligent optimization algorithm--the differential multi-population grey wolf optimizationalgorithm, optimizes the node parameter of the hidden layer, and determine the effective node quantity, so as to reduce the network complexity and improve the learning efficiency of the network; Secondly, the depth structure is introduced into the kernel incremental transfinite learning machine, and the input data is extracted layer by layer to realize the high-dimensional mapping classification of the data and improve the classification accuracy and generalization performance of the algorithm. The simulation experiment results show that the hybrid intelligent depth kernel incremental transfinite learning machine provided by the invention has good prediction accuracy and generalization ability, and the network structure is more compact.

The invention provides an intelligent vehicle path tracking control system and method based on a Gray Wolf Optimizer (GWO). The intelligent vehicle path tracking control system includes a path planning system, a path tracking decision system and a vehicle chassis execution system which are connected in sequence, wherein the path tracking decision system includes two subsystems of a driver prediction controller and a GWO attitude controller, a preview-follower driver model based on the vehicle side acceleration feedback correction is adopted in the prediction controller, and the GWO is adoptedin the GWO attitude controller. According to the intelligent vehicle path tracking control system and method, an ideal path is planned by the path planning system, the path tracking decision system receives the ideal path information, the driving state of a vehicle is adjusted through a wire-controlled steering actuator and a hub motor brake actuator of the vehicle chassis execution system, the accurate tracking of the ideal path is realized, and the driving stability of the vehicle is improved.

The invention relates to a gray wolf optimization algorithm based frequency modulation PID control method of a doubly-fed induction wind-powered generation unit, and belongs to the technical field ofpower system control. The method comprises the following steps: performing PID control on parameters of a current converter at the power grid side of the doubly-fed induction wind-powered generation unit in an offline state; comparing the reference rotor speed of a generator and the reactive powder with the actual values; inputting the difference into the PID control to obtain corresponding reference current; comparing the obtained reference current with the actual current; inputting the difference into the PID control to obtain rotor voltage; optimizing the PID control link through the gray wolf algorithm; outputting the obtained voltage to the outside of the controller by adding compensation quantity; performing frequency modulation on the wind-powered generation unit through the outputvoltage; recycling the frequency modulation of the wind-powered generator in order to realize continuous tracking of the maximum power point. With the adoption of the method, the problem of local optimization of the maximum power point tracking target function in the frequency modulation process of the doubly-fed induction wind-powered generation unit can be solved.

The invention discloses a bridge crane PID control method based on a grey wolf algorithm. An adopted PID controller structure is a double-closed-loop structure, collected swing angle and displacementlength of goods in a crane operation process are taken as an input of a control system, a difference between actual work and a target value is taken as an output, and the PID control of the crane is carried out. An inner loop is an angle control system, a target swing angle is set, the closed-loop feedback is carried out to adjust a swing angle according to a swing angle error between an actual working swing angle and a target swing angle, and the swing angle error is controlled within a certain angle. An outer loop is a displacement control system with the same feedback principle as the innerloop. According to the method, the complicated mechanical model is rationally simplified to a two-dimensional plane, the global search property of a gray wolf bionic algorithm is used to quickly findoptimal control parameters, a closed-loop control strategy is proposed, the intelligent operation of a large-scale working device bridge crane is achieved, the problems of large swing angle and slowpositioning speed of a bridge cane in the operation process are overcome, and the work efficiency and safety performance are improved.

The invention belongs to the technical field of computers and provides a data analysis method and device based on a grey wolf optimization algorithm. The method comprises steps of acquiring to-be-analyzed data; according to the preset searching space, through the gray wolf optimization algorithm, calculating a penalty coefficient and a nuclear width of a wolf group; and according to the penalty coefficient, the nuclear width and a preset analysis model, analyzing the to-be-analyzed data so as to obtain an analysis result. According to the invention, by introducing a reverse learning mechanism,the penalty coefficient and the nuclear width of the wolf group are calculated in the gray wolf optimization algorithm, so a parameter optimal solution of the preset analysis model is obtained. Thus,the specifies of the gray wolves in the wolf group are extended; the global detection ability of the algorithm is improved; and by analyzing the to-be-analyzed data is performed through the analysismodel, so the precise analysis result is rapidly obtained.

The invention relates to a robust multiuser detector design method, which solves the technical problem of high error rate of the traditional robust multiuser detector in an impact noise channel environment. The robust multiuser detector design method comprises the steps of: initializing algorithm parameters; using an opposition-based learning method to initialize a parent population, and determining three wolves in the parent population; updating the parent population by adopting an improved gray wolf algorithm position updating equation, and sorting population individuals according to fitnessvalues from large to small; generating offspring crossover mutants by utilizing the parent population, performing position information differential operation on an evolutionary direction of the offspring crossover mutants and successful crossover mutation probabilistic information when the fitness values of the offspring crossover mutants are superior to that of the parent population, acquiring new evolutionary direction information and saving the new evolutionary direction information, and updating positions of the three wolves. With the adoption of the robust multiuser detector design method adopting a Huber theory and utilizing a residual non-rapid-increasing function to design a multi-user detector in an impact noise channel, the mentioned problem is effectively solved, and the robustmultiuser detector design method can be used in multi-user detector design.

The invention relates to a photovoltaic MPPT (Maximum Power Point Tracking) method based on an improved grey wolf optimization algorithm. According to the technical scheme, the method comprises the steps: S1, selecting N random values from [0, 1] to serve as the initial position of a grey wolf population, wherein grey wolf individuals represent the duty ratio in a photovoltaic system; S2, acquiring the output power of the photovoltaic system under the control of each duty ratio, taking the output power as the fitness of gray wolf individuals, and recording the three gray wolves with the maximum fitness as alpha wolf, beta wolf and delta wolf in sequence; S3, shrinking the search range of the grey wolf population, and updating the upper and lower limits of a search interval; S4, performing position updating on the grey wolf population by using an improved position updating formula; S5, executing reverse search, comparing with gray wolf population fitness, and updating alpha wolf, beta wolf and delta wolf; S6, judging whether a termination condition is met or not, and if yes, stopping iteration and stabilizing the photovoltaic system at the duty ratio corresponding to the alpha wolf; otherwise, returning to the step S2, and continuing iteration.

Disclosed is a WSN three-dimensional coverage enhancement method based on an improved grey wolf optimizer. The contribution degree of an alpha wolf with the most abundant experience is enhanced by theweight calculation formula in the step 3, so that the hunting process of the wolf group is more accurate; to solve the defect that the early global search performance of a classic grey wolf algorithmis poor, the dynamic convergence factor in the step 4 is defined, so that the global optimization capability of the algorithm is enhanced while the complexity of the algorithm is not sacrificed; fora coverage rate threshold of a classic virtual force algorithm, the dynamic stepping formula in the step 4 is defined, so that the algorithm is higher in early optimization speed and more accurate inlater convergence position. To solve the defect that the earlier-stage optimization speed of a classic grey wolf algorithm is relatively low, a virtual force factor is introduced into the wolf pack position updating formula in the step 4, so that the optimal guidance quality of the global search process of the algorithm is enhanced, and the convergence speed of the algorithm is increased.

The invention relates to an Alpha-guided gray wolf algorithm and other applications in two-stage operational amplifier design, belonging to the improvement and application field of swarm intelligencealgorithm. It is characterized in that: Alpha guidance mechanism is introduced into the grey wolf algorithm, which provides the direction of potential prey for the wolf pack leading group, and furtherimproves the convergence speed of the algorithm; as that standard GWO fall into the local optimum for a long time, the population concentrates in the area near Alpha, Beta and Delta and is difficultto jump out of the area. The invention introduce Alpha mutation mechanism into the grey wolf algorithm, and provides a channel for the population to jump out of the local optimum; In the invention, the algorithm parameter adopts the scalar design, and the algorithm structure is simplified. In the optimization design of two-stage unbuffered CMOS operational amplifiers, it has a good effect of optimization.

The invention discloses a path planning method based on a grey wolf algorithm, and the method comprises the following steps of S1, obtaining the geographic position information of each city, calculating the distance between any two cities, and constructing a graph G (V, E); S2, setting the maximum number of iterations, and initializing the current number of iterations to be 1; S3, describing a plurality of initial solutions of a TSP problem according to the representation form of the set; S4, calculating the adaptive values of the initial solutions; S5, updating the current position and the adaptive value of each grey wolf; S6, operating according to the updated adaptive value of the grey wolf solution; s7, updating the number of iterations, and adding 1 to the current number of iterations; S8, judging whether the algorithm reaches a termination condition or not, if the algorithm does not meet the termination condition, skipping to S5 to continue, and if the algorithm reaches the termination condition, continuing the subsequent steps; and S9, defining the position of the alpha wolf as an optimal solution meeting the termination condition, wherein the position of the alpha wolf is an optimal path, and the adaptive value of the alpha wolf is a path value.

The invention discloses an intelligent electric meter wireless sensor network layout method based on an improved grey wolf algorithm. The method comprises the following steps: S1, setting wolf populations including an original population and an assistant population, and initializing parameters including an original population scale M, an assistant population scale M, a maximum iteration frequencyIntermax, a search space dimension N, a search space upper bound ub and a search space lower bound lb; S2, initializing grey wolf positions of the original population and the assistant population; S3,calculating the fitness value of each grey wolf in the original population and the assistant population, and respectively selecting the wolf with the best fitness value in the first three populationsas a decision wolf; S4, disturbing the positions of the decision wolves in the original population and the assistant population; S5, sorting all decision-making wolves in the original population andthe assistance population, and selecting the first three wolves with the highest fitness values; S6, respectively updating the positions of the original population and the assistant population; S7, judging whether an iteration termination condition is met or not: if so, outputting the position of the alpha wolf as a layout coordinate of the wireless sensor, if not, repeating the steps S3 to S6.

The invention discloses a thermal error prediction model method for an electric spindle with variable bearing pre-tightening force, which comprises the following steps: constructing an electric spindle temperature field model, and analyzing the temperature of a heat source and the temperature of a key component; establishing a motorized spindle statics finite element model by using different pretightening force conditions and spindle component parameters changed due to temperature change under the conditions, and analyzing the relationship between the thermal error of the motorized spindle and the pretightening force and the temperature; establishing a grey wolf optimization algorithm (GWO) model, adopting a mode of randomly generating a grey wolf population, initializing positions of alpha, beta and delta wolf of a grey wolf pack, globally searching a fitness optimal value of each body of the wolf pack, and searching a penalty factor (C) and a kernel function width (g) of a support vector regression (SVM) model; building a thermal error prediction model of the SVM variable pre-tightening force motorized spindle, and training the model to enable the model to reach training precision; and finally, through comparison between a BP neural network thermal error prediction model and a DE-GWO-SVM thermal error prediction model, showing that the method has better performance compared with a traditional model.

The invention discloses a generator excitation system parameter identification algorithm based on an improved grey wolf algorithm. The generator excitation system parameter identification algorithm comprises: establishing an original model and an actual system model of an excitation system in a no-load state; identifying the actual system model entering the linear region through an improved grey wolf algorithm to obtain linear part parameters; inputting the linear part parameters into an actual system model; and identifying the actual system model which enters the non-linear region and is substituted with the linear part parameters through an improved grey wolf algorithm to obtain the non-linear part parameters. On the basis of the grey wolf algorithm, a convergence factor nonlinear decreasing strategy and a grey wolf grouping alternate chasing strategy are provided, population diversity of wolf groups is enhanced, and the defect that the algorithm is prone to falling into local optimum is overcome. The grey wolf algorithm is applied to identification of excitation system parameters, identification of the excitation system parameters is effectively achieved by improving the grey wolf algorithm, and an identification result proves that the identification precision and stability of the improved grey wolf algorithm are superior to those of a traditional grey wolf algorithm.

The invention provides a fixed-wing unmanned aerial vehicle cluster collaborative path planning method based on an optimal pheromone grey wolf algorithm. The method comprises the following steps: s1, establishing a three-dimensional map containing M threat sources; s2, establishing a cluster route planning cost function; s3, planning the tracks of all fixed-wing unmanned aerial vehicle clusters by adopting a grey wolf algorithm based on optimal pheromones. According to the method, a cooperative flight path planning problem of a fixed-wing unmanned aerial vehicle cluster in a three-dimensional scene is solved by adopting an optimal pheromone grey wolf algorithm, and a plurality of optimal grey wolves are selected based on a wolf pack grade system in the algorithm to update pheromones on flight path points; wherein, the exploration stage of each gray wolf follows the guidance of pheromones so as to improve the convergence speed and the solution stability of the algorithm, and the rotation angle constraint of the unmanned aerial vehicles and the collision avoidance constraint between the unmanned aerial vehicles are met. According to the technical scheme, the problem of fixed-wing unmanned cluster collaborative path planning in a high-dimensional and multi-constraint complex scene is solved.

The invention relates to a product disassembly sequence optimization method and device, computer equipment and a storage medium. The method comprises the steps: enabling parts contained in a to-be-disassembled product to generate a population according to a grey wolf algorithm, wherein one population comprises a plurality of grey wolves, different grey wolves have different gene sequences and represent different alternative schemes of the to-be-disassembled product, the gene sequence of each grey wolf is a sequence composed of identifiers of all parts of a to-be-disassembled product, and different gene sequences correspond to the disassembling sequence of the parts in the to-be-disassembled product; in the current iteration process of the population, determining the grade of each grey wolf based on the gene sequence of each grey wolf in the population, and updating the gene sequences of all grades of grey wolves to obtain the population of the next iteration; and determining the gene sequence of the grey wolf at the highest level in the population after the iteration process is ended as an optimal sequence, wherein the optimal sequence is used for disassembling the to-be-disassembled product. In the way, generation of a local optimal solution can be inhibited to a certain extent, and finding of a global optimal sequence is facilitated.

The invention relates to a camera calibration method based on Levy flight and mutation mechanism grey wolf optimization, and the method comprises the following steps: S1, building a nonlinear camera model, and determining calibration parameters; S2, setting the number of grey wolf populations and the maximum number of iterations; S3, acquiring a calibration parameter upper limit and a calibrationparameter lower limit of the camera, and generating a grey wolf position; S4, establishing a target function, and obtaining a back projection error and positions of an alpha wolf, a beta wolf and a gamma wolf; S5, generating a grey wolf intermediate by using a grey wolf optimization algorithm based on Levy flight and a mutation mechanism; S6, updating the position of the grey wolf, and judging whether iteration is continued or not; and S7, obtaining a back projection error according to the target function, wherein the grey wolf position with the minimum back projection error is an optimal calibration parameter. Compared with the prior art, the algorithm can be combined with actual engineering cases, can be accurately and effectively used for multi-dimensional nonlinear problem optimizationsolution, effectively improves the calibration precision, and has good stability and accuracy.

The invention discloses a feature selection method of a grey wolf optimization algorithm based on a fused random black hole strategy. The method comprises the steps of initializing a grey wolf population, obtaining the position update of the grey wolf population, updating the positions of the three head wolves by fusing a random black hole strategy, solving the relative distance between the threehead wolves and the remaining wolves, adjusting the positions of the remaining wolves, obtaining the position of the current wolf after the position update according to the approximate distance, and predicting the value of the optimal solution according to the positions of the three head wolves. The method provided by the invention is used in the intrusion detection process of the industrial control system, the characteristics of the intrusion detection data set are calculated, analyzed and selected, and then the characteristic sequence which is beneficial to improving the intrusion detectionaccuracy and efficiency of the industrial control system is obtained, so that the intrusion detection accuracy and efficiency can be effectively improved.

The application of a variable weight gray wolf algorithm optimization method, including setting social class to act on the whole process of gray wolf population search and predation, and the gray wolf population surrounds the target during the search process, and surrounds the target at the center during the predation process; During the iterative search process, the positions of α gray wolves, β gray wolves and δ gray wolves with high social class are always the first, second and third closest to the target in the population, and the positions of each gray wolf in the population are changed from α gray to The variable weight function combination of wolf, β gray wolf and δ gray wolf is described, in which the weight w1 of α gray wolf position gradually decreases from 1 to 1 / 3, and the weights w2 and w3 of β and δ gray wolf gradually increase from 0 to 1 / 3, and use w1+w2+w3=1 and w1≥w2≥w3. Advantages: Significantly speeds up the search process and can complete optimization calculations faster.

The invention discloses a power control method for a wave power generation device based on Fourier analysis and an improved gray wolf algorithm. On the basis of retaining the most essential characteristics of the gray wolf optimization algorithm, a brand-new elite wolf search strategy and an overall wolf group expansion strategy are added to optimize the The encirclement formation strategy, the wolf pack search method and the wolf pack structure ensure that the improved grey wolf algorithm can avoid falling into local optimum due to the nonlinearity of float hydrodynamics. The Fourier analysis method is introduced to decompose the ocean incident wave and the response of the motor moving parts. For each wave component in the frequency range captured by the float, the improved gray wolf algorithm is used to solve the corresponding optimal motor control parameters to maximize the capture of the power carried by it. This enables maximum power point tracking.

The invention discloses a parameter tuning method of the speed loop self-disturbance rejection controller of a permanent magnet synchronous motor, establishes a double closed-loop control structure of speed and current, and initializes the gray wolf population strategy according to the Tent mapping reverse learning, and adopts the non-linear method as the number of iterations increases Change the convergence factor, and introduce the levy flight strategy in the update link of the wolf pack position in the algorithm for mutation operation, and finally obtain the adjusted parameters. The present invention improves the gray wolf optimization algorithm population from the initialization, initializes the gray wolf population according to the tent mapping reverse learning strategy, designs a convergence factor that changes non-linearly with the increase of the number of iterations, and introduces the levy flight strategy in the wolf group position update link. An improved gray wolf optimization algorithm is proposed, which can increase the diversity of the initial population of the algorithm, better adaptability and adjustment to complex searches, avoid falling into local optimum, improve the convergence speed of the algorithm and the global optimization ability, and the effect It is better than other improved gray wolf optimization algorithms.

The invention discloses a grey wolf optimization method based on a dimension learning strategy and Levy flight, and belongs to the technical field of system identification. The grey wolf optimization method based on the dimension learning strategy and the Levey flight aims to construct a model wolf by using the dimension learning strategy in three leading wolves, and solves the problems that global exploration ability is weakened and early maturity convergence is possibly generated due to the fact that a wolf group learns to the model wolf by combining the Levey flight. The method comprises the following steps: constructing a model wolf in a leader layer of a grey wolf optimization method by using a dimension learning strategy; applying the model wolf in the step 1 to wolf pack updating, and a grey wolf optimization method based on a dimension learning strategy is designed in combination with Levy flight. The performance of the standard grey wolf optimization algorithm is greatly improved, and the convergence speed and the convergence precision of the grey wolf optimization algorithm are improved.

The invention discloses a virtual network function deployment method based on gray wolf algorithm, which includes the following steps: (1) input the node information and link information of the underlying network, input the service function chain, and the service function chain request includes multiple sequential Constrained virtual network functions, the resource requirements of all virtual network functions deployed on the underlying network computing nodes in the service function chain request are taken as the first constraint condition of the service function chain deployment scheme, and the relationship between all virtual network functions in the service function chain request The bandwidth requirement of the service function chain is used as the second constraint condition of the service function chain deployment scheme, and the total delay of the service function chain deployment is taken as the optimization goal of the service function chain deployment plan; (2) Initialize the gray wolf population; (3) Evaluate the gray wolf population; (4) Update the location information of each individual gray wolf in the gray wolf population; (5) Determine whether the maximum number of iterations is reached, and output the globally optimal gray wolf. The invention has high efficiency, short time and wide application range.

The invention discloses a random forest and grey wolf optimized coal body gas content prediction method, which comprises two stages, namely construction of a random Sendor prediction model and use ofa grey wolf optimization algorithm, and aims to select and determine optimal values of relevant parameters of the random forest prediction model by using the grey wolf optimization algorithm. Therefore, the identification and prediction effects of the media gas content prediction model are enhanced. The gas basic data is initialized through two steps of missing value filling and feature selection.The geological characteristics of the coal bodies in different regions are different, and the characteristic selection can be used for carrying out targeted selection on the mine coal body data in different regions, so that the gas content prediction accuracy is enhanced, and the method is suitable for mines in different regions. According to the method, effective prediction of the gas content can be achieved, the working efficiency is improved, meanwhile, implementation of gas disaster prevention measures in the coal mining process can be guided, and the method has important significance inprevention and control of gas disaster accidents and accurate mining of coal.

The invention discloses a rectangular piece layout method and equipment based on the gray wolf algorithm, and belongs to the field of optimized blanking of structural parts. The method includes: setting the algorithm objective function and constraint conditions; initializing the wolf pack, decoding and positioning the lowest level method based on fitness, obtaining the material utilization rate and ranking the alpha wolf; all gray wolves wandering, obtaining the material utilization rate and α wolf update; gray wolf siege prey, obtain material utilization rate and α wolf update; loop iteration until the termination condition is reached to obtain the optimal solution, that is, the position code X of α wolf α and its perceived prey odor concentration Y α , which is the optimal layout scheme and its corresponding material utilization rate. The rectangular piece layout method based on the gray wolf algorithm of the present invention can improve the utilization rate of the plate and the solution speed, and has good practicability and applicability in solving the rectangular piece layout problem.