Research On Swarm Intelligent Optimization Method And Application Of Power Vision Technology

With the development of information technology,many optimization problems have appeared in computer,artificial intelligence,communication,automation and other fields.Simple methods(Newton method,gradient descent method,conjugate gradient method)have been unable to adapt to the requirements of obtaining the optimal solution quickly.Therefore,the application of swarm intelligence optimization algorithm in the study of such problems has attracted more and more scholars’ attention.The core of swarm intelligence optimization algorithm is derived from the simulation of biological behavior in nature,and the optimal solution can be obtained by searching for the optimal value in global exploration and local development.Compared to other algorithms,the swarm intelligence optimization algorithm has simple structure,high efficiency and easy to implement.However,the traditional optimization algorithm has some defects,resulting in low convergence accuracy and slow speed.Although the principle is simple and easy to program,there are also many shortcomings.This paper mainly studies the theory and application,and the research contents are as follows:(1)An improved algorithm combining elite disturbance opposition-based learning and dynamic spiral updating(OWOA)was proposed.Firstly,the whale population is initialized by opposition-based learning strategies,and the elite value is obtained by sorting,and then the elite whale is randomly disturbed to avoid falling into the local optimal solution.Secondly,a dynamic spiral updating strategy is adopted to dynamically adjust the spiral shape with the increase of iterations to improve the optimization accuracy of the algorithm.Finally,the experiment was verified by the function test set and achieved good results.(2)A Marine Predator algorithm based on adaptive weight and chaos factor was proposed(ACMPA).Firstly,the overall performance of the improved adaptive weight strategy balancing algorithm is improved through the early exploration and later development stages.Secondly,the Logistic chaos factor was used to replace the random factor,and the ergodic characteristics of the chaos factor was used to make it easier for predators to jump out of the local optimum and enhance the overall performance of the algorithm.(3)A whale optimization algorithm(TWOA)which integrates tent chaos and nonlinear convergence factor is proposed.Firstly,the ergodicity of tent chaotic map is used to make the individual distribution of whale population more uniform in the random walk stage,which enhances the global searching ability of the algorithm.Then,the linear convergence factor is changed to nonlinear convergence factor.which makes the algorithm in the exploration stage search more comprehensive,improve the algorithm optimization accuracy.Experimental results show that the overall performance of the algorithm is greatly improved after the addition of the improved strategy.(4)A tracker based on improved whale optimization(TWOA)is proposed.The improved whale optimization algorithm is used as the search strategy in target tracking,and the application is verified by experiments.The experimental results show that the proposed tracker has a good effect in tracking.This paper focuses on whale optimization algorithm and Marine predator algorithm,It focuses on how to improve the search ability of the algorithm in the global stage,when the population individual does not find the optimal solution,and applies the improved whale algorithm to the target tracking application of power equipment.