Research On Dynamic Optimization Based On Ion Motion Algorithms

In order to solve the dynamic optimization problem(DOPs)in reality,many scholars have proposed the dynamic optimization problem based on evolutionary algorithm,but these algorithms all have the problems of slow optimization speed and low convergence accuracy.In order to improve the ability of dynamic evolutionary algorithm to solve dynamic optimization problems,the main research contents of this paper are divided into two aspects: one is to propose a new dynamic optimization algorithm,the other is to apply the new dynamic optimization algorithm to solve practical problems,which can not only improve the theoretical system of dynamic algorithm,but also supplement the application system of the algorithm.Among them,the research on dynamic evolutionary algorithm includes two aspects: improving the evolutionary strategy of the algorithm and improving the dynamic processing technology.The following will introduce the research content in detail.In order to enhance the ability of dynamic evolutionary algorithm to solve dynamic optimization problems,a new dynamic optimization algorithm based on memory strategy(DIMOMS)is proposed in this paper.First of all,we need to determine the evolutionary strategy of dynamic optimization algorithm.In view of the superior performance of Ion Motion Optimization(IMO)in terms of convergence speed and accuracy,we adopt the powerful evolutionary strategy of IMO algorithm.However,the convergence speed and accuracy of IMO algorithm still need to be improved when solving function optimization problems,and the evolutionary strategy of evolutionary algorithm plays a decisive role in the performance of dynamic algorithm,which directly affects the optimization speed and accuracy of each environmental transient solution in dynamic optimization problems.In order to improve the speed and accuracy of the algorithm in searching the optimal solution for each environmental transient,the position updating of ionic individuals in the liquid and solid phases of IMO algorithm was improved,and the serial start-up conditions which are difficult to execute in the solid phase were modified to parallel mode,and the start-up conditions at the early stage of evolution were relaxed.An improved Ion Motion Operating algorithm was proposed(I-IMO).The experimental results show that the convergence speed and accuracy of I-IMO algorithm are obviously better than those of the original algorithm and other excellent algorithms.Secondly,the dynamic processing technology,i.e.memory strategy,is studied in this paper.Because the dynamic processing technology reflects the processing ability of evolutionary algorithm for dynamic environment transformation,in order to better integrate with I-IMO algorithm andenhance the performance of dynamic algorithm,the memory strategy is improved by selecting memory individuals according to probability,and the forward and backward memory population is constructed.The strategy of cosine similarity ranking to update the evolutionary population and the method of cosine similarity ranking to increase diversity after environmental changes are used.Finally,the evolutionary strategy of I-IMO algorithm is fine-tuned and the improved dynamic mechanism is effectively integrated.The test results of standard dynamic test functions show that the proposed dynamic optimization algorithm has faster searching speed,higher accuracy and more stable performance in each environment than other dynamic optimization algorithms with better performance.DIMOMS algorithm is used to solve dynamic data stream clustering problem.The core purpose of DIMOMS algorithm is to automatically group a large number of chaotic data streams,so that each group is composed of similar data,which can effectively help cluster the hidden information in the data stream.The experimental results show that compared with the comparison algorithm,the DIMOMS algorithm is more effective in clustering.