Improved Parallel Iterative Chaotic Optimization Algorithm And Its Application

The traditional chaotic optimization algorithm is based on the Logistic sequence and adopts the serial iteration mechanism.It is difficult to find the global optimal solution for the optimization problem with a large range of independent variables and a high dimension of independent variables.At the same time,the search time is too expensive and it depends too much on the iteration times of the algorithm.Based on the above problems,a novel parallel chaotic optimization algorithm is proposed.For single-objective and multi-objective optimization problems,parallel chaotic optimization algorithms are divided into single-objective parallel chaos optimization(SOPCO)algorithm and multi-objective parallel chaos optimization(MOPCO)algorithm.Then the algorithms are applied to standard test functions and specific problems respectively,and good results are achieved.Firstly,eight common chaotic sequences are summarized an d their chaotic maps and Lyapunov exponents are analyzed.Different chaotic sequences are introduced into the chaotic optimization algorithm.The simulation results show that the chaotic optimization algorithm using Tent map has the best performance.On the basis of Tent map,the single-objective parallel chaotic optimization algorithm introduces parallel iteration and elite mechanism,and adopts a strategy of shrinking the range of independent variables according to iteration results and iteration times un der random probability,which significantly improves the global and local search performance of the algorithm.Through the simulation experiments of 23 standard test functions,SOPCO algorithm shows good optimization performance,and is superior to grey wo lf optimization,particle swarm optimization and differential evolution in general.Multi-objective parallel chaotic optimization algorithm follows the framework of single-objective parallel chaotic optimization algorithm as a whole.Non-dominant relation and crowding ranking operation are introduced into the algorithm.While ensuring the optimization performance of the algorithm,the algorithm obtains as many non-dominant solutions as possible,and guarantees the distribution and diversity of the optimal solutions to a certain extent.In order to test the performance of the algorithm,three performance evaluation indexes are introduced for eight typical multi-objective test functions.The simulation results show that MOPCO algorithm has more advantages than NSGA-II,SPEA2 and multi-objective Grey Wolf optimization in solving multi-objective optimization problems.The firepower allocation of the region air-denfense is not only a singleobjective optimization problem,but also an integer optimization problem.SOPCO algorithm distributes different weapons to defence different enemies according to the threat matrix of the enemies and the damage coefficient matrix of weapons,which maximizes the fitness value of our air-defense firepower allocation.At the same time,the result error of the algorithm is less than 1%,and the running time is less than 2 S.Aiming at the hot issues in the field efficiency measurement of AC servo motors,an equivalent circuit model of AC servo motors is established.According to the model,a multi-objective optimization problem is constructed.The problem has two objective functions and five independent variables.The multi-objective parallel chaotic optimization algorithm is used to solve the problem,and the field efficiency of AC servo motor is indirectly calculated.The error between the method and the actual measurement is less than 2%.Moreover,the method does not need to remove the motor from location or do some experimental items to obtain parameters.It has lowintrusive level and simple operation.