Research Of A Novel Atmosphere Clouds Model Optimization Algorithm And Its Application

With the development of science and technology, more and more industrial areas are inseparable from the optimization design, especially in the precision high-end products with highly integrated of automation, the optimization method with excellent performance is needed urgently. In recent years, based on the inspiration from a variety of phenomena and rules of nature, researchers proposed numerous intelligent optimization methods to settle with complex optimization problem, such as genetic algorithm(GA), particle swarm optimization algorithm(PSO), ant colony optimization(ACO), etc. in the application of multi-model functions, for the global search ability, the algorithms existing, include the improved algorithms are mainly using the forward search(the whole population convergent from the search space to the global optimal position), complemented with various methods for improving the population diversity of algorithm and easily escaping from the local optimal.Based on the behavior rules of cloud in the natural world, this paper proposed a novel stochastic optimization algorithm-Atmosphere Clouds Model Optimization Algorithm (ACMO), which is tried to simulate the generation behavior, move behavior and spread behavior of cloud in a simple way. The reverse search method, composed by the move behavior and spread behavior of clouds, disperses the whole population to the search space. As the global search method of ACMO algorithm, this method can enhance the diversity of population; the generation behavior is mainly used to search in the vicinities of current global optimal, keeping the convergence of ACMO algorithm.A comprehensive set of benchmark multimodal functions was used to test the performance of ACMO algorithm. The simulation experiment has two parts. The first past tested the effects of several control parameters to the ACMO algorithm; the second part compared the results obtained by ACMO algorithm with Particle Swarm Optimization algorithm (PSO) and Genetic Algorithm (GA). The results demonstrate that the ACMO algorithm is an effective method in solving multimodal optimization problems, while the PSO algorithm is superior to the other algorithms in the convergence precision. Anyway, ACMO algorithm is an effective method to solve multimodal function optimization problems.In the last, this paper used ACMO algorithm in the PID parameters tuning to verify the efficiency of algorithm. The simulation has done by combining the m files programming and the Simulink in MATLAB. Through the result comparison with PSO algorithm and GA algorithm, ACMO was proofed that the PID parameter tuning method based on ACMO algorithm is an efficient method and it has utility value.