Parameter Adaptive Multi-objective Harmony Search Algorithm And Its Application

With the improvement of social productivity and computer development level,more and more complex multi-objective optimization problems have emerged.Using traditional optimization algorithms to deal with these problems often does not achieve satisfactory results.A new heuristic global search algorithm-harmonic search algorithm is widely used in engineering optimization problems because of its simple logic,few control parameters and strong global search ability.A large number of research results show that the harmony search algorithm has better application effect than traditional mathematical optimization methods and some traditional intelligent optimization algorithms.Therefore,combining the ability of harmony search algorithm to deal with high-dimensional problems and applying it to solving multi-objective optimization problems is the trend of this algorithm research field.The traditional harmony search algorit hm has some shortcomings: the lack of flexibility in parameter setting leads to the algorithm being easy to fall into local optimum and the convergence speed is slow,so the performance of the harmony search algorithm needs to be improved.This paper focuses on the improvement of the performance of the harmony search algorithm and the optimization problem in the distributed stability control of Distributed Drive Electric Vehicles(DDEV).The research content can be summarized as follows:This paper proposes a Parameter Adaptive Multi-objective Harmony Search(PSAMOHS)algorithm.The PSAMOHS algorithm combines the chaos idea and uses the chaotic map to control the adaptive adjustment of the Pitch Adjusting Rate.At the same time,the acoustic memory candidat e probability and bandwidth are dynamically adjusted according to the information of the solution diversity and the number of iterations in the harmony memory,respectively.In order to obtain a more diverse Pareto front when the harmony search algorithm solves multi-objective problems,the truncation program and fast non-dominated sorting are applied to the update of the harmony memory.Simulation results show that the PSAMOHS algorithm has better performance than the other three multi-objective optimization evolution algorithms and two multi-objective harmony search algorithms.The experiment further analyzed the sensitivity of the PSAMOHS algorithm performance to HMCR.Finally,the PSAMOHS algorithm and two other multi-objective harmony search algorithms are used in the multi-objective optimization problem of satellite heat pipe design.The simulation proves the better optimization ability of PSAMOHS algorithm.Considering that different road adhesion factors bring different degrees of difficulty to DDEV stability control,the PSAMOHS algorithm is applied to the DDEV lateral stability control system.According to the size of the road surface adhesion coefficient,this paper classifies the working conditions of DDEV into various types.A corresponding controller is established for each type of working condition,and the BP neural network is used to complete the optimization of the controller parameters.This paper focuses on the control precision of the controller,combined with the global search ability of PSAMOHS algorithm,the PSAMOHS algorithm is used to train the parameters of BP neural network offline.The PSAMOHS algorithm comprehensively considers the trajectory followability and lateral stability of the vehicle.The total error of the yaw rate and the centroid side yaw during the running of the vehicle are two optimization objectives.The parameters of the BP neural network are trained offline to establish the BP neural network.By adding the off-line optimization link of the PSAMOHS algorithm,the BP neural network can be prevented from fall ing into local optimum when the controller is working online.Finally,the simulation outputs a good tracking curve of the yaw rate and the centroid side angle.