会议专题

Analysis of Low Voltage Electric Appliances System Reliability Optimization Based on Improved PSO

Optimization of the reliability of the low voltage electric appliances system is an NP hard problem, in the methods that has been used to solve the problem, no one can search the precise solution, in this paper ,an advanced artificial intelligent method is adopted. The traditional intelligent methods, such as genetic algorithm (GA), simulated annealing algorithm (SA), and so on, had some flaws including early convergence and easily falling in local peak, etc. The improved Particle Swarm Optimization with adaptive inertia weight is proposed to solve the problem of reliability optimization. The key point of the improved algorithm is that it is able to adjust the inertia weight according to the searching result in the algorithm iteration. Thus it can get faster convergence speed and better global search capability .The concrete strategy to solve the reliability optimization by the improved PSO is proposed, according to the mathematical model, the program steps of the solution is listed. The results of the simulation test shows that the algorithm can fulfill the requirement of the convergence speed and the convergence precision, and it has high searching efficiency .Comparing with other algorithms, the improved PSO convergence faster and it has higher precision, so the improved PSO is feasible and efficiency to solve the low voltage electric appliances system reliability optimization problem.

Reliability Optimization Particle Swarm Optimization Inertia Weight

Guo Zhitao Wang Yao Yuan Jinli Dong Yongfeng Gu Junhua

School of Information Engineering of Hebei University of Technology, Tianjin, 300130 School of Electrical engineering of Hebei University of Technology, Tianjin, 300130 School of Computer Science and Engineering of Hebei University of Technology, Tianjin, 300130

国际会议

第三届电工产品可靠性与电接触国际会议(The 3rd International Conference on Reliability of Electrical Products and Electrical Contacts)

温州

英文

282-285

2009-10-18(万方平台首次上网日期,不代表论文的发表时间)