基于混沌粒子群优化的系统级故障诊断策略优化

Optimizing strategy of system level fault diagnosis based on chaos particle swarm optimization

针对诊断设计优化过程中的关键问题——故障诊断策略优化,提出了基于混沌粒子群优化算法的系统级故障诊断策略优化方法。该算法利用混沌优化不重复遍历系统所有状态的特点,引导粒子在全局范围内搜索,从而克服了粒子群算法"早熟"收敛的缺点。这使算法不仅具有较快的收敛速度,又保证了获得的最优解的可靠性,为获得有效的系统级故障诊断策略提供了可行的方法。最后,给出了该算法在诊断策略优化过程中的关键步骤,通过仿真证明了该算法对于系统级故障诊断策略优化的有效性。

Aiming at the key problem, the optimization of fault diagnostic strategy,in the process of diagnostic design, an optimization method of system-level fault diagnosis strategy based on chaos particle swarm optimization （CPSO） is put forward. This method makes use of the character of chaos optimization which can traverse all the system states non-repetitiously guide the particle searching in the global scale. So it can overcome the PSO^s defect of ＂precocity＂. This improved algorithm not only has a faster convergence rate, but also can ensure the reliability of optimum solution. It can provide a feasible method for optimizing system-level fault diagnosis strategy. Finally, the key steps in the process of diagnosis strategy effectiveness of the proposed method is verified by a simulation experiment.