基于免疫混沌PSO的液体发动机知识获取方法

为解决液体发动机故障诊断中知识获取的瓶颈问题,提出了一种基于免疫混沌改进粒子群优化算法的知识获取方法.该方法首先将描述系统性能的非线性方程组,转化为带约束的最优化形式,通过设置式中参数可以仿真不同状况下的系统性能.由于此方程为多维、非线性的优化方程,运用标准粒子群优化算法求解进,容易陷入局部最优,为此提出了一种基于免疫混沌粒子群优化算法.该算法在初始化时,运用混沌的思想,使得粒子分布遍历所有状态;在进化时,运用的免疫的思想,设置了基于适应度的克隆算子和基于适应度的混沌变异算子增加粒子的多样性,增加了粒子寻找全局最优的能力.实验结果表明,运用改进的粒子群优化算法进行仿真,能够获取系统知识.

Knowledge acquisition of liquid propellant engine based on immune chaos particle swarm optimization

In order to solve the problem of knowledge acquisition in liquid propellant engine fault diagnosis,a method based on immune chaos particle swarm optimization was proposed.Firstly the nonlinear equations describing the system performance was transformed into optimization form with constriction.By setting different parameters,system performance under different conditions was simulated.Since the equation is nonlinear and multidimensional,standard particle swarm can't solve the problem due to the weakness of pr...