改进的二进制粒子群算法的传感器优化配置  

The Study on Determination of the Optimal Sensor Placement Based on Improved Binary Particle Swarm Optimization

作　　者：郝晋峰[1] 康建设[1] 史宪铭[1] 张贝贝[1] 

机构地区：[1]军械工程学院,石家庄050003

出　　处：《火力与指挥控制》2013年第8期65-68,共4页Fire Control & Command Control

摘　　要：传感器组合优化问题是设备状态检测系统设计的重要问题。在研究传感器与故障有向图的基础上,提出了传感器优化配置模型;根据传感器组合优化自身特点,从位改变率、惯性权重两个方面对BPSO算法进行了参数分析。结合免疫算法的"亲和度"思想以及非线性惯性权重递减公式,提出了改进的二进制粒子群算法。案例显示,改进的二进制PSO算法提高了算法在整个解空间的搜索能力,加快了收敛速度,能够很好地用于解决传感器优化问题。The Optimal Sensor Placement is the important parts of the designing of the equipment condition detected system.This paper proposes the new Optimal Sensor Placement model based on the studying the digraph of sensor and fault.Based on the determination of the Optimal Sensor Placement self-characteristic,the parameters of Binary Particle Swarm Optimization has been analyzed through bit change rate and inertia weight which shows the effect of previous velocity vector on the new vector,connecting with the affinity of immune algorithm and nonlinear-digress function of inertia weight.The example has been shown that enhanced the computation speed and the ability searching in the entire solution space and solved the determination of the Optimal Sensor Placement commendably.

关 键 词：传感器优化配置 有向图 改进的二进制粒子群算法 故障诊断 

分 类 号：TN212[电子电信—物理电子学]