粮情测控系统中传感器布局的改进离散粒子群算法  

Development of distribution for sensors in grain condition monitoring system using improved discrete particle swarm optimization

作　　者：廉飞宇[1,2] 张李建 

机构地区：[1]河南工业大学信息科学与工程学院,河南郑州450001 [2]河南工业大学粮食信息处理与控制教育部重点实验室,河南郑州450001

出　　处：《粮食与饲料工业》2016年第11期9-13,共5页Cereal & Feed Industry

摘　　要：粮情测控系统的检测往往涉及到传感器布局的优化问题。基于改进的检测故障-传感器相关性矩阵,提出一种传感器布局优化模型,该模型在传感器故障自检测能力约束下,基于混沌理论,根据粒子群算法的早期收敛程度自适应调整其惯性权重,并重新定义了粒子的位置更新方式。对于建立的检测故障-传感器优化模型,应用改进的粒子群算法进行仿真实验,实验结果表明,优化结果能够满足粮情测控系统的测试指标,为粮情测控系统中传感器的优化设计提供了新的思路。The detection of grain condition monitoring system usually refers to the optimization problem for sensors distribution. Based on the improved detection fault-sensor relevant matrix, we proposed a sensor distribution optimization model. In this model, under the self-test ability of sensors, we used chaos theory to adjust inertia weight of algorithm adaptively according to early convergence degree of particle swarm algorithm, and redefined location update mode. For the detection of fault-sensor model,the simulation result using the improved particle swarm algorithm showeds that： the optimization results could meet the test index for grain condition monitoring system, and provide a new idea for optimization design of sensors in complex system.

关 键 词：粮情测控系统 布局优化 粒子群 惯性权重 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] S379[自动化与计算机技术—控制科学与工程]