PSO-SVM算法在智能建筑环境监控系统中的应用  

Application of PSO-SVM Algorithm in Environmental Monitoring System of Intelligent Building

作　　者：傅军栋[1] 邹欢[1] 康水华 

机构地区：[1]华东交通大学电气与电子工程学院,江西南昌330013

出　　处：《华东交通大学学报》2016年第1期121-127,共7页Journal of East China Jiaotong University

摘　　要：针对智能建筑环境监控系统中,多个传感器独立工作可能会造成系统误判的问题,提出了一种基于粒子群优化支持向量机(PSO-SVM)的环境质量综合评价模型,即利用粒子群算法快速优化支持向量机的惩罚参数和核函数参数,然后采用Zig Bee无线传感网络采集的环境数据对PSO-SVM分类模型进行训练和测试。实验结果表明PSO-SVM分类器对环境质量判断的平均精度达到94.44%,且分类结果稳定。将这种方法应用于智能建筑环境监控系统中,可以增加系统监测数据的准确性,提高系统工作的可靠性。Abstract：In environmental monitoring system of intelligent buildings, independent work of multiple sensors may cause misjudgment. Aiming at this problem, this paper proposes an environmental quality comprehensive evalua tion model for optimizing the Support Vector Machine （SVM） by using the Particle Swarm Optimization （PSO） al- gorithm, in which the SVM parameters of the punishment and the kernel function are optimized by PSO and the established SVM classifiers are trained and tested by using the sample data collected by ZigBee wireless sensor networks. The experimental results show that the average recognition rate of the PS0-SVM classifier is up to 94.44% in evaluating environmental quality, and the classification results are stable. It suggests that the pro- posed method increase the accuracy of monitoring data and improve working reliability for the environmental monitoring system of intelligent buildings.

关 键 词：环境监控 多传感器 粒子群优化 支持向量机 

分 类 号：TM715[电气工程—电力系统及自动化]