基于LPWAN物联网的PM_(2.5)监测系统  

Monitoring system for PM_(2.5)based on LPWAN internet of things

作　　者：海涛[1] 王秋晨 杨嘉芃 李康 林国忠 HAI Tao;WANG Qiu-chen;YANG Jia-peng;LI Kang;LIN Guo-zhong(College of Electrical Engineering,Guangxi University,Nanning 530004,China;Guangxi Mengchuang Wisdom Technology Co.,LTD.,Nanning 530004,China)

机构地区：[1]广西大学电气工程学院,南宁530004 [2]广西盟创智慧科技有限公司,南宁530004

出　　处：《桂林理工大学学报》2022年第4期1002-1008,共7页Journal of Guilin University of Technology

基　　金：国家自然科学基金项目(51867003);广西重点研发计划项目(桂科AB22035037);广西科技计划项目(桂科AB16380193);科技创新平台(基地)与服务能力建设项目(AE30200067)。

摘　　要：针对传统空气质量监测系统稳定性差、预测精度不足等问题,提出了一种基于麻雀搜索算法优化BP神经网络(SSA-BP)的预测方法,构建了基于低功耗广域物联网LPWAN和云平台技术于一体的PM_(2.5)智能监测系统。系统以物联网云平台为主体框架,LoRa通信传输环境参数,在云端部署服务,使用SSABP网络模型,达到提高系统整体稳定性和预测精度的目的。通过将采集的特征参数用SSA-BP神经网络和BP神经网络进行训练、仿真测试和对比,结果表明SSA-BP模型相较于传统BP网络模型有着更好的预测精度,预测误差率在4%以内,系统实现了对PM_(2.5)的精准监测。To improve poor stability and insufficient prediction accuracy in traditional air quality monitoring system,a prediction method based on sparrow search algorithm optimized BP neural network(SSA-BP)was proposed.An intelligent PM_(2.5)monitoring system based on LPWAN internet of things and cloud platform technology was constructed.In the system,IoT cloud platform is the main framework.LoRa communicates and transmits environmental parameters,provides services in the cloud.The SSA-BP network model is used to improve the overall stability and prediction accuracy of the system.At the same time,the collected characteristic parameters were trained,simulated and compared with the SSA-BP neural network and BP neural network.The results show that the SSA-BP model works with better prediction accuracy than the traditional BP network model.The system achieves accurate monitoring of PM_(2.5)and the prediction error rate is less than 4%.

关 键 词：PM_(2.5)监测 远程监控 云平台 SSA-BP 物联网 低功耗广域网 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP29[自动化与计算机技术—控制科学与工程]