首页|基于主元分析的温室物联网空气温湿度传感器故障诊断

基于主元分析的温室物联网空气温湿度传感器故障诊断

扫码查看
原文链接
  • 万方数据
  • 维普
针对温室物联网测控系统中空气温湿度传感器受恶劣环境影响易发生故障的问题,研究温室物联网测控系统中空气温湿度传感器的故障诊断,保障温室物联网测控系统整体工作的稳定可靠性,提出一种基于主元分析的空气温湿度传感器故障诊断方法.首先对空气温湿度传感器数据进行主元分析,通过监控平方预测误差统计量的变化实现传感器的故障检测;再针对检测出的空气温湿度传感器故障数据,利用加权后的平方预测误差统计量来计算传感器的累积贡献率,并将其作为传感器故障识别的指标,识别出现故障的空气温湿度传感器.利用空气温湿度传感器数据在不同故障条件下进行传感器故障诊断方法验证,验证结果表明:所提方法可用于温室物联网测控系统中空气温湿度传感器偏差故障和漂移故障的检测,偏差故障的检测效果要好于漂移故障,偏差故障综合检测率为100％,漂移故障综合检测率为51.25％;同时所提方法能够正确识别出故障传感器,有效提高温室物联网测控系统中空气温湿度传感器故障诊断结果的准确性.
Fault diagnosis of greenhouse IoT air temperature and humidity sensors based on principal components analysis
Aiming at the problem that the air temperature and humidity sensor in the greenhouse IoT measurement and control system is prone to failure due to the harsh environment,the fault diagnosis of the air temperature and humidity sensor in the greenhouse IoT measurement and control system is studied to ensure the stability and reliability of the overall work of the greenhouse IoT measurement and control system.A fault diagnosis method of air temperature and humidity sensor based on principal component analysis is proposed.Firstly,the principal component analysis of the air temperature and humidity sensor data is carried out,and the fault detection of the sensor is realized by monitoring the change of the square prediction error statistics.Then,for the detected air temperature and humidity sensor fault data,the weighted squared prediction error statistics are used to calculate the cumulative contribution rate of the sensor,and it is used as an indicator of sensor fault identification to identify the faulty air temperature and humidity sensor.The sensor fault diagnosis method is verified by using the data of air temperature and humidity sensor under different fault conditions.The verification results show that the proposed method can be used to detect the deviation fault and drift fault of air temperature and humidity sensor in the measurement and control system of greenhouse IoT.The detection effect of deviation fault is better than that of drift fault.The comprehensive detection rate of deviation fault is 100％,and the comprehensive detection rate of drift fault is 51.25％.At the same time,the proposed method can correctly identify the fault sensor,and effectively improve the accuracy of the fault diagnosis results of the air temperature and humidity sensor in the greenhouse IoT measurement and control system.

greenhouseIoTair temperature and humidity sensorsfault diagnosisprincipal components analysismeasurement and control system

范卫强、柳平增、朱珂、孟宪勇、刘力宁

展开 >

山东农业大学信息科学与工程学院,山东泰安,271018

山东农业大学农业大数据研究中心,山东泰安,271018

农业农村部黄淮海智慧农业技术重点实验室,山东泰安,271018

山东农业大学机械与电子工程学院,山东泰安,271018

展开 >

温室 物联网 空气温湿度传感器 故障诊断 主元分析 测控系统

山东省农业重大应用技术创新项目山东省重大科技创新工程项目山东省科技特派员项目山东省重点研发计划

SD2019ZZ0192019JZZY0107132020KJTPY0782022CXGC010609

2024

10.13733/j.jcam.issn.2095-5553.2024.05.011

中国农机化学报
农业部南京农业机械化研究所

中国农机化学报

CSTPCD北大核心
影响因子：0.684
ISSN：2095-5553
年,卷(期)：2024.45(5)
  • 25