基于偏最小二乘法的冷水机组传感器故障检测  

Sensor Fault Detection of Chiller Based on Partial Least Square Method

作　　者：吴棒 胡云鹏[2] 陈乖乖 周传辉[1] 李冠男 WU Bang;HU Yunpeng;CHEN Guaiguai;ZHOU Chuanhui;LI Guannan(Institute of HVAC and New Energy,Wuhan University of Science and Technology,Wuhan 430065,Hubei,China;Department of Building Environment and Energy Engineering,Wuhan Business University,Wuhan 430056,Hubei,China)

机构地区：[1]武汉科技大学暖通空调与新能源研究所,湖北武汉430065 [2]武汉商学院机电工程学院,湖北武汉430056

出　　处：《制冷技术》2023年第6期20-26,共7页Chinese Journal of Refrigeration Technology

基　　金：湖北省高等学校优秀中青年科技创新团队计划(No.T201829)。

摘　　要：本文以偏最小二乘(PLS)为主线开展了冷水机组传感器故障检测研究。采用正常的冷水机组运行数据作为训练集建立单输出PLS模型,用于冷水机组传感器故障检测。在测试集中引入不同程度的故障,使用故障检测率(FDR)分析了算法的故障检测性能。结果表明:基于PLS的冷水机组传感器故障检测方法有效,但对不同传感器,故障检测性能存在差异,如冷冻水供水温度传感器故障幅值绝对值大于1.5℃时,故障检测率达到80%,而冷冻水回水温度传感器故障幅值绝对值大于2℃时,故障检测率才能达到80%。Partial least squares(PLS)are used as the main line to carry out the research on the sensor fault detection of the chiller in this paper.The normal chiller operating data is used as the training set to establish a single-output PLS model,which is used for the sensor fault detection of the chiller.Introduce different levels of faults in the test set,and use the fault detection rate to analyze the fault detection performance of the proposed method.The results show that:the PLS-based sensor fault detection method for chillers is effective,but the fault detection performance is different for different sensors.For example,when the absolute value of the chilled water supply temperature sensor's fault amplitude is greater than 1.5℃,the fault detection rate reaches 80%.When the absolute value of chilled water returns temperature sensor's fault amplitude is greater than 2℃,fault detection rate can reach 80%.

关 键 词：冷水机组 传感器故障 偏最小二乘 故障检测 检测性能 

分 类 号：TQ051.5[化学工程] TP306.3[自动化与计算机技术—计算机系统结构]