基于主成分法的空调新风阀故障检测与诊断

Fault Detection and Diagnosis Technique of Outdoor Air Valve in Air Conditioning System Based on Principal Component Analysis

建立了主成分分析法的空调系统故障检测模型,根据平方预测误差和其阈值大小的比较,判断系统是否发生故障。采用ASHRAE 1312现场实测数据进行方法的测试验证,以新风阀故障的诊断为例,比较并分析了故障贡献图和相似系数两种故障诊断方法。结果表明:当新风阀开度偏离正常运行开度15%时,主成分分析能够有效检测出故障;故障贡献图不需要先验知识便可检测出故障变量,但只能确定出故障设备引起的单一故障变量;与贡献图法相比,相似系数法诊断结果更加可靠,两种故障诊断方法各有优缺点,两者组合能够更加准确地诊断出故障。

A fault detection model of air conditioning system based on principal component analysis was established, the fault is judged according to the comparison of the square prediction error and the threshold value. ASHRAE 1312 field measured data were used to test and verify the method. Taking fault diagnosis of fresh air valve as an example, two fault diagnosis methods, fault contribution plot and similarity factor, were compared and analyzed. The results show that the principal component analysis can effectively detect the fault if the fresh air valve stuck severity exceeds 15%. The fault contribution plot can detect the fault variable without prior knowledge, but only single fault variable caused by fault equipment can be determined. Compared with the fault contribution plot, the diagnosis results of the similarity factor method is more reliable, and the two fault diagnosis methods have their own advantages and disadvantages, the combination of them can diagnose faults more accurately.