康嘉,谷波,韩华
KANG Jia,GU Bo,HAN Hua(Shanghai Jiaotong University

• 1:上海交通大学

摘要(Abstract)：

研究了多变量统计分析方法在制冷装置故障检测和诊断中的应用。对ASHRAE资助下的一组实验数据进行预处理,对其进行故障检测和故障诊断。对于故障检测,可利用平方预测误差(Q统计量)等统计控制变量来判断系统是否在正常运行状态。而对于故障诊断,第一次尝试采用各变量对于平方预测误差的负荷结合各变量的变化率来得到其对于平方预测误差异变的贡献率。从而快速利用变量的变化方向和程度判断故障类型。从结果分析,这种方法可以在众多变量中过滤掉不显著的变化,迅速找到故障主因。利用多变量统计分析方法可以实现对制冷装置的运行状态进行实时监控和诊断。
According to the characteristics of gradual faults in refrigeration systems,a fault detection model based on Multivariate Statistical Analysis was presented.An experimental study was introduced,which was conducted on a 90-ton centrifugal chiller to produce a database that will be used in the development and evaluation of Fault Detection and Diagnostic(FDD) methods applied to chillers.The presented FDD model was validated and evaluated by the experiments.The results show that The PCA-FDD model can meet the detection performance requirement.Moreover,the model was applied to determine the type of faults for the first time,which achieved satisfactory results.

关键词(KeyWords)： 故障检测和诊断;多变量统计分析方法;主元分析;冷水机组
fault detection and diagnostic(FDD);multivariate statistical analysis;principal component analysis(PCA);chiller

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金(50876059)

作者(Author): 康嘉,谷波,韩华
KANG Jia,GU Bo,HAN Hua(Shanghai Jiaotong University

参考文献(References)：

• [1]胡昌华,许化龙.控制系统故障诊断与容错控制的分析和设计[M].北京:国防工业出版社,2000.7.
• [2]Patton R J,Frank P M,Clark R N.Fault Diagnosis inDynamic Systems:Theory and Applications[M].Pren-tice Hall,1989.
• [3]Frank P M.Fault diagnosis in dynamic systems using an-alytical and knowledge-based redundancy-a survey andsome newresults[J].Automatica.1990,26:459-474.
• [4]Isermann,P.Process fault detection based on modelingand estimation methods-a survey[J].Automatica.1984,20:387-404.
• [5]Zhang J,Roberts P D,Ellis J E.Fault diagnosis of amixing process based on qualitative representation ofphysical behaviours.Journal of Intelligent and RoboticSystems[J].1990,3(2):103-115.
• [6]Zhang J,Roberts P D,Ellis J E.A self-learning faultdiagnosis system[J].Transactions of the Institute ofMeasurement and Control.1991,13(1):29-35.
• [7]Zhang J,Roberts P D.Process fault diagnosis with di-agnostic rules based structural decomposition[J].Journal of Process Control.1991,1:259-269.
• [8]周东华,叶银忠.现代故障诊断与容错控制[M].北京:清华大学出版社,2000.6.
• [9]张杰,阳宪惠.多变量统计过程控制[M].北京:化学工业出版社,2000.8.
• [10]王惠文,吴载斌,孟洁.偏最小二乘回归的线性与非线性方法[M].北京:国防工业出版社,2006.9.
• [11]梅长林,范金城.数据分析方法[M].北京:高等教育出版社,2006.2.
• [12]Comstock M C,Braun J E.Development of analysistools for the evaluation of fault detection and diagnos-tics for chillers[D].1999a,HL 99-20,Report#4036-3.ASHRAE Research Project 1043.
• [13]任能.制冷系统故障检测、诊断及预测研究[D].上海:上海交通大学机械与动力工程学院,2008.
• [14]王志毅,谷波,欧永梅.通过实验选择空调制冷系统故障诊断检测参数[J].制冷与空调,2003,3(2):14-18.
• [15]Han H,Cao Z K,Gu B,et al,2010a.PCA-SVM-Based Automated Fault Detection and Diagnosis(AFDD)for Vapor-Compression Refrigeration Sys-tems[J].HVAC&R Research.16(3):295-313.