基于小波去噪的冷水机组传感器故障检测
|
摘要
对基于主元分析方法的冷水机组传感器故障检测效率取决于训练数据和被测数据的质量的问题进行了研究.采用小波变换剔除测量数据中的噪声,提高数据质量,从而提高故障检测效率.结果表明:在-1.0℃故障下,基于小波去噪的主元分析方法的故障检测效率达到了91%.在同等数值的正负偏差故障下,基于小波去噪的主元分析方法的故障检测效率对称性更好.故障检测效率与小波基函数的分解层次关系密切.分解层次越多,故障检测效率越高.所有的db族小波基函数在5层分解的情况下,-0.5℃故障下的检测效率均能达到90%以上.
Chiller sensor fault detection based on principal component analysis is a data-based analysis method.The fault detection efficiency relies on the quality of the training data and the measured data.The measurement noise was removed by wavelet transfer.The fault detection efficiencies were promoted because of the promotion of the data quality.Results show that the fault detection efficiency is 91% on the-1.0 ℃ introduced fault level by the PCA-based method combined with wavelet de-noising.On the same values of the positive and negative fault levels,the symmetry of the presented method is well than the normal PCA-based method.The fault detection efficiencies rely on the decomposed layer of the wavelet transfer.The more decomposed layers are,the well the fault detection efficiencies are.On the-0.5 ℃ fault level,the fault detection efficiencies of all the db′s wavelet function on the 5 layers decomposition are greater than 90%.
Chiller sensor fault detection based on principal component analysis is a data-based analysis method.The fault detection efficiency relies on the quality of the training data and the measured data.The measurement noise was removed by wavelet transfer.The fault detection efficiencies were promoted because of the promotion of the data quality.Results show that the fault detection efficiency is 91% on the-1.0 ℃ introduced fault level by the PCA-based method combined with wavelet de-noising.On the same values of the positive and negative fault levels,the symmetry of the presented method is well than the normal PCA-based method.The fault detection efficiencies rely on the decomposed layer of the wavelet transfer.The more decomposed layers are,the well the fault detection efficiencies are.On the-0.5 ℃ fault level,the fault detection efficiencies of all the db′s wavelet function on the 5 layers decomposition are greater than 90%.
引文
[1]Katipamula S,Brambley M R.Methods for fault de-tection,diagnostics,and prognostics for building sys-tems-a review,part I[J].HVAC&R Research,2005,11(1):3-25.
[2]Wang S W,Cui J T.Sensor-fault detection,diagno-sis and estimation for centrifugal chiller systems usingprincipal-component analysis method[J].AppliedEnergy,2005,82(3):197-213.
[3]Hu Y P,Chen H X,Xie J L,et al.Chiller sensorfault detection using a self-adaptive principal compo-nent analysis method[J].Energy and Buildings,2012,54:252-258.
[4]Chen Y M,Lan L L.A fault detection technique forair-source heat pump water chiller/heaters[J].Ener-gy and Buildings,2009,41(8):881-887.
[5]Chen Y M,Lan L L.Fault detection,diagnosis anddata recovery for a real building heating/cooling bill-ing system[J].Energy Conversion and Management,2010,51(5):1015-1024.
[6]肖赋,王盛卫,徐新华,等.基于主成分分析法的空调系统传感器自动故障诊断[J].建筑科学,2008,24(6):34-39.
[7]Wang S W,Xiao F.AHU sensor fault diagnosis u-sing principal component analysis method[J].Energyand Buildings,2004,36(2):147-160.
[8]Du Z M,Jin X Q,Wu L Z.PCA-FDA-based fault di-agnosis for sensors in VAV systems[J].Hvac&RResearch,2007,13(2):349-367.
[9]Wang S W,Chen Y M.Sensor validation and recon-struction for building central chilling systems basedon principal component analysis[J].Energy Conver-sion and Management,2004,45(5):673-695.
[10]徐新华,杨春生,王盛卫.基于小波变换的传感器故障诊断研究[J].建筑科学,2007,23(12):72-75.
[11]Xu X H,Xiao F,Wang S W.Enhanced chiller sen-sor fault detection,diagnosis and estimation usingwavelet analysis and principal component analysismethods[J].Applied Thermal Engineering,2008,28(2-3):226-237.
[12]郝小礼,陈友明,张国强.小波滤波在小故障检测中的应用[J].暖通空调,2005,35(8):138-140.
[13]张华,陈小宏,杨海燕.地震信号去噪的最优小波基选取方法[J].石油地球物理勘探,2011,46(1):70-75.
[14]王希武,董光波,谢桂海.基于小波变换的核磁共振FID信号的去噪方法研究[J].核电子学与探测技术,2008,28(2):365-370.
[15]张恒,李安宗,李传伟,等.基于离散平稳小波变换的无线随钻系统测试信号处理[J].石油钻探技术,2007,35(2):49-51.
[2]Wang S W,Cui J T.Sensor-fault detection,diagno-sis and estimation for centrifugal chiller systems usingprincipal-component analysis method[J].AppliedEnergy,2005,82(3):197-213.
[3]Hu Y P,Chen H X,Xie J L,et al.Chiller sensorfault detection using a self-adaptive principal compo-nent analysis method[J].Energy and Buildings,2012,54:252-258.
[4]Chen Y M,Lan L L.A fault detection technique forair-source heat pump water chiller/heaters[J].Ener-gy and Buildings,2009,41(8):881-887.
[5]Chen Y M,Lan L L.Fault detection,diagnosis anddata recovery for a real building heating/cooling bill-ing system[J].Energy Conversion and Management,2010,51(5):1015-1024.
[6]肖赋,王盛卫,徐新华,等.基于主成分分析法的空调系统传感器自动故障诊断[J].建筑科学,2008,24(6):34-39.
[7]Wang S W,Xiao F.AHU sensor fault diagnosis u-sing principal component analysis method[J].Energyand Buildings,2004,36(2):147-160.
[8]Du Z M,Jin X Q,Wu L Z.PCA-FDA-based fault di-agnosis for sensors in VAV systems[J].Hvac&RResearch,2007,13(2):349-367.
[9]Wang S W,Chen Y M.Sensor validation and recon-struction for building central chilling systems basedon principal component analysis[J].Energy Conver-sion and Management,2004,45(5):673-695.
[10]徐新华,杨春生,王盛卫.基于小波变换的传感器故障诊断研究[J].建筑科学,2007,23(12):72-75.
[11]Xu X H,Xiao F,Wang S W.Enhanced chiller sen-sor fault detection,diagnosis and estimation usingwavelet analysis and principal component analysismethods[J].Applied Thermal Engineering,2008,28(2-3):226-237.
[12]郝小礼,陈友明,张国强.小波滤波在小故障检测中的应用[J].暖通空调,2005,35(8):138-140.
[13]张华,陈小宏,杨海燕.地震信号去噪的最优小波基选取方法[J].石油地球物理勘探,2011,46(1):70-75.
[14]王希武,董光波,谢桂海.基于小波变换的核磁共振FID信号的去噪方法研究[J].核电子学与探测技术,2008,28(2):365-370.
[15]张恒,李安宗,李传伟,等.基于离散平稳小波变换的无线随钻系统测试信号处理[J].石油钻探技术,2007,35(2):49-51.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心