Fault diagnosis for refrigeration system based on PCA-PNN

doi:10.11949/j.issn.0438-1157.20151301

Abstract

Abstract:

The diversity of internal physical form of refrigeration system and the deep coupling between the system parameters make the system more intricate and the detection and diagnosis more complicated. Seven typical degrading faults of a refrigeration system, including system-level and component-level, were explored. The principal component analysis (PCA) was applied to extract the principal characters and reduce the dimension of faults samples. The probabilistic neural network (PNN) was used for fault diagnosis. The PCA could decompose the original 62 parameters into independent principal components and select a certain amount of principal components according to the cumulative contributions. Import these principal components as input data into PNN for fault diagnosis. Results indicate that the PNN combined with PCA is not sensitive to the spread value within a certain range. The combination also increased the correct rate and saved the elapsed time of diagnosis. Obviously, the use of PCA could effectively optimize the diagnosis performance of PNN.

Key words: principal component analysis, probabilistic neural network, refrigeration system, fault diagnosis, optimization

摘要：

制冷系统由于内部物质形态的多样性以及系统参数间的高度耦合而较为复杂,也增加了出现故障后的检测及诊断难度。针对制冷系统常见的7种故障,包括局部故障与系统故障,运用主元分析法提取故障样本主要特征,对样本进行降维处理后,基于概率神经网络进行故障诊断。主元分析法可将原始的62个参数分解为相互独立的主元,根据累计贡献率选取一定量的主元,并将其样本输入概率神经网络进行故障诊断,结果表明结合主元分析后的概率神经网络在一定范围内对spread值不敏感,不仅诊断正确率有所提高,而且缩短了诊断耗时。可见,主元分析法的使用可有效优化概率神经网络的诊断性能。

关键词: 主元分析, 概率神经网络, 制冷系统, 故障诊断, 优化

CLC Number:

TB65

LIANG Qingqing, HAN Hua, CUI Xiaoyu, GU Bo. Fault diagnosis for refrigeration system based on PCA-PNN[J]. CIESC Journal, 2016, 67(3): 1022-1031.

梁晴晴, 韩华, 崔晓钰, 谷波. 基于主元分析-概率神经网络的制冷系统故障诊断[J]. 化工学报, 2016, 67(3): 1022-1031.

References 25

[1]	贺丁, 赵劲松. 基于Hopfield网络的时滞分析故障诊断策略[J]. 化工学报, 2013, 64 (2): 633-640. DOI: 10.3969/j.issn.0438-1157.201302030. HE D, ZHAO J S. Fault strategy of time delay analysis based on Hopfield network[J]. CIESC Journal, 2013, 64 (2): 633-640. DOI: 10.3969/j.issn.0438-1157.201302030.
[2]	胡耀斌, 谢静, 胡良斌. 基于神经网络与小波变换的滚动轴承故障诊断[J]. 机械设计与研究, 2013, 29 (6): 33-35. HU Y B, XIE J, HU L B. Fault diagnosis of antifriction bearings based on the neural network and wavelet transform[J]. Machine Design and Research, 2013, 29 (6): 33-35.
[3]	王瑞海, 孙鹏, 吕振江, 等. 概率神经网络在水泵故障诊断中的应用研究[J]. 煤矿机械, 2014, 35 (10): 285-287. DOI: 10.13436/j.mkjx.201410125. WANG R H, SUN P, LÜ Z J, et al. Research on the application of probabilistic neural network for the fault diagnosis in pump[J]. Coal Mine Machinery, 2014, 35 (10): 285-287. DOI: 10.13436/j.mkjx. 201410125.
[4]	MA D Y, LIANG Y C, ZHAO X S, et al. Multi-BP expert system for fault diagnosis of power system[J]. Engineering Applications of Artificial Intelligence, 2013, 26 (3): 937-944. DOI: 10.1016/j.engappai. 2012. 03. 017.
[5]	HABERL J S, CLARIDGE D E. An expert system for building energy consumption analysis: prototype results[C]//Proceedings of the ASHRAE Conferences. New York: ASHRAE Transactions, 1987.
[6]	刘相艳, 谷波, 黎远光. 基于并行感知器的制冷系统故障诊断分析[J]. 上海交通大学学报, 2005, 39 (8): 1233-1239. LIU X Y, GU B, LI Y G. Analysis of fault diagnosis for refrigeration system based on parallel perceptron[J]. Journal of Shanghai Jiaotong University, 2005, 39 (8): 1233-1239.
[7]	马炎坤. 组态软件在冷水机组实时监视与故障诊断方面的应用[J]. 低温与超导, 2008, 36 (10): 77-81. MA Y K. Application of configuration software in the real-time monitoring and fault diagnosis for chillers[J].Cryogenics and Superconductivity, 2008, 36 (10): 77-81.
[8]	LI H R, BRAUN J E. Decoupling features and virtual sensors for diagnosis of faults in vapor compression air conditioners[J]. International Journal of Refrigeration, 2007, 30 (3): 546-564. DOI: 10.1016/j.ijrefrig. 2006. 07.024.
[9]	NAJAFI M, AUSLANDER D M, BARTLETT P L, et al. Application of machine learning in the fault diagnostics of air handling units[J]. Applied Energy, 2012, 96: 347-358. DOI:10.1016/j.apenergy.2012.02.049.
[10]	LI S, WEN J. Application of pattern matching method for detecting faults in air handling unit system[J]. Automation in Construction, 2014, 43: 49-58. DOI:10.1016/j.autcon.2014.03.002.
[11]	王伟, 姚杨, 马最良. 基于BP神经网络的压缩机性能预测模型的建立[J]. 流体机械, 2005, 33 (9):21-24. WANG W, YAO Y, MA Z L. Establishment of the performance forecast model for compressor based on BP neural network[J]. Fluid Machinery, 2005, 33 (9): 21-24.
[12]	YILMAZ S, ATIK K. Modeling of a mechanical cooling system with variable cooling capacity by using artificial neural network[J]. Applied Thermal Engineering, 2007, 27 (13): 2308-2313. DOI: 10.1016/j.applthermaleng. 2007.01.030.
[13]	KAMAR H M, AHMAD R, KAMSAH N B, et al. Artificial neural networks for automotive air-conditioning systems performance prediction[J]. Applied Thermal Engineering, 2013, 50 (1): 63-70. DOI: 10.1016/j.applthermaleng. 2012.05.032.
[14]	ERTUNC H M, HOSOZ M. Artificial neural network analysis of a refrigeration system with an evaporative condenser[J]. Applied Thermal Engineering, 2006, 26 (5/6): 627-635. DOI: 10.1016/j.applthermaleng. 2005. 06.002.
[15]	李中领, 金宁, 朱岩. 人工神经网络应用于空调系统故障诊断的研究[J]. 制冷与空调, 2005, 5 (1): 50-53. LI Z L, JIN N, ZHU Y. Research on artificial neural network for fault diagnosis of air conditioning system[J]. Refrigeration & Air-Conditioning, 2005, 5 (1): 50-53.
[16]	KOCYIGIT N. Fault and sensor error diagnostic strategies for a vapor compression refrigeration system by using fuzzy inference systems and artificial neural network[J]. International Journal of Refrigeration, 2014, 50: 69-79.
[17]	WU J D, CHIANGA P H, CHANGB Y W, et al. An expert system for fault diagnosis in internal combustion engines using probability neural network[J]. Expert Systems with Applications, 2008, 34 (4): 2704-2713. DOI: 10.1016/j.eswa.2007.05.010.
[18]	ZHAO J, WANG X Y, JIN P Q. Feature selection for event discovery in social media: a comparative study[J]. Computers in Human Behavior, 2015, 51: 903-909. DOI: 10.1016/j.chb.2014.11.007.
[19]	韩华, 谷波, 任能. 基于主元分析与支持向量机的制冷系统故障诊断方法[J]. 上海交通大学学报, 2011,45 (9): 1355-1361. HAN H, GU B, REN N. Fault diagnosis for refrigeration system based on principal component analysis and support vector machine[J]. Journal of Shanghai Jiaotong University, 2011, 45 (9): 1355-1361.
[20]	杨淑莹. 模式识别与智能计算MATLAB技术实现[M]. 2版. 北京:电子工业出版社, 2011: 120. YANG S Y. Pattern Recognition and Intelligent Computing— MATLAB Implementation[M]. 2nd ed. Beijing: Electronic Industry Press, 2011:120.
[21]	陈明. MATLAB神经网络原理与实例精解[M]. 北京: 清华大学出版社, 2013: 58. CHEN M. MATLAB Neural Network Principle and Instance Analysis[M]. Beijing: Tsinghua University Press, 2013: 58.
[22]	KARTHIKEYAN B, GOPAL S, VIMALA M. Conception of complex probabilistic neural network system for classification of partial discharge patterns using multifarious inputs[J]. Expert Systems with Applications, 2005, 29 (4): 953-963. DOI:10.1016/j.eswa.2005.06.014.
[23]	COMSTOCK M C, BRAUN J E. Development of analysis tools for the evaluation of fault detection and diagnostics for chillers[R]. ASHRAE, 1999.
[24]	LI H R. A decoupling-based FDD approach for multiple simultaneous faults[D].West Lafayette: Purdue University, 2003.
[25]	HAN H, GU B, WANG T, et al. Important sensors for chiller fault detection and diagnosis (FDD) from the perspective of feature selection and machine learning[J]. International Journal of Refrigeration, 2011, 34: 586-599.