基于改进的线性判别分析识别重叠并发的双故障

工业工程 ›› 2012, Vol. 15 ›› Issue (5): 73-78.

基于改进的线性判别分析识别重叠并发的双故障

李天恩（1984-），男，河北省人，博士研究生，主要研究方向为质量工程，质量诊断.

(天津大学管理与经济学部，天津 300072)

出版日期:2012-10-31 发布日期:2012-11-15
基金资助:
国家自然科学基金资助项目（70931004）

Detection of Overlapping and Simultaneous Two Faults Based on the Improved LDA

(School of Management, Tianjin University, Tianjin 300072, China)

Online:2012-10-31 Published:2012-11-15

摘要/Abstract

摘要： 在两维空间中，当关键质量特性之间存在相关关系并且预定义故障类之间重叠时，传统的模糊聚类算法FCM对双故障并发的识别率会下降。为了提升对重叠并发双故障的识别率，一种新算法PILDA被提出，该算法提出的主成分修整能够消除重叠的影响，而双故障判别区间确定的方法则能够实现对未预定义的并发双故障的识别。经过864种不同相关关系和均值偏移量的故障组合仿真实验，结果表明PILDA能有效识别并发故障及预定义单发故障，平均识别率为84.94%,明显高于FCM的58.13%。该方法具有一定的应用价值。

关键词: 线性判别分析, 主成分修整, 判别区间

Abstract: In the two-dimension space, when there is a correlation between two critical-to-quality (CTQ) attributes, two predefined faults overlap each other. In this case, if fuzzy c-means (FCM) method is applied to detect these two faults that happen simultaneously, the detection rate is low. In order to solve this problem, a new algorithm called linear discriminant analysis (LDA) with principal component analysis (PCA) and discriminant intervals (PILDA) is proposed. By this algorithm, PCA-shaping is used to eliminate the effects of overlapping and to determine the discriminant intervals so as to overcome the difficulty in detecting two faults that are not predefined. To verify the proposed method, simulation is made for the detection of simultaneous faults and single faults by using 864 fault combinations in different variable relations and different variable shifts. By the proposed method, the average detection rate is 84.94% compared with 58.13% by FCM. This is a significant improvement.

Key words: linear discriminant analysis (LDA), PCA-shaping, discriminant intervals

李天恩，何桢. 基于改进的线性判别分析识别重叠并发的双故障[J]. 工业工程, 2012, 15(5): 73-78.

Li Tian-en, He Zhen. Detection of Overlapping and Simultaneous Two Faults Based on the Improved LDA[J]. Industrial Engineering Journal , 2012, 15(5): 73-78.

参考文献

［1］Li J H, Cui P L. Kernel scatter-difference-based discriminant analysis for nonlinear fault diagnosis［J］. Chemometrics and Intelligent Laboratory Systems, 2008, 94(1):80-86. ［2］Venkatasubramanian V, Rengaswamy R, Yin K, et al. A review of process fault detection and diagnosis Part I: Quantitative model-based methods［J］. Computers & Chemical Engineering, 2003, 27(3):293-311. ［3］沈道义,庞彦伟,王雷,等.基于遗传算法的线性判别分析方法［J］.数据采集与处理, 2008, 23(3):327-332. Shen Dao-yi, Pang Yan-wei, Wang Lei, et al. Linear discriminant analysis based on genetic algorithm［J］. Journal of Data Acquisiti n & Processing, 2008, 23(3): 327-332. ［4］刘青山,卢汉清,马颂德.综述人脸识别中的子空间方法［J］.自动化学报, 2003, 29(6):900-911. Liu Qin-shan, Lu Han-qin, Ma Song-de. A survey：subspace analysis for face recognition［J］. Acta Automatica Sinica, 2003, 29(6):900-911. ［5］Chiang L H, Russell E L, Braatz R D. Fault diagnosis in chemical processes using Fisher discriminant analysis, discriminant partial least squares, and principal component analysis［J］.Chemometrics and Intelligent Laboratory Systems, 2000, 50(2):243-252. ［6］Fuente M J, Garcia G, Sainz G I. Fault diagnosis in a plant using Fisher discriminant analysis［C］. Proceeding of the 16th Mediterranean Conference on Control and Automation Congress Centre, France:Ajaccio, 2008: 53-58. ［7］Du Z M, Jin X Q, Wu L Z. PCA-FDA-based fault diagnosis for sensors in VAV systems［J］. HVAC & R Research, 2007, 13(2): 349-367. ［8］Jiang L Y, Wang S Q. Fault diagnosis based on independent component analysis and Fisher discriminant analysis［C］. Proceedings of the Third International Conference on Machine Learning and Cybernetics. Shanghai: IEEE, 2004: 26-29. ［9］Zhou Y F, Hahn J, Mannan M S. Process monitoring based on classification tree and discriminant analysis［J］. Reliability Engineering & System Safety, 2006, 91(5): 546-555. ［10］Du Z M, Jin X Q. Multiple faultsdiagnosis for sensors in air handling unit using Fisher discriminant analysis［J］. Energy Conversion and Management, 2008, 49(12): 3654-3665. ［11］林吉良, 蒋静坪. 基于模糊聚类的移动机器人并发故障诊断［J］. 浙江大学学报：工学版, 2010, 44 (3): 453-457. Lin Ji-liang, Jiang Jing-ping. Diagnosis of simultaneous faults for mobile robots based on fuzzy clustering method［J］. Journal of Zhejiang University: Engineering Science, 2010, 44 (3): 453-457. ［12］Zhang K, Chai Y, Yang S X. Self-organizing feature map for cluster analysis in multidisease diagnosis［J］. Expert Systems with Applications, 2010, 37(9): 6359-6367. ［13］Du Z M, Jin X Q. Multiple faults diagnosis for sensors in air handling unit using Fisher discriminant analysis［J］. Energy Conversion and Management, 2008, 49(12):3654-3665. ［14］Ma B, Qu H Y, Wong H S. Kernel clustering-based discriminant analysis［J］. Pattern Recognition, 2007, 40(1):324-327. ［15］Ronald D, Fricker Jr. Directionally sensitive multivariate statistical process control procedures with application to syndromic surveillance［J］.Advances in Disease Surveillance, 2007, 3(1):1-17. ［16］Das N, Prakash V. Interpreting the out-of-control signal in multivariate control chart-a comparative study［J］. The International Journal of Advanced Manufacturing Technology, 2008, 37(9/10): 966-979. ［17］He X B, Yang Y P, Yang Y H. Fault diagnosis based on variable-weighted kernel Fisher discriminant analysis［J］. Chemometrics and Intelligent Laboratory Systems, 2008, 93(1):27-33.