Research on Data-Based Nonlinear Fault Prediction Methods in Multi-Transform Domains for Electromechanical Equipment
Safety of equipment has significant impact on production and human resources as well as environment. Ensuring safe operation of equipment is an important problem faced. One of the important and difficult key technologies in guaranteeing equipment operation is fault prediction. In this paper, research on fault prediction methods based on field data mainly is carried out to achieve predictive maintenance for large rotating electromechanical equipment as most of its faults being trendy ones with long course characteristics. This paper studies new way to make fault prediction in multi-transform domains, perform feature frequency band decomposition based on wavelet packet or HHT, explore nonlinear dimension reduction method to extract fault sensitive characteristics, applies Elman neural network methods to perform nonlinear associative intelligent prediction based on historical and present fault sensitive characteristics so as to realize long course fault prediction. The research is important for large electromechanical equipment to achieve early fault prediction, guarantee safe operation, save maintenance costs, improve utilization, and implement scientific maintenance.
KeywordsWavelet Packet Fault Prediction Wavelet Packet Transform Predictive Maintenance Nonlinear Manifold
Unable to display preview. Download preview PDF.
- 1.Jay Lee,Jun Ni,Dragan Djurdjanovic,Hai Qiu,Haitao Liao.(2006) Intelligent Prognostics Tools and E-Maintenance. Computers in Industry, (57): 476~489Google Scholar
- 2.Andrew K.S.Jardine,Daming Lin,Dragan Banjevic.(2006) A Review On Machinery Diagnostics and Prognostics Implementing Condition-based Maintenance. Mechanical Systems and Signal Processing, (20):1483~1510Google Scholar
- 3.P.Caselitz,J.Giebhardt.Condition Monitoring and Fault Prediction for Marine Current Turbine. http://www.iset.unikassel.de/abt/FB-E/papers/Paper_CM4MCT_dl.pdfGoogle Scholar
- 4.Ammar Iqbal,Rakesh Tanange ,Shafqat Virk. (2006) Vehicle Fault Prediction Analysis. SwedenGoogle Scholar
- 5.Liang Xu, Li Xingshan, Zhang Lei, Yu Jinsong. (2007) Survey of Fault Prediction Supporting Condition Based Maintenance, Measurement & Control Technology, 26(6):5-8, 14Google Scholar
- 6.P.FrankPai,AnthonyN.Palazotto. (2008) HHT-based Nonlinear Signal Processing Method for Parametric and Non-Parametric Identification of Dynamical Systems. International Journal of Mechanical Sciences, (50): 1619~1635Google Scholar
- 8.Hu Qiao,He Zhengjia, Zhang Zhousuo,Zi Yanyang. (2007) Fault Diagnosis of Rotating Machinery Based on Improved Wavelet Package Transform and SVMs Ensemble.Mechanical System and Signal Processing, 21(2):688-705Google Scholar
- 9.TENENBAUM J B, SILVA V, LANGFORD J C. (2000) A global geometric framework for nonlinear dimensionali-ty reduction.Science, (290):2319-2323.Google Scholar
- 10.Yin Junsong, Xiao Jian, Zhou Zongtan, Hu Dewen. (2007) Analysis and Application of Nonlinear Manifold Learning Method, Progress in Natural Science, 17(8):1015-1025.Google Scholar
- 11.Abhinav Saxena,Ashraf Saad. (2007) Evolving an Artificial Neural Network Classifier for Condition Monitoring of Rotating Mechanical System. Applied Soft Computing, (7): 441-454Google Scholar
- 12.J.L. Elman. (1990) Finding structure in time.Cognitive Sci, (14): 179-211Google Scholar
- 13.Meng Lingqi, Meng Meng. (2008) Application of Elman Neural Network on Wide Spread Prediction in Medium Plate Mill, Journal of Jilin University (Engineering and Technology Edition), 38(1):193-196Google Scholar