The Research of Warning Model of Hidden Failure Based on Data Mining

Abstract:

Complex equipment is mainly used in important areas of national defense, health care, banking, etc. Consequences of failure are relatively severe, while the hidden failures are contained in the most complex devices as the process is running. Hidden failures in the normal operation of the device is difficult to find, and only under certain conditions will be triggered, while other faults may be led. The stability of the running system will be undermined. In order to monitor the occurrence and development of hidden failure of complex equipment, a hidden failure warning model based on data mining has been put forward, and the theory of the model has been analyzed, the selection gist of the model parameters has been given. The result shows that the accuracy of hidden failure impact value forecast by the model is 93.33%, the impact degree of the hidden failure effect on the dominant failure can be effectively monitored, and the model makes a good preventative effect against the sudden failure caused by the hidden failure.

• Periodical: Key Engineering Materials (Volume 693)
• Pages: 1844-1848
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.693.1844
• Online since: May 2016
• Authors: Jing Liu, Qing Xiang Zhu*, Xin Yu, Jing Xin Wang, Yi Ge Huang
• Keywords: Data Mining, Fault Diagnosis, Hidden Failure, Neural Network

References

[1] Barakat M, Druaux F, Lefebvre D, et al. Self-Adaptive Growing Neural Network Classifier for Faults Detection and Diagnosis[J]. Neurocomputing, 2011, 78(18): 3865-3876.

DOI: 10.1016/j.neucom.2011.08.001

[2] Purarjomandlangrudi A, Ghapanchi A H, Esmalifalak M. A Data Mining Approach for Fault Diagnosis: An Application of Anomaly Detection Algorithm[J]. Measurement, 2014, 2241(14): 250-253.

DOI: 10.1016/j.measurement.2014.05.029

[3] Lv Yunrong, Chen Xuedong, Gao Jinji, Fan Zhichao, Yang Qichao. China's large polytechnic compressor failure research and Failure Prevention Measures [J]. Fluid Machinery, 2013, 01: 14-20.

[4] Ma Jin, Jiang Zhinong, Gao Jinji. Based on the track of the location of rod axis reciprocating compressor fault diagnosis technology [J]. Vibration Engineering, 2012, 04: 453-459.

[5] Ma Jin, Jiang Zhinong, Gao Jinji. Extraction of feature based on chaos fractal theory apply in valve fault diagnosis [J] Vibration and Shock, 2012, 19: 26-30.

[6] Jing Luyang, Wang Taiyong, Chen Dongxiang, Zhang Pan. Design and implementation of multi-parameter monitoring CNC machine tools diagnosis system [J] Manufacturing Automation, 2013, 11: 178-180.

[7] Hao Yan, Wang Taiyong, Wan Jian, Zhang Pan, Liu Lu. Based on empirical mode decomposition and generalized dimension of mechanical fault diagnosis [J]. Jilin University (Engineering Science), 2012, 02: 392-396.

[8] Hao Yan, Wang Taiyong, Wan Jian, Zhang Pan. Based on cascaded bistable stochastic resonance and multifractal method of mechanical fault diagnosis research [J]. Vibration and Shock, 2012 preceding: 181-185.