Research on Vibration Signal Feature Extraction Method to the Wind Turbine Generator

Abstract:

Aiming at the safe operation of the wind turbine, a feature extraction method of vibration signal based on the principle of blind source separation was proposed. Blind source and the current state of fault signal was separated and predicated by Using historical data of wind turbine vibration signal as the observation noise, and then fault diagnosis signal mechanical operation was analyzed, the HMM/SVM series fault diagnosis models structure was proposed. By calculating the matching degree of unknown signal and wind power equipment in the state using HMM, the features for SVM was formed to achieve the finally discriminant. The experimental results showed that the fault diagnosis method can precisely and effectively complete the wind power equipment, higher than pure HMM or SVM diagnostic accuracy.

• Periodical: Advanced Materials Research (Volume 902)
• Pages: 370-377
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.902.370
• Online since: February 2014
• Authors: Guo Xin Wu, Yun Bo Zuo, Yan Hui Shi
• Keywords: Blind Source Separation, HMM/SVM Series, Vibration Signal Analysis, Wind Turbine (WT)

References

[1] Wang Hongjun, Zhang Jianmin, Xu Xiaoli. Study On Combination Trend Prediction Technology for Mechaninery System Based on SVM . Journal of vibration engineering，2006(6)：242-245.

[2] Liu Xinmin, Liu Guanjun, Qiu Jing. Hybrid HMM and SVM Approach for Fault Diagnosis. Chinese journal of scientific instrument, 2006(01): 45-53.

[3] Rabiner L R. An Introduction to The Hidden Markov Models. IEEE ASSP Mag，1986(1): 4-16.

DOI: 10.1109/massp.1986.1165342

[4] B. X Xu, Q. S. Han, X. L. Xu. A Method of Trends Predictions of Machine Running Conditions Based on Hidden Markov Models. Precise manufacturing & automation. 2003(9): 11-13.

[5] Vapnik V. The Nature Of Statistics Learning Theory. New York: Springer-Verlag, 1995: 40-50.

[6] Vapnik V. Statistical Learning Theory. NewYork: JohnWiley and Sons, 1998, 102-110.

[7] C. Cortes,V. Vapnik. Support Vector Networks . Machine Learning, 1995, 20: 273-297.

DOI: 10.1007/bf00994018

[8] Yue Xia, Zhang Chunliang, Quan Yanming, Zhu Houyue. Fault Diagnosis of Nuclear-Powered Equipment Based on HMM&SVM. Nuclear power engineering. 2012(6): 104-107.

[9] Wang Hongjun, Zhang Jianmin, Xu Xiaoli. Machinery Trend Condition Support Vector Machine Intelligent Prediction. Machine tool & hydraulics. 2005(5): 170-172.

[10] Zhang Zhousuo, Li Lingjun, He Zhengjia. Research on Diagnosis Method of Machinery Fault Based on Support Vector Machine. Journal of xi'an jiaotong university. 2002, 36 (6): 116-120.

[11] Morteza Daneshkar, Reza Ebrahimi. A Novel Initialization Approach For Solving Permutation Ambiguity Of Frequency Domain Blind Source Separation. IEEE ISCI 2012, p: 189-192.

DOI: 10.1109/isci.2012.6222692

[12] Ma Jiancang, Shi Qingbin. Blind Source Separation for Nonlinearly Mixed Mechanical Vibration Signals. Noise and vibration control，2008(6): 5-8.

[13] John A, Ismial I. Analysis of Subbanding Technique in Blind Source Separation. Proceedings of SPIE - The International Society for Optical Engineering, v 6236, (2006).

[14] Tang Yan, Tang Jingtian. Removal of Ocular Artifact From EEG Using JADE. ICBBE'2007: 566-569.

[15] Nakayama, Kenji Hirano, Akihiro Horita. A Learning Algorithm for Convolutive Blind Source Separation With Transmission Delay Constraint. Proceedings of the International Joint Conference on Neural Networks, 2002, v2: 1287-1292.

DOI: 10.1109/ijcnn.2002.1007680

[16] TheodorD. Popescu. Blind Separation of Vibration Signals And Source Change DetectionApplication To Machine Monitoring. Applied mathematical modelling , 2010(34): 3408–3421.

DOI: 10.1016/j.apm.2010.02.030