Diagnosis for Engine Misfire Fault Based on Torsional Vibration and Neural-Network Analysis

Abstract:

Misfire is a common fault which affects the engine performances. Because the signal-to-noise ratio of torsional vibration signal is high, torsional vibration test and analysis for the engine were performed in a variety of operating conditions, including healthy condition and single-cylinder misfire condition. In order to improve the accuracy of analysis, energy centrobaric correction method was used to correct the amplitude. Taking the corrected amplitude of main order as the fault feature, and then a BP neural-network diagnostic model can be established for misfire diagnosis. The result shows that the method of combining torsional vibration signal analysis and neural-network can diagnose engine misfire fault correctly.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 7240-7246
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.7240
• Online since: January 2012
• Authors: Can Yi Du, Kang Ding, Zhi Jian Yang, Cui Li Yang
• Keywords: Energy Centrobaric, Engine, Misfire Fault, Neural Network (NN), Torsional Vibration

References

[1] P. Charles, Jyoti K. Sinha, F. Gu, etc. Detecting the crankshaft torsional vibration of diesel engines for combustion related diagnosis[J]. Journal of Sound and Vibration 321(2009) 1171-1185.

DOI: 10.1016/j.jsv.2008.10.024

[2] Teng Qin. Torsional Vibration of the Digital Measurement[J]. Journal of Electrical Measurement and Instrumentation, 1996, 7 (1) : 31-33.

[3] Yang Zhijian. Discrete spectrum correction in theoretical research and application of engine vibration analysis[D]. Guangzhou: South China University of Technology, (2009).

[4] Ding Kang, Jiang Liqi. Energy Centrobaric Correction Method for Discrete Spectrum[J]. Journal of Vibration Engineering, 2001, 14(3) : 354-358.

[5] Ding Kang, Zheng Chunsong, Yang Zhijian. Frequency Estimation Accuracy Analysis and Improvement of Energy Barycenter Correction Method for Discrete Spectrum[J]. Journal of Mechanical Engineering, 2010, 46(5) : 43-48.

DOI: 10.3901/jme.2010.05.043

[6] Xiang Jianhua, Liao Ridong, Zuo Zhengxing. Study on the Torsional Vibration Characteristics of Engine Crankshaft System When Single-Cylinder Misfires[J]. Journal of Beijing University of Technology, 2008, 28 (2) : 121-124.

[7] Wang Huamin, Chen Xia, An Gang, etc. I.C. Engine Misfire Fault Diagnosis Based on Singular Value Theory and Neural Networks [J]. Journal of Internal-combustion Engine, 2003, 21 (6) : 449-452.

[8] Qiao Xinyong, Liu Jianmin, Zhang Xiaoming. A Method for Diagnosing Misfiring Fault of Engine Based on Neural Network Data Fusion [J]. Chinese Internal Combustion Engine Engineering, 2009, 30 (1) : 74-79.

DOI: 10.1109/csse.2008.1409

[9] Ren Hui, Niu Yijie, LiuYanbin. A method for Daignosing the Misfire Fault of Engine Using Vibration Signal based on Wavelet Analysis[J]. Journal of Vibration and Impact, 2006, 25(S) : 24-26. TABLE I. Part of Trainning Samples Unit: ×10-3 deg Order Group No. 0. 5.

[1] 1. 5.

[2] 2. 5.

[3] 3. 5.

[4] Actual State.

[1] [57] 515.

[32] [77] [9] [11] [8] [13] [0] 0.

[2] [42] 511.

[13] 120.

[8] [7] [5] [6] [0] 0.

[3] [60] 613.

[11] 311.

[24] [19] [8] [52] [0] 0.

[4] [84] 574.

[14] 333.

[22] [67] [10] [64] [0] 0.

[5] 345 265.

[10] [50] [5] [6] [4] [3] [0] 0.

[6] 227 197.

[17] [58] [5] [5] [5] [7] [0] 0.

[7] 374 205.

[15] 104.

[9] [8] [7] [5] [0] 0.

[8] 239 181.

[22] 176.

[14] [11] [10] [11] [0] 0.

[9] 270 161.

[15] [66] [7] [4] [6] [1] [0] 0.

[10] 105 156.

[19] [65] [8] [5] [6] [3] [0] 0.

[11] [25] 139.

[21] [50] [9] [5] [9] [2] [0] 0.

[12] 771 660.

[55] 305.

[67] [47] [22] [54] [0] 1.

[13] 932 610.

[71] 237.

[36] [51] [8] [46] [0] 1.

[14] 298 217.

[98] [36] [8] [11] [5] [4] [0] 1.

[15] 332 225 125.

[21] [12] [8] [4] [5] [0] 1.

[16] 528 205 204.

[96] [22] [12] [12] [5] [0] 1.

[17] 543 219.

[1] 983.

[55] [22] [16] [19] [0] 1.

[18] 413 158.

[70] [16] [10] [7] [2] [1] [0] 1.

[19] 781 673.

[54] 305.

[55] [50] [20] [53] [0] 1.

[20] 503 181.

[64] [49] [5] [9] [7] [2] [0] 1.

[21] 538 639.

[63] 299.

[86] [23] [11] [35] [0] 1.

[22] 931 588.

[41] 270 129.

[44] [28] [49] [0] 1 TABLE II. Testing Data Unit: ×10-3 deg Order Group No. 0. 5.

[1] 1. 5.

[2] 2. 5.

[3] 3. 5.

[4] Expected Output Simulation Result.

[1] [72] 496.

[35] 158.

[16] [16] [10] [7] [0] 0 0. 0005 0. 0083.

[2] 343 316.

[8] [50] [5] [8] [4] [4] [0] 0 0. 0024 0. 0097.

[3] [19] 139.

[22] [50] [9] [5] [8] [2] [0] 0 0. 0000 0. 0095.

[4] [31] 277.

[16] [53] [9] [11] [7] [9] [0] 0 0. 0012 0. 0043.

[5] 420 686.

[51] 294.

[56] [24] [9] [24] [0] 1 0. 0042 0. 9142.

[6] 522 615.

[62] 276 102.

[22] [11] [32] [0] 1 0. 0022 1. 0000.

[7] 779 673.

[53] 300.

[57] [49] [19] [54] [0] 1 0. 0046 0. 9793.

[8] 346 176.

[78] [35] [6] [4] [3] [2] [0] 1 0. 0006 0. 9717.