Fault Diagnosis Method for Internal Combustion Engines Based on IHS-RVM Model
Source: B. Yang, G. F. Song, L. Z. Shen, Y. A. Ghuktomova, & J. X. Xu, "Fault Diagnosis Method for Internal Combustion Engines Based on IHS-RVM Model", Journal of Mechanical Engineering Research and Developments, vol. 40, no. 1, pp. 64-71, 2017.
Document Type: Research Article
Abstract: To deal with the problems of existing internal combustion engine misfire fault diagnosis models like low recognition accuracy and long classification time, this paper proposes an IHS-RVM model based on an improved harmony search(IHS) and relevance vector machine(RVM). Firstly, determination methods of the three parameters in HS algorithm, HMCR, PAR and BW, are improved and the improved HS algorithm is obtained. Then, we use IHS to optimize RVM parameters and build the IHS-RVM model of internal combustion engine fault diagnosis model. Finally, testing data of the whole vehicle is collected to testify the diagnosis performance of the model and make comparison of other models. Diagnosis case study has shown that, IHS-RVM model enjoys a higher and faster classification ability than IHS-SVM, IHS-BP and HS-BP models, with a diagnosis accuracy of 100% and a better anti-disturbance performance of data noise.
Keywords: Exhaust; Internal combustion engine; Misfire; RVM; HS.