论文信息 - Study on Diesel Engine Fault Diagnosis Based on Integration of Bayesian and Genetic Neural Network Method

Study on Diesel Engine Fault Diagnosis Based on Integration of Bayesian and Genetic Neural Network Method

The characteristic vector of cylinder vibration signal is extracted by wavelet packet decomposition. The factor of selection is proposed to take adaptive integration on basis of improved super parent one dependence estimator Bayesian method and back propagation genetic algorithm neural network method. Experimental results on WD615 diesel engine showed that the method has high accuracy rate of detection.

[1] Dan Boneh,et al. On genetic algorithms , 1995, COLT '95.

[2] Wang Xin,et al. Improved Naive Bayesian Classifier Method and the Application in Diesel Engine Valve Fault Diagnostic , 2011, 2011 Third International Conference on Measuring Technology and Mechatronics Automation.

[3] Eamonn J. Keogh,et al. Learning augmented Bayesian classifiers: A comparison of distribution-based and classification-based approaches , 1999, AISTATS.