Artificial Neural Network Based Fault Detection in a Four Stroke Engine using Acoustic Signal

Prof. Mr. S. N. Dandare; Dr. S. V. Dudul

Fault detection and isolation technique have been developed for automated processes during the last few years. These methods include numerical methods, artificial intelligence methods or combinations of both. Condition monitoring and fault detection techniques are used to prevent early fault in a mechanical systems. The paper deals with the problem of fault detection in an automobile engine employing an artificial neural network (ANN). The fault detection is not an easy task for an inexperienced mechanic or driver because it needs a lot of knowledge for finding the fault. Many times the trial and error approach has been applied to detect the fault and because of that, the engine may get more damaged instead of getting repaired. To overcome such type of problem the new technique has been suggested to diagnose the fault correctly without opening the engine. Therefore, this paper presents the innovative technique to detect the Air Filter fault, Spark Plug fault, Insufficient Lubricants fault, Piston Ring fault and Rich Mixture fault, in a four stroke automobile engine, using a single sensor. The Artificial Neural Networks have been employed to classify the faults correctly. Performance of Multilayer Perceptron Neural Network and Support Vector Machine Neural Network has been compared on the basis of Average Classification Accuracy. The paper further justifies the use of Support Vector Machine Neural Network for classification of the faults.

Select Volume / Issues:

• Volume 3 Issue 5, Sept. 2012

Year:

2012

Type of Publication:

Article

Keywords:

Four stoke Automobile Engine; Air Filter; Spark Plug; Insufficient Lubricants; Piston Ring; Rich Mixture; Artificial Neural Network; Classification Accuracy and Fault Detection Abbreviations

Journal:

IJECCE

Volume:

3

Number:

5

Pages:

1203-1207

Month:

Sept.