New Intelligent Condition Monitoring and Fault Diagnosis System for Diesel Engines Using Embedded System

Abstract

Diesel engines are power source of various vehicles. Their complex structures make it difficult to detect pending faults in a timely manner. In fact, simultaneous existence of different faults is frequently found in diesel engines. To ensure the security of diesel engines, it is involved with two issues. One is the on-line condition monitoring, and the other is the intelligent fault diagnosis. However, the investigations have been addressed to one of the two problems extensively in previous work while very limited work has been done to develop an on-line intelligent condition monitoring and fault diagnosis (CMFD) system. For this reason, a new intelligent CMFD system is proposed to on-line analyze the vibration signals using the Hessian-based locally linear embedding (HLLE) and support vector machine (SVM) in this work to detect the diesel engine faults and to prevent the malfunction of the machines. The vibration signals were firstly recorded on-line by an embedded system. Then the HLLE was used to reduce the dimensions of original vibration signals to extract useful fault features, and lastly the SVM was adopted for faults detection. Experimental tests have been implemented to evaluate the performance of the on-line intelligent CMFD system. The diagnosis results demonstrate that the proposed method is very effective for the fault diagnosis of diesel engines. The faults can be detected on-line with a high rate of 90%