Misfire Fault Diagnosis Method of Gasoline Engines Using the Cosine Similarity Measure of Neutrosophic Numbers

Abstract

This paper proposes a distance measure of neutrosophic numbers and a similarity measure based on cosine function, and then develops the misfire fault diagnosis method of gasoline engines by using the cosine similarity measure of neutrosophic numbers. In the fault diagnosis, by the cosine similarity measure between the fault knowledge (fault patterns) and required diagnosis-testing sample with neutrosophic number information and its relation indices, the proposed fault diagnosis method can indicate the main fault type and fault trends. Then, the misfire fault diagnosis results of gasoline engines demonstrate the effectiveness and rationality of the proposed fault diagnosis method. The proposed misfire fault diagnosis method not only gives the main fault types of the engine, but also provides useful information for future fault trends. The proposed method is effective and reasonable in the misfire fault diagnosis of gasoline engines.