Identification of Reliability States of a Ship Engine of the Type Sulzer 6AL20/24

The article presents results of tests performed with the use of a ship engine of the type Sulzer 6AL20/24. The goal of the tests was to create and verify an identification procedure for the analyzed object’s reliability states to be used without interfering with the object operation processes. The proposed method is based on an analysis of vibrations and noise generated during the engine operation, which are considered to be the most significant diagnostic signals. The signals of the engine vibrations and noise recorded during the engine operation on a laboratory test stand have been analyzed in the time domain. A number of the recorded signal characteristics are calculated. The characteristics are statistically analyzed in order to choose those which can provide the basis for the identification of reliability states. Next, based on the spaces of ability and inability, states are formulated. The spaces are used for the identification of ability states and selected inability states through the determination of a minimal distance between signals in the Euclidean distance metrics. The proposed method was used in the ship engine tests to yield very good identification effects. A description of the applied method and the results are presented in the final part of the article.