A ship diesel engine abrasive particle type identification method based on information fusion

Abstract

The invention relates to a marine diesel engine abrasive particle type identification method based on information fusion. The method comprises the following steps: obtaining an abrasive particle sample from an on-line oil monitoring system of the marine diesel engine, correcting and denoising an acquired abrasive particle image by utilizing an image processing technology, and extracting morphological characteristics of abrasive particles; determining a reference value set of the input characteristics, and calculating the comprehensive similarity distribution of the input characteristics with respect to the reference values; constructing an input point statistical table reflecting the relationship between the input signal and the five abrasive particle types by utilizing the comprehensive similarity distribution of the sample set; An evidence matrix table of the input characteristic signals is obtained through conversion of the point-throwing statistical table; and fusing the evidence activated by the input sample vector by using confidence rule reasoning and evidence reasoning rules, and reasoning the abrasive particle type corresponding to the abrasive particle sample from a fusion result. According to the invention, intelligent identification of the marine diesel engine abrasive particle type can be realized, the identification complexity is reduced, and the identification precision is improved.

Description

technical field [0001] The invention relates to an information fusion-based identification method for abrasive particle types of marine diesel engines, belonging to the field of safe operation and maintenance of marine diesel engines. Background technique [0002] As the main power source of ships, marine diesel engines are closely related to the safe and reliable operation of ships and the safety of life and property at sea. Since marine diesel engines are composed of a large number of tribological systems, the friction and wear failures of these friction pairs are one of the main types of marine diesel engine failures. Therefore, it is of great significance to carry out the research on friction and wear fault diagnosis of marine diesel engines to prolong the service life of parts and improve the operating efficiency and safety of diesel engines. [0003] Oil monitoring is the main way to monitor the wear state of diesel engines. The abrasive particles contained in the oil...

AI Technical Summary

Problems solved by technology

However, due to the limitations of oil sample collection methods, wear particle feature extraction methods, and limited empirical knowledge, there are usually certain uncertainties or incompleteness in the extracted diesel engine wear particle morphology features. presents a greater challenge

At present, most of the models have limited ability to deal with uncertain information or multivariate information, the interpretability of the model is poor, and they lack the ability to identify incomplete samples, resulting in inaccurate identification results or model failure.

Images