DOI: 10.14489/td.2016.02.pp.046-050 

Ovcharenko S.М., Minakov V.A.
A TECHNICAL STATE OF DIESEL ENGINE DETAILS ESTIMATION BASED ON A NEURAL-NETWORK DATA PROCESSING
(pp. 46-50)

Abstract. It is given the results of the study of a possibility of using artificial neural networks for identification of prefailure condition, for example diesel D49, assessment of technical condition, based on the results of spectral analysis of motor oil. The article describes the problems: the study of the geometry of the wear parts CM and CPG; the development of a model of accumulation of wear in the engine oil, taking into account the filtering process, losses in the frenzy, leaks and replacing the engine oil; the algorithm development division of wear on groups of controlled parts; the establishment and training an artificial neural network to solve the problem of determining the intensity of wear parts diesel type of D49. The proposed method has several advantages due to constructing the estimation method on the basis of an organization and functioning of the data. Artificial neural network is a great way to solve issues of identifying and forecasting system. The results are used in the development of software for assessing the wear of diesel engines.

Keywords: diesel, wear of diesel parts, contamination of the engine oil, artificial neural networks.