Predictive Neural Network Modeling for Diesel Engine Part Wear Assessment via Analysis of Wear Element Concentration in Used Oil

Abstract

Used oil serves as a repository of wear elements, reflective of the friction generated among various engine components. The accumulation of these elements offers vital insights for understanding wear and tear in diesel engine parts. Accurate prediction of such wear is imperative for strategic maintenance planning and cost efficiency.
 This paper presents an innovative methodology that leverages artificial neural network modeling to forecast the degradation of locomotive diesel engine components based on the concentrations of wear elements in used oil and the corresponding vehicle mileage. The proposed supervised artificial neural network comprises two hierarchical architectures: one linking diesel locomotive operation and wear element concentration, and another associating diesel locomotive operation with diesel engine component wear. Through experimental validation, the results highlight the efficacy of the developed neural network model in precisely estimating the wear of diesel engine components. This predictive capability empowers informed maintenance decision-making to bolster engine reliability and performance.