An extrapolation framework for torque spectrum of excavator internal combustion engine via bivariate diffusion-based kernel density estimation

Abstract

The torque extrapolation of the excavator internal combustion engine (ICE) is unknown and challenging. In this paper, an extrapolation framework for ICE torque of excavator is proposed. Data pre-processing, distribution estimation, load extrapolation, and damage evaluation are included in the framework. In estimating measurements, bivariate diffusion-based kernel density estimation is used to obtain the probability density. The cumulative cycle frequency and pseudo damage are compared with the results obtained by the other extrapolation methods. The results show that the proposed method can realize reasonable extrapolation. This extrapolation process can provide a reference for other engineering equipment in cyclic operation.