Working cycle condition construction for electric wheel loader based on principal component analysis and cluster analysis

Abstract

With the rapid development of new energy technology, cycle condition is significant for the design of powertrain system, whole machine economy, endurance, and other performance parameters of pure electric drive machines. Many cycle conditions of the passenger car industry have been defined. While that for the loader which has complex operation modes and working conditions, has less research on this aspect. Cycle conditions suitable for loader have not been constructed. To support the research and development of electric loader power system, the cycle conditions that can reflect the driving characteristics of electric loader are studied in this paper. According to the typical “V” cycle condition of loader, the actual driving condition data are collected and the kinematic segments are divided. The kinematic segments are grouped into five categories by using principal component analysis and K-means clustering algorithm to obtain the clustering center of each category. The working cycle condition is formed by extracting the combination of segments closest to the clustering center. Finally, based on AMESim simulation and whole machine test analysis, the error between the simulation and vehicle test results of the energy consumption is 3.4%. The accuracy and effectiveness of the proposed constructed cycle conditions are verified.