Machine Learning-Based Vehicle Model Construction and Validation—Toward Optimal Control Strategy Development for Plug-In Hybrid Electric Vehicles

Abstract

Advances in machine learning inspire novel solutions for the validation of complex vehicle models, and spur an easy manner to promote energy management performance of complexly configured vehicles, such as plug-in hybrid electric vehicles (PHEVs). A constructed PHEV model, based on the four-wheel drive passenger vehicle configuration, is validated through an efficient virtual test controller (VTC) developed in this paper. The VTC is designed via a novel approach based on the least square support vector machine and random forest with the inner-interim data filtered by the ReliefF algorithm to validate the vehicle model as necessary. The paper discusses the process and highlights the accuracy improvements of the PHEV model that is achieved by implementing the VTC. The validity of the VTC is addressed by examining the PHEV model to mimic the characteristics of internal combustion engine, motor and generator behaviors observed through the benchmark test. Sufficient simulations and hardware-in-loop test are employed to demonstrate the capability of the novel VTC based model validation method in practical applications. The major novelty of this paper lies in the development of a VTC, by which the vehicle model can be efficiently developed, providing solid framework and enormous convenience for control strategy design.

A Dynamic Optimal Scheduling Strategy for Multi-Charging Scenarios of Plug-in-Electric Vehicles Over a Smart Grid

“…In recent years, car manufacturers and federal financial institutions have preferred the adoption of PEVs because they have low oxides emission, better torque, energy saving and easy maintenance [33]- [36]. However, a substantial percentage of PEVs will cause massive instability in the operation of power grids [37], [38]. Due to inherent autonomy of PEVs, an efficient charging model of all PEVs in a given urban area is presented in [39], [40].…”

Development of a Volkswagen Jetta MK5 Hybrid Vehicle for Optimized System Efficiency Based on a Genetic Algorithm

“…However, in Ref. [19], the rule-based method uses pre-defined rules to determine when to switch between engines and electric motors considering speed, load, and battery status. In [1], an advanced RBC strategy based on machine learning was proposed to develop the mode control map.…”

Energy Management in Hybrid Electric Vehicles: A Q-Learning Solution for Enhanced Drivability and Energy Efficiency

“…Four operating modes are allowed where in turn the engine (pure thermal) or electric motor (pure electric) provides traction alone, or simultaneously (power-split) or the engine provides traction while recharging the battery (battery charging). Depending on the final application considered (off-line or on-line), several control logics [4][5][6] can be adopted including rule-based [7][8][9][10][11][12], optimization-based [13][14][15][16][17][18][19], data-driven [20][21][22][23][24][25][26][27][28][29][30][31][32] and reinforcement learning (RL) [33][34][35][36][37][38][39][40][41][42][43][44] among the main ones. Rule-based controllers require a substantial calibration effort and they fail to achieve good performance when applied to a driving scenario other than the calibration one.…”