Fuel Economy Improvement Strategy for Light Duty Hybrid Truck Based on Fuel Consumption Computational Model Using Neural Network

Abstract

This paper describes a strategy for fuel economy improvement of light duty truck with parallel hybrid system. The main objective of this paper is to develop a new hybrid controller which optimizes the torque distribution among various running situations and driver's characteristics with on-line simulation, computing fuel and electric current consumption by using neural network models of the hybrid ECU. Then, fuel and battery current consumption computational models with respect to battery state of charge (SOC), engine and motor torque and engine speed are synthesized by using neural network, and the models are based on experimental data. Finally, the new hybrid controller including the above mentioned models is developed, and its effectiveness on fuel economy improvement is verified by using computer simulation.