Adaptive critic control of the power train in a hybrid electric vehicle

Abstract

Summary form only given. An adaptive critic control algorithm is used to control the flow of electric power in the power train of a hybrid electric vehicle. This system consists of a fuel cell which serves as the primary power source, a flywheel/battery to recover regenerative power and provide power for acceleration, and separate motor/generators on each wheel; connected by an inverter which controls the power flow between the various components. The control system for the power train employs an adaptive critic algorithm which is used to control the power flow to and from the fuel cell, battery/flywheel, and motor generators; while optimizing a performance measure designed to meet performance specification with minimal fuel usage.