Heuristic operating strategies for parallel hybrid vehicles in the context of model-based application

Abstract

Hybrid vehicles provide an opportunity to meet the rising demands on modern vehicles. A central role plays the reduction of fuel consumption by using, for instance, appropriate control strategies. They specify the operating points of the built-in energy converters, of the electric machine and of the combustion engine.

This article compares heuristic control strategies for parallel hybrid vehicles. To realize this, a fuzzylogic controller (FLC), as well as the Electric Assist Control Strategy (EACS) approach and the map-based Equivalent Consumption Minimization Strategy (ECMS) are employed. For calculating the fuel consumption, an empirical model of the power train is utilized. To assess the strategies’ performance in diverse driving situations, the results of six driving cycles are compared to the optimal solution based on dynamic programming (DP). Furthermore, the multiple approaches are optimized regarding their parameters. For this purpose, the Particle Swarm Optimization and the Bees Algorithm are employed as two natural analogue methods. This article demonstrates that heuristic approaches provide online-capable possibilities for operating hybrid power trains including a performance close to the global optimum.