Stochastic Model Predictive Energy Management of Electric Trucks in Connected Traffic

Abstract

This paper proposes a cost-effective power management strategy utilizing the data provided by V2I communication techniques for dual electric machine coupling propulsion trucks. We formulate a bilevel program where the high-level optimizes operation mode implicitly, while the low-level computes an explicit policy for power distribution of two electric machines. Stochastic model predictive control (SMPC) strategy is employed at the high-level, the performance of which highly depends on the prediction accuracy of future driving information. To establish a position-dependent stochastic velocity predictor using limited amount of historical data, two improved approaches are developed: 1) Predictor using multiple features; 2) Predictor combining data and model. Simulations are performed to validate the performance of the proposed predictors compared with a benchmark. The results show that the controllers using the proposed predictors can reduce driving cost by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$3.36 \,\%$</tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$4.26 \,\%$</tex-math></inline-formula> , respectively.