ABOUT THIS POSITION

TNO has developed an energy management strategy called Modular Energy Management Strategy (MEMS) which is based on optimal control and distributed optimization using dual decomposition. This development is for a large part based on the work of Romijn [1], where powertrain topologies are represented as power nets, were a distinction is made between energy storage devices and energy converters, which are connected to each other via nodes.

However, this approach using optimal control has its limits, because to guarantee optimality the whole power demand cycle needs to be known a priori and constraints can only be defined as start and endpoint constraints. Romijn in [1] does however propose an alternative approach to solve the energy management problem by using an model predictive control (MPC) framework instead of an optimal control framework. This approach would only require a limited receding horizon to be known, which is assumed to be extracted from preview information (Route information, ADAS,…). Also the MPC framework allows for explicitly dealing with instantaneous constraints.
References
[1] Romijn, T. C. J. (2017). A distributed optimization approach to complete vehicle energy management. [Phd Thesis 1 (Research TU/e / Graduation TU/e), Electrical Engineering]. Technische Universiteit Eindhoven.
The objective is to convert the existing MEMS energy management strategy, which is based on optimal control, to a MPC based framework. To show a proof of concept for a given use case powertrain configuration, and to compare the performance (e.g. in terms of optimality, robustness, complexity, computational load) with the original framework.