Route Optimized Energy Management of Plug-in Hybrid Electric Vehicles

Plug-in hybrid electric vehicles have the potential to significantly reduce the oil dependence within the transportation sector. However, there will always be some trips that exceed the electric driving range, meaning that both electric energy and fuel must be used. For such trips the fuel economy is intimately connected with the energy management system and its ability to schedule the use of the battery. The fundamental problem is that the optimal fuel economy can be reached only if the future trip is known a priori. It is therefore desirable to have a system that can perform three principal tasks: i) acquire a prediction of the future trip, ii) given the prediction precompute feedforward information for the real-time level, and iii) at the real-time level identify the optimal operating points in the powertrain. This thesis investigates all three of the mentioned tasks. It is shown that frequently travelled routes can be identified from logged driving data using hierarchical clustering. Based on the historical driving conditions along the route, it is then possible to precompute an optimal strategy that can be used as feedforward information for the real-time level. Two different methods for such a precomputation are investigated, convex optimization and Dynamic Programming. Particular attention is given to the implementation of a computationally efficient Dynamic Programming algorithm. A real-time control strategy that is based on a closed-form minimization of the Hamiltonian is also presented. The strategy is derived for a powertrain with two degrees of freedom, and is implemented in a dynamic vehicle model that is used by a vehicle manufacturer. Simulations with a linearly decreasing battery state of charge reference indicate that the fuel economy can be improved with up to 10%, compared to a depleting-sustaining strategy. Real-time compatible controller code is also generated and tested in a production vehicle. The vehicle behaviour during a test drive is similar to simulated behaviour.