Abstract
Due to the recent trend to change from fossil-based fuels to alternative energy sources in the transportsector, many challenges need to be addressed for heavy-duty long-haul transportation. Driving range and
driveability are among these challenges. In order to tackle these challenges, VDL is exploring the feasibility of
a fuel-cell hybrid long-haul coach. The hybrid powertrain is a new concept for long-haul coaches. Therefore,
modelling and analyzing the new vehicle concept is important for optimizing vehicle packaging and the
energy management strategy.
This thesis introduces an optimization algorithm suitable for optimizing the energy consumption of a fuelcell
hybrid coach. The development consists of two main phases; the modeling phase and the optimization
phase. During the modeling phase, low fidelity models suitable for control are developed for the main vehicle
subsystems, i.e., the traction system, battery, fuel cell, and climate system. The low fidelity models are
developed and validated by means of high-fidelity models and measurements performed on component level.
The power demand of the auxiliary subsystems that are not modelled is captured by an exogenous signal that
acts as a disturbance on the control problem. In the optimization phase, the vehicle’s energy consumption
is minimized by formulating and solving an optimal control problem. The developed optimization algorithm
involves convex optimization in order to find the optimal utilization of the energy sources. In order to solve
large scale problems, a split-horizon approach is taken. This approach reduced the computational complexity
of the control problem significantly with minimal impact to the solution. The optimization is performed for
a set of varying operating conditions in order to analyze the impact of external factors and component sizing on the vehicle’s energy consumption.
The optimization results show that the energy consumption of the vehicle is heavily affected by its operation.
Heavy reliance on the fuel cell for energy delivery has shown an increase in consumption up to 66%. The optimization algorithm adapts the fuel cell utilization in order to minimize internal losses and reduce the effort that is required from the climate system. The correct battery sizing has a large impact on the energy consumption of the vehicle, where fuel cell sizing shows a much smaller influence. By reducing the amount of fuel cell units from five to only two, energy consumption increased up to 6.5% under worst-case conditions.
The optimization algorithm provides a platform that can be extended to optimize for running cost or total cost of ownership. Furthermore, additional subsystems, such as cabin climate, can be incorporated in order improve the accuracy of the solution.
| Date of Award | 14 Jan 2022 |
|---|---|
| Original language | English |
| Supervisor | M.C.F. (Tijs) Donkers (Supervisor 1) & A.P.C. Blom (External coach) |