Mission Management for Fuel Cell Electric Trucks - a Bilevel Approach
Ladda ner
Publicerad
Författare
Typ
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
Abstract
Fuel cell electric vehicles (FCEVs) have a crucial role in the transition to fossilfree travel and transport. In order to fully leverage their potential, proper mission planning of such vehicles should be conducted. This thesis aims to solve the mixedinteger problem of mission management with a hierarchical bilevel approach. The first layer solves a simplified mixed-integer problem of choosing hydrogen stations to
refuel at and the amount to refuel with the help of stochastic dynamic programming, allowing us to account for uncertain external factors to obtain a more robust solution. The second layer becomes a smooth problem that can be solved with more complex and comprehensive dynamics. In addition, the benefits of using non-equidistant stage sampling were evaluated. To evaluate the robustness and optimality of the proposed approach 3 tests were conducted. The simulations were conducted with a 45-ton truck using 3 routes: Zagreb–Paris (1559 km), Gothenburg–Kiruna (1793 km), and Gothenburg–Rødby (461 km). Overall, according to the results, the bilevel approach can yield solutions comparable in optimality to the ones that were obtained with different strategies. In some cases, the proposed approach produced better results when it comes to robustness in the presence of disturbances, but the results were not conclusive enough to answer either of the research questions with a high level of confidence. Even though the non-equidistant stage sampling did not show any substantial increase in simulation accuracy, it was shown that stage and DP stage resolutions can be reduced without significant losses in optimality, making the
computation time almost 3 times smaller.
Beskrivning
Ämne/nyckelord
Keywords: Fuel cell electric trucks, Mission Management, Stochastic optimal control, Mixed-integer nonlinear program.