Optimising fuel cell hybrid train operation: a convex optimisation approach

Jibrin, Rabee Raed Sari ORCID: 0000-0001-7743-8930 (2024). Optimising fuel cell hybrid train operation: a convex optimisation approach. University of Birmingham. Ph.D.

Preview

Jibrin2024PhD.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (3MB) | Preview

Abstract

This work focuses on optimising the operation of fuel cell hybrid trains with the primary objective of reducing fuel consumption to improve financial viability. It builds on the basic premise that conventional practices are suboptimal for fuel cell hybrid trains due to the novelty of this class of trains. Given the hybrid nature of this powertrain, it is hypothesised that it is necessary to jointly optimise the energy management system along the driving style and timetable to achieve true optimal operation. Furthermore, it is hypothesised that implementing measures against fuel cell degradation to extend lifetime would come at the expense of additional fuel consumption.

Jointly optimising multiple variables that are dynamically coupled increases computational complexity, especially when they are coupled non-linearly. Therefore, this work sets out to use convex optimisation, as it comes with practical guarantees on global optimality and computational complexity. Formulating a convex problem that models the system adequately is non-trivial which is why it is hypothesised whether it is possible.

A series of optimisation problems with increasing levels of fidelity are built and convexified with various techniques, such as replacing non-linear variables, introducing surrogate variables, approximating discrete data by convex polynomials, and relaxing non-linear equality constraints. Moreover, a hybrid modelling approach that relies on both spatial and temporal domains is proposed. All of this enables building a joint convex optimisation formulation for speed, energy management and timetable.

Simulation results prove that jointly optimising speed and energy management is considerably more fuel efficient than coasting, as the latter fails to effectively leverage regenerative braking to recover kinetic energy. Adding measures against fuel cell degradation was found to come at the expense of additional fuel consumption, though the percentage reduction in degradation was found to be substantially greater than the additional fuel use. Lastly, jointly optimising speed, energy management and timetable was found to yield better fuel economy by distributing running time between stations more efficiently.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):