Speed profile optimization of catenary-free electric trains with lithium-ion batteries

2021

Preprint

We look into modeling fuel cell hybrid trains for the purpose of optimizing their operation using convex optimization. Models and constraints necessary to form a physically feasible yet convex problem are reviewed. This effort is described as holistic due to the broad consideration of train speed, energy management system, and battery thermals. The minimized objective is hydrogen fuel consumption for a given target journey time. A novel battery thermal model is proposed to aid with battery thermal management and thus preserve battery lifetime. All models are derived in the space-domain which along constraint relaxations guarantee a convex optimization problem. First-principle knowledge and real-world data justify the suitableness of the proposed models for the intended optimization problem.

Section: Battery State-of-chargesupporting

confidence: 55%

Mathematical Modeling for Holistic Convex Optimization of Hybrid Trains

Jibrin¹,

2021

Preprint

“…Predicting the battery's state-of-charge ζ is vital in order to guarantee charge-sustaining operation. The battery considered herein is a lithium-ion battery, modeled as a fixed voltage source U oc with a fixed internal resistance R. Experimental results from a real battery-powered train confirm the validity of such a model [21].…”

Section: Battery State-of-chargementioning

confidence: 63%

Convex Optimization of Speed and Energy Management System for Fuel Cell Hybrid Trains

Jibrin¹,

et al. 2021

Preprint

We look into minimizing the hydrogen fuel consumption of hydrogen hybrid trains by optimizing their operation. The powertrain considered is a fuel cell charge-sustaining hybrid. Convex optimization is utilized to compute optimal speed and energy management trajectories. The barrier method is used to solve the optimization problems quickly on the order of tens of seconds for the entire journey. Simulations show a considerable reduction in fuel consumption when both trajectories-speed and energy management-are optimized concurrently within a single optimization problem in comparison to being optimized separately in a sequential manner-optimizing energy management after optimizing speed. It is concluded that the concurrent method greatly benefits from its holistic powertrain knowledge while optimizing all trajectories together within a single optimization problem.

“…Predicting the battery's state-of-charge ζ is vital in order to guarantee charge-sustaining operation-terminal battery charge identical to initial. The battery is modeled with a fixed open-circuit voltage U oc and a fixed internal resistance R, a model that is accurate for the narrow state-of-charge range employed by hybrid vehicles and is validated by [24].…”

Section: E Battery State-of-chargementioning

confidence: 99%

Convex Optimization for Fuel Cell Hybrid Trains: Speed, Energy Management System, and Battery Thermals

Jibrin¹,

2021

Preprint

We optimize the operation of a fuel cell hybrid train using convex optimization. The main objective is to minimize hydrogen fuel consumption for a target journey time while considering battery thermal constraints. The state trajectories: train speed, energy management system, and battery temperature, are all optimized concurrently within a single optimization problem. A novel thermal model is proposed in order to include battery temperature yet maintain formulation convexity. Simulations show fuel savings and better thermal management when temperature is optimized concurrently with the other states rather than sequentially-separately afterwards. The fuel reduction is caused by reduced cooling effort which is motivated by the formulation's awareness of active cooling energy consumption. The benefit is more pronounced for warmer ambient temperatures that require more cooling.