Algorithmic Power Optimisation in Constrained Railway Networks: A Systematic Review

TL;DR

This review highlights the need for advanced, multi-train energy management algorithms that balance computational complexity and human usability to optimize railway network capacity without hardware upgrades.

Contribution

It identifies the limitations of current single-train models and emphasizes the importance of multi-train simulations and human-compatible solutions for capacity enhancement.

Findings

Traditional models are 'grid-blind' and insufficient for multi-train scenarios.

A trade-off exists between deterministic models' accuracy and heuristic approaches' speed.

Current optimal profiles are too complex for practical human implementation.

Abstract

The decarbonisation of heavy-duty railway networks requires maximising the capacity of existing electrical infrastructure. Integrating heavy freight alongside fast passenger services exposes the hard physical limits of conventional alternating current traction networks, causing severe localised power quality degradation, phase unbalance, and low-voltage behaviour that triggers protective substation tripping. Because upgrading physical hardware is highly capital-intensive, software-based Energy Management Strategies have the potential to offer viable solution for preventing these power capacity challenges. This systematic review demonstrates that traditional, single-train optimisations are fundamentally "grid-blind", necessitating a shift toward multi-train simulations to protect the network's Firm Service Capacity. However, evaluating this shift reveals a critical tension between the computational bottlenecks of deterministic models and the latency of heuristic approaches. Furthermore, a fundamental operational gap exists: while current algorithms generate theoretically optimal speed profiles to increase efficiency and therefore reduce power consumption from the grid, these profiles are excessively complex and inappropriate for human execution. Consequently, future energy management frameworks must bridge this human-machine interface gap to realise capacity improvements on constrained mixed-traffic networks.