Advanced definition and analysis of anti-icing methods to balance energy usage and dependability in railways

Aiming at promoting intelligent railway systems, this paper proposes advancements in the efficient management of heater-based anti-icing systems for railroad switches. The goal is to reduce energy consumption while ensuring reliable performance. The main novel contributions comprise both a richer set of software policies for an efficient anti-icing system management by looking at the overall lifetime of the anti-icing system, where normal operating conditions alternate to periods where failures may be experienced, and a more realistic representation of weather conditions by removing some simplifying assumptions. An advanced stochastic modeling framework supports the novel extensions, and allows for the assessment and comparison of a comprehensive range of heating policies in terms of energy and dependability-related indicators, as demonstrated by an extensive analysis campaign performed in a variety of weather scenarios. Given its generality, both in terms of weather profiles and weather-based switch heating policies (in principle, even beyond those developed in the paper) that it can manage, the proposed framework positions itself as a powerful support for railway companies in selecting policies that achieve an appropriate balance between energy consumption and dependability.