Optimisation of hydrogen fuelling station operation and maintenance to maximise performance and resilience of key infrastructure

This project is in collaboration with ITM Power , who are leading the sector in the design and deployment of electrolyser based hydrogen solutions, including hydrogen refuelling stations (HRS). The deployment of early-to-the-market plants provides the opportunity to optimise not only the operation of the existing plants but to also improve the design of next generation fuelling stations.

It is obviously desirable to maximise performance of the HRS not just for economic reasons, but in order to deliver the best customer experience. This project seeks to optimise the plant operation, where planning preventative maintenance can help reduce disruption to service and improve the commercial case of a plant. There are two approaches being proposed for the project:

1. Component-based asset management

Investigation of the individual "components" (such as electrolyser, refrigerator system, compressor, etc.), analysing their operational, maintenance and fault data. Looking at the performance of a "component", we can identify how long it will be before a failure occurs and then recommend a preventative maintenance schedule. Operational parameters (e.g. flow rates, temperature, pressure, power demand, etc.) can provide early indications of a failure event in the near future when the component is no longer within normal operational parameters and identify any requirement for unscheduled preventative maintenance.

The project will analyse existing plant data using pattern recognition and machine learning methods to look for early signs of failure. 2. System-based asset management

Taking a more system-wide or network-wide approach can help improve resilience. In a system we need to understand redundancies between components, as one component failure can be compensated by a redundant component. Where there is no or limited redundancy, a cost benefit analysis will be undertaken to guide future design principles for resilient HRS.

In a network, there is a need to understand where the stations are located and what are the effects to reliable delivery if one fails. This analysis will help guide strategic clustering of HRS defining the density and optimised geographic location.

To model a system or a network, a suitable alternative to a more traditional Fault Tree Analysis method would be a Petri net-based approach. It would be used to simulate system performance and evaluate its resilience when failures occur.

The approach taken (or a blend from the above) will be tailored to the successful candidate's expertise and interests. The project will involve periods of working at ITM Power, either at their new factory in Sheffield or at one of their HRS, to facilitate effective knowledge transfer between the University research group and the experts at ITM Power.