Advanced Offshore Wind Farm Maintenance Scheduling (with Sennen) [£10,000]

Project: Knowledge Exchange

Project Details


EP/P019544/1 focused on improved decisions in an operational context for wind power cost reduction.
One of the new areas for improvement identified on the project was the prioritisation of maintenance tasks for the offshore despatch teams. No standard approach exists, and in discussion with a major turbine OEM it became clear that there is no optimisation done (as current business as usual) to prioritise
jobs. Strathclyde has developed an optimisation approach which shows real promise [Dawid, R., McMillan, D., Revie, M. Time series semi-markov decision process with variable costs for maintenance planning, ESREL 2016, p. 183).
This IAA will facilitate the initial build of such an optimisation capability within Sennen (Primary job of the impact champion), who will then use it to optimise maintenance at their client sites (primarily London Array).
We have demonstrated a potential availability uplift of 0.1 – 0.4% by utilising such an optimised approach [Browell, J., Dinwoodie, I., McMillan, D.; Forecasting for day-ahead offshore maintenance scheduling under uncertainty, ESREL 2016, p. 182.]. This translates to a cost saving (in terms of yield uplift) of £40,000 - £160,000 for every 100MW of offshore wind capacity (assuming 40% capacity factor).
Sennen provide software and services direct to the renewables industry, principally offshore wind. Their software platform has been extensively used for marine operations at London Array. However, the vessel routing and task
prioritization aspects of the marine operations module are fairly basic, and they would look to work with the impact champion to integrate the optimization work through the KTP. The IAA would enable this work to begin, by facilitating the necessary training of the impact champion in the technology required to integrate the algorithms.
Effective start/end date1/11/1715/03/18


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.