Skip to main content
WAM

Project WAM 2025

Posted March 04, 2025

Project WAM, short for Wind farm Aram Minimisation, explores the optimal trade-off when designing a wind farm layout to maximise its annual energy production and minimise its footprint. The project is funded by the Department for Energy Security and Net Zero (DESNZ).

Historically, there have been no strong incentives for wind energy developers to consider shrinking the area coverage of an array, as reducing the space between turbines can lead to more energy being lost to wakes, which negatively impacts farm production and revenues. However, with the introduction of non-price assessment criteria into offshore wind auction designs, new incentives are being created for developers to consider deviating away from design norms.

Collaboration

Project funding was split 50:50 between DESNZ and EDF power solutions, who are supporting the project. The government awarded the grant following a successful application to its AI for decarbonisation stream 3 innovation competition. Project WAM is a collaborative project with contributions from R&D, EDF power solutions, Digital Innovation and the Natural Hazards teams, with a dedicated project manager from R&D Operations.

Modelling

Single and multi-objective optimisation techniques were used to develop internal models and tools. These were used to find optimal wind farm layouts that would reduce the footprint while maximising annual energy production. Modelling work demonstrated the use and accessibility of both approaches to wind farm developers, and signalled to the government that wind farm area is a valid new dimension on which they can ask developers to compete. A web platform was also built by the Digital Innovation team, allowing users to upload a simple text file of inputs and receive results by email when the model has run on the platform.

Next Steps

The project is in its final stages. The team will now engage potential end users at EDF Renewables to trial newly developed tools in a commercial setting.