Collaborative learning of wave energy converters

This project aims to develop new types of control algorithms for wave energy conversion.

Compared to other renewable energy technologies wave energy conversion have the advantage of potentially very high energy absorption.

However, to implement high energy absorption for wave energy conversion are a big challenge since it involves complex numerical models and/or complicated mechanical constructions with gear boxes.

In our approach we will avoid the traditional approach of using complex models and/or complicated mechanical constructions and use a machine learning approach where the wave energy converters collaborate and learn each other how to move in order to increase the energy capture.

With sufficient amount of training the wave energy conversion should be able to learn how to maximize energy absorption based on the recent history but still use a simple mechanical construction. A functional machine learning algorithm will therefore decrease the cost for electricity.