Green energy networks, games, and getting optimal solutions

Speaker: Liam Timms, University of Melbourne Title: Green energy networks, games, and getting optimal solutions Summary: Power banks present an increasingly accessible, adaptable, and eco-friendly way to use and store power. Multiple power banks can be coordinated to create a virtual power plant (VPP) that can satisfy larger power demands. However, coordinating a VPP introduces difficulties regarding how individual charging decisions interact. In this talk, we model VPP coordination as a non-cooperative game with individual power banks as agents. To solve this model and find optimal charging schedules, we develop an optimisation algorithm based on monotone operator splitting. This algorithm is distributed to reflect the independent agents making up the VPP, and decentralised to avoid the drawbacks of a centrally coordinating agent. Bio: Liam Timms is a PhD student at the University of Melbourne, working on improved algorithms for distributed optimisation. Their research focuses on new algorithms for optimisation problems of ethical and environmental significance, grounded in mathematical theory and computational science. They are particularly interested in the mathematical framework of operator inclusions, which allows a wide array of complex decision-making problems to be addressed and solved with the same techniques.