Rapid changes are being experienced by the energy systems requiring to house an increasing amount of renewable energy sources, and also new kinds of loads like those obtained from decarbonized transport.
Credit: HW University
Energy communities will play a key role in building the more decentralised, less carbon intensive, and fairer energy systems of the future. Such communities enable local prosumers (consumers with own generation and storage) to generate, store and trade energy with each other using locally owned assets, such as wind turbines, rooftop solar panels and batteries. In turn, this enables the community to use more locally generated renewable generation, and shifts the market power from large utility companies to individual prosumers.
Energy community projects often involve jointly-owned assets such as community-owned wind turbines or shared battery storage. Yet, this raises the question of how these assets should be controlled - often in real time, and how the energy outputs jointly-owned assets should be shared fairly among community members, given not all members have the same size, energy needs or demand profiles.
New machine learning method accurately predicts battery state of health
Researchers from the Smart Systems Group at Heriot-Watt University in Edinburgh, UK, working together with researchers from the CALCE group at the University of Maryland in the US, have developed a new method to estimate battery health irrespective of operating conditions and battery design or chemistry, by feeding artificial intelligence (AI) algorithms with the raw battery voltage and current operational data.
A paper describing the method is published in the journal
Nature Machine Intelligence.
In the reported study, the team designed and evaluated a machine learning pipeline for estimation of battery capacity fade a metric of battery health on 179 cells cycled under various conditions. The pipeline estimates battery state of health (SOH) with an associated confidence interval by using two parametric and two non-parametric algorithms. Using segments of charge voltage and current curves, the pipeline engin