AI Battery Optimizer


Funding institution: Regional Research Funds Viken

Electrification is an important step in transforming energy systems to address the climate crisis. New energy sources and changing demands cause large fluctuations in electricity prices, which motivate the use of energy storage systems combined with renewable energy in in households. The project builds on Smart Energy System’s battery energy storage system EnergyBank for developing AI-driven battery optimization algorithms that efficiently control battery use to reduce energy consumption and increase electricity cost-savings in residential buildings. By evaluating and developing prediction algorithms that take input data such as weather forecasts, electricity prices, expected short-term power needs and expected power generation, the production, storage, and management of electricity within the residential sector may be optimized. Evaluation of different prediction and control algorithms and combinations of them into an overall control system for managing energy use will lead to a robust general-purpose solution for various buildings in different areas. Energy forecasting remains a complicated task despite considerable research work, partially due to increasing volatility, changing user behavior and a more diverse energy mix. There is also a need for research on the system behavior of energy installations in households, especially when renewable electricity generation and electric vehicle charging are used. Such an AI-based household Battery Optimizer does not currently exist in the European market and the solution has great potential for cost savings, grid investment deferral and value creation through additional jobs and industry activity. The Battery Optimizer will be demonstrated in pilot households and the main goal is a minimum 20% reduction in energy cost. The project could contribute to several of the UN’s Sustainable Development Goals and benefit Viken, Norway and the rest of the world.