AI Slashes Energy Costs in Commercial Buildings—Here's How It Works

A recent pilot project has shown how AI can improve energy management in commercial buildings. The initiative, known as SET Hub Pilot 4, tested an AI-driven system to optimise electric vehicle charging, heating, and solar power use. Results suggest the technology could cut costs while supporting the electricity grid.

The project focused on developing an AI-based energy management system (AI-EMS) for non-residential buildings. Researchers installed the system at a Fraunhofer IEE office equipped with EV chargers, a heat pump, and a photovoltaic (PV) array. The AI agent then adjusted charging times and energy use to reduce expenses and align with solar generation.

The system relied on reinforcement learning, a method that continuously improves decisions based on real-time data. It also featured a cloud-edge architecture, allowing it to respond quickly to dynamic tariffs and variable grid fees. Tests confirmed the AI-EMS could balance cost savings, energy efficiency, and occupant comfort.

Key factors behind the project's success included a secure communication network, open standards like EEBUS, and standardised interfaces. The report also highlighted new revenue opportunities, such as selling flexibility to energy aggregators or virtual power plants. Regulatory adjustments and wider smart meter adoption were recommended for broader implementation.

The pilot demonstrated that AI-driven control works well in complex energy systems. Recommendations now call for faster smart meter deployment, fewer regulatory hurdles, and better integration of storage and building management. These steps could help scale up the technology across more commercial properties.