Traditional electricity meters only reported total household electricity use. While this provides useful billing data, it doesn’t reveal where the energy is going. Our disaggregation technology looks to bridge that gap.

Smart meters capture high-resolution consumption data (voltage, current, and power over time). Using advanced algorithms, we extract the unique “signatures” that different appliances leave in the data stream. For example, a refrigerator cycling on and off looks very different from a washing machine or a kettle.​

We train AI models on large, diverse datasets of appliance usage. These models learn to identify and classify energy patterns with high accuracy, even in complex households where multiple devices run simultaneously. The system converts raw meter data into a clear breakdown of energy use across major appliances and categories — heating, cooling, hot water, cooking, lighting, and more.​

Unlike sensor-based solutions that require extra plugs or monitors, our approach works directly on existing smart meter infrastructure. This makes it highly scalable, cost-effective, and easy to roll out across millions of households.