Using Machine Learning for smart dynamic pricingSession
Blockchain, AI, VR, AR, Digital TwinTrack
Digitalisation - Theatre 3
Hub Sessions Programme
- Flexibility is needed by many energy stakeholders for different purposes (balancing, congestion management, self-consumption). One way to leverage flexibility is through explicit markets, where flexibility is offered under offers (quantity, price…) format. Another way is to incentivize flexibility providers through price signals (implicity market), leaving them the freedom to react or not
- In this presentation we explain how machine learning/AI is used for dynamically determining the best price signal to be sent to the end customers in order to achieve a given objective (e.g. reduce system imbalance), based on the desired reaction and important features (e.g. user preference/behavior, events, external factor such as temperature, weather, season, etc.)
- The Smart Price Engine is tested in the framework of the IO.Energy ecosystem (IoE), allowing to test the whole communication loop from the price engine to the smart appliance.