BDA.17 Intelligence Detection and Prediction of Energy at the Device Level

Renewable energy sources (RES) are increasingly becoming an important component in the future power grids.
However, with high dependence on weather conditions, such as wind and sunshine, the integration of RES into the power grid creates huge challenges, both in regards to the physical integration and in regard to management of the expected demand. To confront these challenges, the TotalFlex project will provide a mechanism to extract and utilize the flexibility in demand as a cheaper and environmental solution using the concept of flex-offer proposed in the EU FP7 project MIRABEL. The TotalFlex project addresses the challenges of balancing the energy consumption and demand by utilizing the novel concept of flex-offer, which utilizes the flexibility for device operation in household level to create the flexible consumption offer. In its simplest form, a flex-offer specifies the amount of energy required, the duration, earliest start time, latest end time, and a price. E.g.,"I need 2KW to operate a dishwasher over 2 hours between 11 PM to 8 AM and I will pay 0.40DKK/KWh for it". If needed, the amount and duration of operation can be flexible without exceeding total energy consumption and latest end time. The term flexibility can be defined as customer willing to be flexible with respect to time or energy profile for the operation of their electric devices.

The introduction of the flex-offer concept and the requirement of designing a mechanism for load-forecast and flex-forecast to support the concept, has led to the research issue in which this Ph.D. project targets. This Ph.D. project focuses on the analyses of historical device operation behaviors of the consumers and extraction of flexibility from their device operation patterns. More specifically, it deals with the issue regarding accurate and precise load-forecast, flex-detection, and flex-forecast at device level, and use the forecast for automated generation of flex-offer. Further, we perform an econometric analysis of the flexibility on the electricity market.

The research will utilize current data mining and machine learning techniques and propose new algorithms and state-of-the-art methodologies for load-forecast, flex-detection, flex-forecast, and generation of flex-offer providing substantial information for different perspectives of energy market actors for supply and demand management and enabling mutual benefits. The research will be carried through the acquisition of various relevant datasets which is preceded by a comprehensive literature survey, problem analysis and followed by solution design. Further, this research will disseminate the results, finding and view from the conducted research through discussion with experts and publications of the accomplished work. Finally a complete working demo for TotalFlex project and Ph.D. thesis as a collection of papers will be submitted. Though current ex-offer research mainly focuses on disaggregation of demand and flexibility at the device level, but can be easily extended to household, transformer, and industrial level.