Data Predictive Control: Bridging Machine Learning and Control Synthesis for Volatile Energy Systems


In December 2014, the average price of wholesale electricity in the PJM market surged from $25/MWh to $2680/MWh - an 83x increase in 5mins. Demand response (DR) is becoming increasingly important as the volatility on the grid continues to increase. Current DR approaches are predominantly manual and rule-based or involve deriving first principles based models which are extremely cost and time prohibitive to build. We consider the problem of data-driven end-user DR for large buildings which involves predicting the demand response baseline, evaluating fixed rule based DR strategies and synthesizing DR control actions. We provide a model-based control with regression trees algorithm, which allows us to perform closed-loop control for DR strategy synthesis for commercial buildings. Our data-driven control synthesis algorithm outperforms rule-based DR by 17% for a large DoE commercial reference building and leads to a curtailment of up to 380 kW and over $45,000 in savings. Our methods have been integrated into a tool called DR-Advisor, which acts as a recommender system for the building’s facilities manager and provides interpretable control actions to meet the desired load curtailment while maintaining operations and maximizing the economic reward. Built upon DR-Advisor is IAX, an Interactive Energy Analytics engine - think of it as a Siri for querying buildings’ energy use. We are developing IAX to procedurally generate energy dashboards for open-ended questions.

Current Members
Rahul Mangharam's Profile Photo
Rahul Mangharam
Madhur Behl's Profile Photo
Madhur Behl
Profile photo of Achin Jain
Achin Jain