Building Energy Management - Heating Efficiency with No Comfort Loss

Our client is a Nordic utility company that is one of Europe’s largest producers and retailers of electricity and heat
  • Meeting the peak electricity and heat demand is costly and requires fossil-based, carbon-emitting fuels to be used
  • Current heating optimization methods are not successful in keeping the heat at the comfort level and result in client dissatisfaction
  • Time-series forecasting models were used to predict room, hall and aisle temperature for various temperature set point scenarios considering weather forecast
  • Data-driven heuristics algorithms were used to approximate the optimal temperature set point dynamically at every 5 minutes to keep the temperature at the desired level while reducing peak energy demand and cost of heating
Tool Stack
  • AWS, Post-gre, Python



Peak Demand Shaving (aka Cost Reduction)



Increased Comfort



Reduced CO2 Emission