THE PEAK ENERGY LOAD MANAGEMENT SYSTEM (PELMS): A CASE STUDY IN MONTREAL

Abstract Extreme warm temperature events cause energy consumption and resultant electrical power demand spikes in large metropolitan areas and pose challenges to energy service providers, while enhancing operational blackout risk. This results in the disruption of essential services and impacts on public health. These events are projected to increase in frequency, especially in further northern latitudes historically less used to them. Current methods to forecast short-term electricity demands lack the complexity required for dense urban environments, in addition to the insufficient resolution and physics of current numerical weather prediction models. The Peak Energy Load Management System, or PELMS, is a novel approach to forecasting energy consumption at sufficient lead times for its information to be employed by utilities. Using the Weather Research and Forecasting Model (WRF), PELMS downscales Numerical Weather Prediction Models to a resolution of 1km and integrates multi-layer urban parameterization and integrated building energy models. With this framework, fine-scale peak load forecasts are possible in advance of extreme events, adjusting forecasts dynamically to minimize grid disruption. Results for PELMS's pilot from the heatwave of June 24-25, 2025, in Montreal, Quebec, Canada, are shown. The model was able to accurately predict the magnitude of the heatwave across the city well in advance and precisely depict building air conditioning demand across the Island of Montreal for operational purposes. The viability of PELMS as a forecast tool is discussed, as well as the initiative's next steps.