About the Project

WillowWatt is partnering with Willow to revolutionize energy management in buildings through advanced AI, Machine Learning (ML), and Optimization techniques. Buildings in the US consume 76% of all electricity and contribute 40% of total CO₂ emissions, making energy efficiency a critical challenge.

Our capstone project focuses on co-optimizing buildings and assets to maximize energy and operational efficiencies while minimizing costs and environmental impact. By leveraging Willow's cutting-edge digital twin technology and real-time data, we are building a solution that dynamically optimizes energy usage across hundreds of facilities and assets.

We’ve designed and trained a machine learning model using Random Forest Regression to forecast weekly energy consumption. The model is built in Python using Scikit-learn and is packaged with ONNX for deployment directly within Willow’s platform. Our system visualizes these forecasts, identifies peak energy usage periods, and simulates strategies for reducing loads, such as adjusting HVAC or lighting schedules.

This approach allows decision-makers to act before inefficiencies occur, rather than reacting after the fact. The goal is to reduce energy costs, lower greenhouse gas emissions, and enhance energy resilience, contributing meaningfully to NAU’s carbon neutrality target for 2030.

The initial concept for this project was provided by our sponsor, in the form of a Capstone project proposal, but the resulting system reflects months of collaborative design, prototyping, and refinement with direct input from Willow's Director of Energy Transformation.