Building a predictive model for occupancy and energy usage usage using Machine Learning

Closed
Feedback Solutions
Burlington, Ontario, Canada
Preferred learners
  • Ontario, Canada
  • Academic experience or paid work
Categories
Data analysis Data modelling Software development Machine learning
Skills
project management software
Project scope
What is the main goal for this project?

To build a predictive model using occupancy and energy usage data to forecast expected occupancy and energy availability (kWh) to off-load to the grid.


The system to be developed will use historical occupancy data as well as energy utilization data over a time period to build a model to predict future usage. The usage data will be compared to actual available energy as per design occupancy and ventilation load. The delta between the design mode and the actual usage will determine the available energy that can be sold back to the grid.

What tasks will learners need to complete to achieve the project goal?
  1. Analyze occupancy trends and build a model to predict expected space usage
  2. Analyze energy usage to predict expected energy usage and availability to off load to the grid.
  3. Design a flow diagram to interface with an external grid management software and controller to automatically adjust how much energy can be made available to the grid during peak demand.


About the company

Feedback Solutions leverages best-in-class people counting sensors with its patented technology platform to continuously calculate highly accurate real-time occupant counts based on user-defined zones within a building. This data is then communicated via: BACnet/IP, cloud platform or DDC controller – so that ventilation requirement are optimized seamlessly in real-time based on actual occupant demand. These real-time adjustments reduce energy consumption by as much as 40%, reduce GHG emissions and result in less wear and tear on critical HVAC equipment – all while meeting space ventilation requirements as per ASHRAE guidelines.