Paper Conference

Proceedings of Building Simulation 2021: 17th Conference of IBPSA


HVAC performance evaluation and optimization algorithms development for large buildings

Mike Stock 1, Mohamed Kandil 1, Brian Macleod 2, Brad Pilgrim 2, Jennifer McArthur 1
1 Ryerson University, Toronto, Canada
2 Parity Inc, Toronto, Canada

Abstract: Data-driven models can be coupled with optimization algorithms to predict optimal setpoints for heating and cooling systems in large buildings. Informed by equipment thermodynamics, these models ingest available building data to infer building performance under a range of weather and control conditions and use this to develop an optimization strategy. This paper presents the development of these models using a case study building in ASHRAE CZ5. Initial results indicate savings up to 70% for heating applications can be achieved beyond standard outdoor air reset control without any physical changes to equipment, offering a low-cost, rapidly deployable solution to reduce building energy consumption and GHG emissions.
Keywords: Existing building performance, data analytics, HVAC system simulation, optimization, continuous commissioning
Pages: 3080 - 3087