TA-23-C042 – Development of Three Inverse Modeling Approaches for Zone-Level Virtual Metering of Heat Supplied by Perimeter Heaters

The increasing concern to make buildings more sustainable calls for the development of various control strategies and algorithms that improve energy efficiency, without compromising occupant comfort. However, analyzing the performance of these strategies requires extensive sub-metering to sufficiently account for the energy used at the zone-level, which can be costly and raises many practical issues. Therefore, Virtual Meters (VM) are needed to estimate the heat supplied at the zone-level. This approach uses data obtained from the Building Automation System (BAS) and leverages inverse modeling approaches to estimate the heat added by perimeter heaters. This paper presents three approaches to creating VMs, namely 1) steady-state inverse grey-box modeling, 2) transient inverse grey-box modeling, and 3) linear regression steady-state modeling. The accuracy of a VM depends on the mathematical selection and the approach used for inverse modeling. These VMs can then be used to quantify the effectiveness of any retrofits or control sequence changes at the zone level. To validate the developed models, they were applied to data obtained from the BAS of an academic building in Montreal, Canada. This study conveys methodologies that provide a basis for measurement and verification strategies that are focused on improving zone-level energy efficiency. Comparative assessments based on assumptions, data requirements from the BAS and the estimates provided for zonal energy consumption by these models are presented, which is beneficial for building operators aiming to optimize operations.