A data-driven methodology for enhanced measurement and verification of energy efficiency savings in commercial buildings

Methods to obtain accurate estimations of the savings generated by building energy efficiency interventions are a topic great importance, and considered be one keys increase capital investments in conservation strategies worldwide. In this study, novel data-driven methodology is proposed for measurement verification savings, with special focus on commercial buildings facilities. The presented approach involves use characterization means clustering technique that allows extract typical consumption profile patterns. These then used, combination an innovative evaluate building’s weather dependency, design model able provide dynamic achieved savings. method was tested synthetic datasets using simulation software EnergyPlus, as well monitoring data from real-world buildings. results obtained were compared ones provided applying time-of-week-and-temperature (TOWT) model, showing up 10% CV(RMSE) improvement, depending case analysis. Furthermore, comparison deterministic EnergyPlus showed median estimated error always lower than 3% total reporting period consumption, similar accuracy retained even when reducing training available.