Assessment of Model-Based peak electric consumption prediction for commercial buildings

Utility programs have successfully delivered energy efficiency for decades. Today, increasing emphasis is being placed on demand response (DR) that incentivize customers to reduce, or “shed” electric load during grid peak periods. The most common methods used predict building peaks and quantify DR reductions rely simple averaging algorithms using hourly temperature data from the days preceding event. In contrast, regression-based been decades annual savings. availability of smart meter has enabled application regressions more accurate savings estimation, often referred as “advanced measurement verification (M&V).” This project explored whether advanced M&V regression approaches offer improvements over simpler prediction in commercial buildings. We present evaluation results eight (based three baseline modeling approaches). findings show all underpredicted consumption across 453 meters 1,100 days. Median bias values varied between 4.5 18.7 percent, indicating evaluated would tend understate achieved applications these did not a notable advantage commonly methods.