Apply weather normalization (degree days, IPMVP) to measure energy savings

Verified steps

1. Establish a reporting period baseline: collect 12–24 months of pre-retrofit interval or monthly energy consumption data aligned to the same meter and billing boundaries that will be used post-retrofit; document any baseline adjustments for non-routine events (equipment changes, occupancy shifts) per IPMVP Option C requirements.
2. Retrieve Heating Degree Days (HDD) and Cooling Degree Days (CDD) for the facility's location from a weather station within an acceptable proximity: use NOAA Climate Data Online (ncei.noaa.gov/cdo-web/) for historical daily HDD/CDD data, or compute from hourly dry-bulb temperature relative to a base temperature appropriate to the building type (commonly 65°F / 18°C).
3. Build a regression model between baseline energy consumption (dependent variable) and HDD, CDD, and other independent variables (production, occupancy hours) using ordinary least squares; validate goodness-of-fit with CV(RMSE) < 25% and R² > 0.75 per ASHRAE Guideline 14 thresholds.
4. Apply the regression model to reporting-period weather data to produce a weather-normalized baseline: predicted_consumption = intercept + coeff_HDD * HDD_reporting + coeff_CDD * CDD_reporting + other_terms.
5. Calculate savings = weather_normalized_baseline - actual_reporting_period_consumption; express uncertainty bounds using the regression confidence interval propagated through the prediction, as required by IPMVP for savings verification.
6. Document the M&V plan including: baseline period, independent variables, regression equation, weather data source and station ID, and the routine and non-routine adjustment policy — this documentation is required for third-party verification under IPMVP.