Normal Year outputs are generated by using CalTRACK models to predict energy consumption and energy consumption changes in a normal weather year (a hypothetical calendar year where the weather is the average of the weather for 15 or more prior years). For analyses that require estimates of energy consumption changes, two models are fit - one for the baseline (pre-intervention) period and one for the reporting (post-intervention) period. For analysis where only weather-normalized baseline metrics are required (program planning and customer targeting), a single model is fit to the baseline period. This approach allows for apples-to-apples comparisons--savings and other metrics are calculated under the same weather conditions, and there is less worry about unusual weather. This makes it the preferred approach for retrospective program review and for long term demand forecasting, including efficiency initiatives. The following derived quantities are typically generated from normal year analyses:
Normal Year Consumption: By substituting normal year weather conditions into the baseline model, one can obtain an estimate of Normal Year energy consumption. Similar quantities can be derived for post-intervention conditions by substituting normal year weather conditions into the reporting model.
Normalized Energy Consumption Changes / Energy Savings: Energy consumption changes or savings can be determined by subtracting the normal year consumption for the reporting model from that of the baseline model. This method is useful for estimating long term impacts of efficiency programs and it requires a full 12 months of data in both the baseline and reporting periods.
Savings Uncertainty: Since two models are being used to estimate energy savings, their uncertainty needs to be aggregated. This can be done by taking the square root of the sum of squares of the uncertainties for the two models (assuming these are independent).