An assessment of the accuracy of statistically adjusted engineering (SAE) models of end-use load curves

SAE models combine engineering estimates of end-use loads with conditional demand analysis of customer-level loads. The statistical procedure adjusts the engineering loads for observed customer-level loads. Using data from PG & E, SAE end-use loads are estimated for a sample of households. These estimated loads are then compared with metered end-use loads for the same households. It is found that the SAE models improve the engineering estimates considerably for wall and central air conditioning but add error to the engineering estimates for stoves, refrigerators, and dryers.