Power demand prediction for buildings at a large scale is required for power grid operation. The bottom-up prediction method using physics-based models is popular, but has some limitations such as a heavy workload on model creation and long computing time. Top-down methods based on data driven models are fast, but less accurate. Considering the similarity of power demand patterns of single buildings and the superiority of generative adversarial network (GAN), this paper proposes a new method (E-GAN), which combines a physics-based model (EnergyPlus) and a data-driven model (GAN), to predict the daily power demand for buildings at a large scale. The new E-GAN method selects a small number of typical buildings and utilizes EnergyPlus models to predict their power demands. Utilizing the prediction for those typical buildings, the GAN then is adopted to forecast the power demands of a large number of buildings. To verify the proposed method, the E-GAN is used to predict 24-hour power demands for a set of residential buildings. The results show that (1) 4.3% of physics-based models in each building category are required to ensure the prediction accuracy; (2) compared with the physics-based model, the E-GAN can predict power demand accurately with only 5% error (measured by mean absolute percentage error, MAPE) while using only approximately 9% of the computing time; and (3) compared with data-driven models (e.g., support vector regression, extreme learning machine, and polynomial regression model), E-GAN demonstrates at least 60% reduction in prediction error measured by MAPE.

Quantifying the energy savings of various energy efficiency measures (EEMs) for an energy retrofit project often necessitates an energy audit and detailed whole building energy modeling to evaluate the EEMs; however, this is often cost-prohibitive for small and medium buildings. In order to provide a defined guideline for projects with assumed common baseline characteristics, this paper applies a sensitivity analysis method to evaluate the impact of individual EEMs and groups these into packages to produce deep energy savings for a sample prototype medium office building across 15 climate zones in the United States. We start with one baseline model for each climate zone and nine candidate EEMs with a range of efficiency levels for each EEM. Three energy performance indicators (EPIs) are defined, which are annual electricity use intensity, annual natural gas use intensity, and annual energy cost. Then, a Standard Regression Coefficient (SRC) sensitivity analysis method is applied to determine the sensitivity of each EEM with respect to the three EPIs, and the relative sensitivity of all EEMs are calculated to evaluate their energy impacts. For the selected range of efficiency levels, the results indicate that the EEMs with higher energy impacts (i.e., higher sensitivity) in most climate zones are high-performance windows, reduced interior lighting power, and reduced interior plug and process loads. However, the sensitivity of the EEMs also vary by climate zone and EPI; for example, improved opaque envelope insulation and efficiency of cooling and heating systems are found to have a high energy impact in cold and hot climates.