The limitations and applicability of four different residential building energy simulation models at different levels of sophistication have been examined and compared . These models are:

1. The Standard Heat Loss Methodology (SHLM);
2. SUNDAY, a microcomputer based energy simulation model;
3. WORK4, a temperature bin method of heat loads calculation; and
4. DOE-2, a state-of-the-art public domain building energy simulation model.

The comparisons were performed over a wide range of parametric conditions including changes in climate, inside temperatures, internal gains, and UA. We found reasonably close agreement between the WORK4 and SUNDAY models. These two models were also in reasonable agreement with DOE-2, except when modeling well-insulated homes and air infiltration changes. The SRLM model yielded substantially lower energy use estimates than the other models.

In addition, the simulation results for both WORK4 and SUNDAY were collapsed into a smooth trend using a normalization procedure. This simple trend is solely defined by climatic conditions.

Units: I-P