Determining Appropriate Heating and Cooling Thermostat Set Points for Building Energy Simulations for Residential Buildings in North America

Existing building energy simulation tools have a purported tendency to over-predict heating use and, in some cases, to under-predict cooling use, an outcome often attributed to inappropriate assumptions about thermostat management. As a result, a systematic review was conducted for the purposes of determining typical heating and cooling set points for single-family houses in North America. The preliminary consensus results provide an empirical basis for establishing typical inputs for building energy simulation models, increasing prediction accuracy of heating and cooling loads. The results of this review have been applied as default values in the Lawrence Berkeley National Laboratory’s Home Energy Saver and Home Energy Scoring Tools, significantly improving on values previously in use. The review evaluated various data sources showing measured data on heating and cooling temperatures in buildings. Measured data from eighteen studies spanning more than three decades were examined. Our evaluation also considered recent 2009 survey data on heating and cooling set points from the Residential Energy Consumption Survey (RECS) as well as a very detailed recent series of measurements in eighty homes geographically distributed around the United States. The evaluation found that there is diurnal variation in the temperature maintained in homes in all studies and that accounting for this variation will slightly reduce simulated heating and cooling loads (compared to an assumption of constant temperatures) since the interior temperatures largely vary in phase with the outdoor temperature. For central systems, typical evening heating thermostat set points averaged 67F with setbacks to 64F during nighttime hours. Similarly, for cooling thermostat set points averaged 75F with daytime setups to 78F and 77F during the occupied evening hours. These recommendations account for evidence of unequal room temperature distributions of at least 1-2F, even with central space conditioning equipment. If a constant value is used, a thermostat setback of 3°F is recommended as a measure for heating; for cooling, the daytime set up should be assumed should be 2°F. A realistic assumption for conditioned basement temperatures is 7F below the main zone. Further, the data clearly demonstrated that zoned heating and cooling systems exhibit relaxed thermostat set points of a least 1F beyond those recommended for central systems (eg. the occupied evening thermostat setting would be 66F for a zoned heating system). It should be mentioned, that there was clearly shown to be geographic differences in cooling and heating temperature preference, but the data sources were not of sufficient quality to include the effect.