Weather is a key driving factor of electricity demand. Among all the weather variables included in load forecasting models over the past half a century, temperature is the most commonly used one. Although humidity has also been discussed in the load forecasting literature, it is not as formally studied as temperature. In reality, a large portion of the electricity demand is caused by heating, ventilation, and air conditioning in order to meet people’s comfort level, which is primarily determined by temperature and humidity. In this thesis, how relative humidity affects electricity demand will be investigated.The case study is conducted at North Carolina Electric Membership Corporation, a large generation and transmission cooperative in the United States, for its system total load and the loads of three power supply areas. It is found that relative humidity plays a vital role in driving electricity demand during the warm months (June, July, August and September). This thesis proposes a systematic approach to include relative humidity variables in a regression analysis framework, resulting in the recommendation of a group of relative humidity variables. The proposed models with the recommended addition of relative humidity variables improve the forecast accuracy of Tao’s Vanilla Model and its three derivatives in 24-hour ahead, one-week ahead, two-week ahead and one-year ahead ex post load forecasting settings. The improvement obtained from this case study ranges from 4.05% to 9.39% for NCEMC total ex post load forecasting on the test data (holdout sample).
