Analysis of direct load control using multiple linear regression techniques

A method for estimating the load and energy deferred due to load management (LM) at North Carolina Electric Membership Corporation (NCEMC) is described. Two rural feeders are studied using multiple linear regression (MLR). The development of the model for diversified load and a way to show the LM effectiveness are described. It is shown that the NCEMC LM system is effective for reducing demand, but it is also shown that most of the energy that is deferred by load control is lost. While more benefits are gained by controlling during the power supplier´s demand peak than not controlling, it is important to minimize the length of control in order to maximize energy sales