Neural Network Based Demand Side Management Using Load Shifting

Abstract

Demand Side Management (DSM) is one of the emerging areas that focus on management of demand at the customer side in order to achieve various benefits such as reduction in electricity cost, reducing peak demand, improving load factor etc. While the objective of any DSM activity considered so far is peak demand reduction, peak to average ratio (PAR) improvement, load factor improvement, user satisfaction maximization etc., the main objective discussed in this paper is the minimization of electricity consumption cost since by reducing the electricity cost, the remaining objectives can be achieved indirectly. Further the potential of DSM is analyzed for a sample test system and DSM is implemented using load shifting technique since this technique typically does not alter total electricity consumption. Neural network is used to create a network and train it according to test system data to minimize mean square error.