A novel information technology of load events detection for the energy management information systems

Coordinating economic load demand response (ELDR) strategy with energy efficiency and information technology (IT) of e-business management multiplies the reduction in electricity usage. The steady-state power signatures (PS) contain plenty of information needed for detecting state transition and aging of loads. On the other hand, adopting the values of PS directly has the drawbacks of taking a longer time and much memory for the datasets of energy management information system (EMIS). To effectively reduce the number of PS representing load state transition and aging signals, a feature extraction technique of the PS in the EMIS, Hellinger distance, is proposed in this paper. The high success rates of identifying state transition and aging of loads from the back-propagation artificial neural network (BP-ANN) have been proved via experiments to be feasible in load operations of EMIS applications.