Residential micro-hub load model using neural network

This paper presents the modeling of a residential micro-hub load based on real measurements and simulation data obtained using the Energy Hub Management System (EHMS) model of a residential load. A neural network (NN) is used to estimate the load model as a function of time, temperature, peak demand, and energy price. Different NN training approaches are compared to determine the best function to be used, based on the available data. Also, the number of hidden layer neurons are varied to obtain the best fit for the NN model. The results show that the proposed NN model is able to properly represent the behavior of an actual residential micro-hub.