Use of neural networks for modeling energy consumption in the residential sector

This chapter investigates the use of neural networks (NN) for modeling of residential energy consumption. Currently, engineering and conditional demand analysis (CDA) approaches are mainly used for residential energy modeling. The studies on the use of NN for residential energy consumption modeling are limited to estimating the energy use of individual or a group of buildings. Development of a national residential end-use energy consumption model using NN approach is presented in this chapter. The comparative evaluation of the results of the model shows NN approach can be used to accurately predict and categorize the energy consumption in the residential sector as well as the other two approaches. Based on the specific advantages and disadvantages of three models, developing a hybrid model consisting of NN and engineering models is suggested.