Soft computing methodologies for estimation of energy consumption in buildings with different envelope parameters

In this study, soft computing methods are designed and adapted to estimate energy consumption of the building according to main building envelope parameters such as material thicknesses and insulation K-value. In order to predict the building energy consumption, novel intelligent soft computing schemes, support vector regression (SVR), and adaptive neuro-fuzzy inference system (ANFIS) are used. The polynomial, linear, and radial basis function (RBF) is applied as the kernel function of the SVR to estimate the optimal energy consumption of buildings. The performance of proposed optimizers is confirmed by simulation results. The SVR results are compared with the ANFIS, artificial neural network (ANN), and genetic programming (GP) results. The computational results show that an improvement in predictive accuracy and capability of generalization can be achieved by the ANFIS approach in comparison to the SVR estimation. Based on the simulation results, the effectiveness of the proposed optimization strategies is verified. The data used in soft computing were obtained from 180 simulations in EnergyPlus for variations of building envelope parameters.