Hybrid support vector machine and ARIMA model in building cooling prediction

Accurate building cooling load forecasting is the precondition for the optimal control and energy saving operation of HVAC systems. Many forecasting approaches such as artificial neural network (ANN), support vector machine (SVM), autoregressive integrated moving average (ARIMA) and grey model, have been proposed in the field of building cooling load prediction. However, none of them has enough accuracy to satisfy the practical demand. Therefore, a novel hybrid predictor integrating seasonal ARIMA and Support Vector Regression (SVR) is presented to forecast building cooling load. SARIMA is suitable for linear prediction and SVR is suitable for nonlinear prediction. Experimental results demonstrate that the proposed hybrid building cooling load forecasting model can be an effective way to improve forecasting accuracy achieved by either of the models used separately.