Abstract: In order to solve the problems of single forecasting models such as low accuracy and easy falling into local optimization, a hybrid model is presented basing on time series analysis method and SVM (support vector machine), which could improve the predicting accuracy. An improved chaotic time series model is presented.
Chaotic time series model, SVM and hybrid model of time series are used for the modeling and simulation of energy consumption of Tianjin Binhai International Airport. Evaluation of the three models is based on the estimation of average behavior of mean squared error. Experimental results show that the hybrid model is an effective way to improve the forecasting accuracy achieved by any one of the models separately.

Key words: time series, SVM, hybrid prediction model, airport energy prediction