Study Of The Energy Consumption Prediction Method Of Airport

With the rapid development of civil aviation and the national economy level,more and more people chose aircraft as their first choice when travelling.The trouble with this trend is large-scale airport expansion and a lot of new airport put into operation.During the process of airport construction and renovation,the designers usually only consider to meet the standard requirement and make the energy saving calculation at the end of the process.It makes the energy saving calculation meaningless.So the monitoring of airport operating energy consumption has become the most important part of the energy saving problem.Due to the application time of energy consumption prediction method used in the energy saving design in not long,there are still a lot of problems in the specific scene.It needs a standardization simulation method.Three widely used prediction methods are took out after the research at home and abroad:Support vector machine,the time series and the BP neural network.A detailed algorithm of these three methods is introduced and after that the improvement of these three methods used in airport energy consumption prediction is put out.Firstly,SVM shows out a strong advantage in the process of energy consumption prediction for monthly settlement data cause it can adapt to small sample data.In order to solve the problems of single forecasting model,such as low accuracy and easily falling into local optimization,this paper presented a hybrid model based on the time series analysis method and the support vector machine(SVM),which could improve the accuracy of prediction.An improved chaotic time series model had been presented.Three methods chaotic time series model?SVM and hybrid model of time series were used to the modeling and simulation.The evaluation of the three models is based on the estimation of the average behavior of the mean squared error.At last these three models were used to the modeling and simulation of the energy consumption of the Tianjin Binhai International Airport by Matlab.The experimental results show that the hybrid model is an effective way to improve the forecasting accuracy achieved by any one of the models separately.At last an abnormalenergy consumption model based on BP neural network is build.Simulation results show that this model can effectively found the energy consumption abnormal points.An effective energy consumption prediction model and an abnormal energy consumption model have been built through these work and they will better serve the energy consumption prediction work and make an important valuable benefit to the airport operations.