The Research Of The Prediction Model Of Airport Noise Based On Machine Learning

With the rapid development of civil aviation of china in recent years, the throughput and the scale of the airport is constantly expanding, the problem of airport noise has become increasingly serious, the affected area has become more and more large, it is becoming one of the obstacles affecting the sustainable development of civil aviation. An important issue currently faced is how to control the airport noise effectively. Because the prediction of the airport noise is the most important basis for the airport noise control measures, the accuracy of the prediction is very crucial.Effective prediction model is the key to ensure the accuracy of the prediction.Through the analysis of the influence factors of single aircraft noise event, the regression prediction model based on BP neural network is established. Then, the prediction model of neural network ensemble for single aircraft noise event is constructed by selecting neural networks with the aid of adaptive genetic algorithm.Simultaneously, in order to maintain the diversity of neural networks, different number of hidden units and Bagging algorithm are used. Experimental results show that the prediction model of neural network ensemble is better than the model of single BP neural network and it has better generalization ability and higher stability.According to the characteristic of cumulative noise events, the prediction model of SVM is constructed for cumulative noise events. Then, an improved particle swarm optimization algorithm is adopted to optimize the parameters of SVM. Experimental results show the effectiveness of the prediction model based on SVM of cumulative noise events.