基于EEMD和BiLSTM的航班准点率预测研究

Prediction of Flight Punctuality Based on EEMD and BiLSTM

航班准点率问题是民航业最为关心的问题之一,准确地预测出航班的准点率能够有效降低航班延误所带来的不利影响、提高乘客满意度。为解决普通深度学习预测模型存在的航班准点率数据挖掘程度不足、预测准确度较低的问题,提出一种基于集合经验模态分解法(EEMD)和双向长短时记忆神经网络(BiLSTM)的机场短期航班准点率预测模型。模型使用EEMD算法将准点率时间序列进行分解,采用BiLSTM结构作为预测模型,使模型能够更深层、高效地处理航班准点率数据,提高预测准确度。实验数据为2018年上海虹桥机场航班准点率及天气数据,实验建立了多个参照模型与所提模型进行对比分析。结果表明:提出的EEMD-BiLSTM模型相较于一般模型预测误差更小,准确度更高。

Flight punctuality is one of the most concerned issues in the civil aviation industry.Accurate prediction of flight punctuality can effectively reduce the adverse impact caused by flight delay and improve passenger satisfaction.In order to solve the problems of insufficient data mining and low prediction accuracy of flight punctuality in common deep learning prediction models,this paper proposes a short term airport flight punctuality prediction model based on Ensemble Empirical Mode Decomposition(EEMD)and Bi directional Long Short Term Memory neural network(BiLSTM).The model uses EEMD algorithm to decompose the punctuality time series,and adopts BiLSTM structure as the prediction model,so that the model can process the flight punctuality data more deeply and efficiently,and improve the prediction accuracy.The experimental data are the flight punctuality rate and weather data of Shanghai Hongqiao Airport in 2018.Several reference models are established in the experiment for comparative analysis with the model proposed in this paper.The results show that the prediction error of the proposed EEMD BiLSTM model is smaller and the accuracy is higher than that of the general model.