基于Bi IndRNN和PSO的航班延误预测

Flight Delay Prediction Based on Bi IndRNN and PSO

针对普通循环神经网络在航班延误预测问题上精度不高、调试时间长的问题,提出基于Bi IndRNN和粒子群的机场短期航班延误预测模型。模型采用Bi IndRNN结构作为预测模型,使用粒子群算法对模型超参数进行全局寻优,使模型不仅能够处理长序列的数据,还能够高效率地选择合适的参数,使其预测准确度更高。实验在2018年国内某机场航班延误数据集上构建预测模型,并对所提模型预测结果与其他预测模型预测结果进行比较分析。结果表明,提出的预测模型提高了航班延误预测的准确率。

Aiming at the low accuracy and long debugging time of ordinary recurrent neural network in flight delay prediction,an airport short term flight delay prediction model based on Bi IndRNN and particle swarm is proposed.The model adopts the Bi IndRNN structure as the prediction model,and uses the particle swarm algorithm to optimize the hyperparameters of the model globally,so that the model can not only process long sequence data,but also select appropriate parameters efficiently.In the experiment,a prediction model was constructed on the flight delay data set of a domestic airport in 2018,and the prediction results of the proposed model were compared and analyzed with those of other prediction models.The results show that the prediction model proposed in this study improves the accuracy of flight delay prediction.