| 443 | 1 | 98 |
| 下载次数 | 被引频次 | 阅读次数 |
针对当前单一模型预测航空器滑出时间精度提升存在瓶颈的问题，提出一种结合最大互信息系数（Maximal Information Coefficient, MIC）的迭代注意力特征融合模块（iterative Attentional Feature Fusion, iAFF）和Stacking集成学习框架组合的航空器滑出时间预测模型。首先利用MIC提取出与滑出时间相关性较高的因素作为模型原始特征序列；然后以支持向量回归（SVR）、随机森林（RF）、多层感知机（MLP）和极限梯度提升机（XGBoost）为基学习器模型对原始特征进行特征构造，并利用iAFF模块对基学习器得到的构造特征和原始特征进行特征融合，通过MLP对融合后的特征进行学习，最终得到预测滑出时间。经实际算例对比验证表明，与单一模型相比，MIC-iAFF-Stacking集成学习模型在±2、±3、±5 min误差范围内的预测精度分别提升了6.14%、6.40%、2.31%，证明了该模型在滑出时间预测中的有效性。
Abstract：An aircraft departure taxi time prediction model, which combines maximal information coefficient(MIC), iterative attentional feature fusion(iAFF), and Stacking ensemble learning framework, was developed to improve the prediction accuracy of a single model for aircraft taxi-out time.First, MIC was used to extract the factors that highly correlate with taxi-out time to serve as the original feature sequence of the model. Subsequently, support vector regression(SVR), random forest(RF), multilayer perceptron(MLP), and extreme gradient boosting machine(XGBoost) were used as the components of the base learner model to construct the original features. The iAFF module was used to fuse the original and structural features obtained by the base learner model, and the fused features were learned using the MLP to finally predict the taxi-out time. The comparison and verification of actual examples show that the prediction accuracy of MIC-iAFF-Stacking ensemble learning model was higher than that of a single model by 6.14%, 6.40% and 2.31% in the error range of ±2,±3, and ±5 min, respectively, which proves the effectiveness of the model in predicting the taxi-out time.
[1]魏云,王莉莉.多跑道机场离场航班推出时刻的优化研究[J].交通运输工程与信息学报, 2015, 13(2):70-73, 79.WEI Yun, WANG Lili. Research on the optimization of launch time of departure flights of multi-runway airport[J]. Journal of Transportation Engineering and Information, 2015, 13(2):70-73, 79.
[2] CLEWLOW R, SIMAIAKIS I, BALAKRISHNAN H.Impact of arrivals on departure taxi operations at airports[C]//AIAA Guidance, Navigation, and Control Conference. Toronto:AIAA, 2010:AIAA2010-7698.
[3] RAVIZZA S, ATKIN J A D, MAATHUIS M H, et al. A combined statistical approach and ground movement model for improving taxi time estimations at airports[J].Journal of the Operational Research Society, 2013, 64(9):1347-1360.
[4]冯霞,孟金双.基于KNN和SVR的航班滑出时间预测[J].西南交通大学学报, 2017, 52(5):1008-1014.FENG Xia, MENG Jinshuang. Flight taxi-out time prediction based on KNN and SVR[J]. Journal of Southwest Jiaotong University, 2017, 52(5):1008-1014.
[5] ZHANG Y, WANG Q. Methods for determining unimpeded aircraft taxiing time and evaluating airport taxiing performance[J]. Chinese Journal of Aeronautics, 2017, 30(2):523-537.
[6]刘继新,尹旻嘉,朱学华,等.基于航空器滑出时间的离场影响因素研究[J].武汉理工大学学报(交通科学与工程版), 2018, 42(2):195-200.LIU Jixin, YIN Minjia, ZHU Xuehua, et al. Research on departure influencing factors based on aircraft taxi-out time[J]. Journal of Wuhan University of Technology(Transportation Science&Engineering), 2018, 42(2):195-200.
[7]李楠,焦庆宇,张连东,等.离场航空器滑行时间预测研究[J].重庆交通大学学报(自然科学版), 2021, 40(3):1-6.LI Nan, JIAO Qingyu, ZHANG Liandong, et al. Taxi time prediction of departure aircraft[J]. Journal of Chongqing Jiaotong University(Natural Science), 2021,40(3):1-6.
[8] WANG X, BROWNLEE A E I, WOODWARD J R, et al.Aircraft taxi time prediction:Feature importance and their implications[J]. Transportation Research Part C:Emerging Technologies, 2021, 124:102892.
[9]唐小卫,陈祯,张生润,等.繁忙机场机坪空间构型对航班离港滑行时间的影响[J].交通运输系统工程与信息,2022, 22(5):309-317.TANG Xiaowei, CHEN Zhen, ZHANG Shengrun, et al.Impact of apron spatial configuration on flight departure taxi time at busy airports[J]. Journal of Transportation Systems Engineering and Information Technology, 2022,22(5):309-317.
[10]唐小卫,丁叶,张生润,等.进港航班滑入时间预测研究[J].北京航空航天大学学报, 2024, 50(7):2018-2224.TANG Xiaowei, DING Ye, ZHANG Shengrun, et al.Taxi-in time prediction of arrival flight[J]. Journal of Beijing University of Aeronautics and Astronautics,2024, 50(7):2018-2224.
[11] LIAN G, ZHANG Y, XING Z, et al. RETRACTED:a new dynamic pushback control method for reducing fuel-burn costs:using predicted taxi-out time[J]. Chinese Journal of Aeronautics, 2019, 32(3):660-673.
[12] LI Nan, JIAO Qingyu, ZHU Xinhua, et al. Prediction of departure aircraft taxi time based ondeep learning[J].Transactions of Nanjing University of Aeronautics and Astronautics, 2020, 37(2):232-241.
[13]刘家学,白明皓,郝磊.基于ARIMA-SVR组合方法的航班滑出时间预测[J].中国科技论文, 2021, 16(6):661-667.LIU Jiaxue, BAI Minghao, HAO Lei. Prediction of flight taxi-out time based on ARIMA-SVR combination method[J]. China Sciencepaper, 2021, 16(6):661-667.
[14]黄龙杨,夏正洪.离港航班可变滑出时间预测方法及应用[J].科学技术与工程, 2021, 21(33):14434-14439.HUANG Longyang, XIA Zhenghong. Prediction method of departure flight estimated taxi-out time and its application[J]. Science Technology and Engineering, 2021,21(33):14434-14439.
[15] WANG F, BI J, XIE D, et al. A data-driven prediction model for aircraft taxi time by considering time series about gate and real-time factors[J]. Transportmetrica A:Transport Science,2023,19(3):2071353.
[16]杜婧涵,胡明华,尹嘉男,等.基于强化学习方法的航班滑出时间预测研究[J].航空计算技术, 2022, 52(6):26-29, 34.DU Jinghan, HU Minghua, YIN Jianan, et al. Research on flight taxi-out time prediction based on reinforcement learning method[J]. Aeronautical Computing Technique,2022, 52(6):26-29, 34.
[17]赵征,冯事成,宋梅雯,等.基于XGBoost的航空器动态滑行时间预测方法研究[J].航空工程进展, 2022, 13(1):76-85.ZHAO Zheng, FENG Shicheng, SONG Meiwen, et al.Prediction method of aircraft dynamic taxi time based on XGBoost[J]. Advances in Aeronautical Science and Engineering, 2022, 13(1):76-85.
[18] DU J, HU M, ZHANG W, et al. Finding similar historical scenarios for better understanding aircraft taxi time:a deep metric learning approach[J]. IEEE Intelligent Transportation Systems Magazine, 2023, 15(1):101-116.
[19]章月,周洁敏.基于特征选择的BP神经网络算法滑行时间预测[J].航空计算技术, 2024, 54(1):71-75.ZHANG Yue, ZHOU Jiemin. Taxiing time prediction based on the feature selection of BP neural network algorithm[J]. Aeronautical Computing Technique, 2024, 54(1):71-75.
[20]史佳琪,张建华.基于多模型融合Stacking集成学习方式的负荷预测方法[J].中国电机工程学报, 2019, 39(14):4032-4042.SHI Jiaqi, ZHANG Jianhua. Load forecasting based on multi-model by stacking ensemble learning[J]. Proceedings of the CSEE, 2019, 39(14):4032-4042.
[21] RESHEF D N, RESHEF Y A, FINUCANE H K, et al.Detecting novel associations in large data sets[J]. Science, 2011,334(6062):1518-1524.
[22] DAI Y, GIESEKE F, OEHMCKE S, et al. Attentional feature fusion[C]//IEEE Winter Conference on Applications of Computer Vision(WACV). Waikoloa:IEEE,2021:3559-3568.
[23] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Las Vegas:IEEE, 2016:770-778.
[24] PROKHORENKOVA L, GUSEV G, VOROBEV A, et al. CatBoost:unbiased boosting with categorical features[EB/OL].(2019-01-20)[2024-01-09]. https://arxiv. org/abs/1706.09516.
[25] KE G, MENG Q, FINLEY T, et al. LightGBM:a highly efficient gradient boosting decision tree[C]//31st Conference on Neural Information Processing Systems(NIPS 2017), Long Beach:Curran Associates Inc.,2017:3149-3157.
基本信息:
DOI：10.19961/j.cnki.1672-4747.2024.04.009
中图分类号:V355
引用信息:
[1]李浩,卢朝阳,谈翌平等.基于MIC-iAFF-Stacking集成学习的航空器滑出时间预测[J].交通运输工程与信息学报,2024,22(04):142-153.DOI:10.19961/j.cnki.1672-4747.2024.04.009.
基金信息:
国家自然科学基金民航联合基金重点项目（U2033203）