College of Air Traffic Management, Civil Aviation University of China, Tianjin 300300, China
Show less
文章历史+
Received
Published
Revised
2025-07-18
2026-05-28
2025-11-28
Online Date
2026-08-07
PDF (13621K)
摘要
针对基于飞机性能数据库(base of aircraft data, BADA)模型的航空器油耗模型精度较低、反向传播神经网络(back propagation neural network,BPNN)等机器学习模型可解释性不强的问题。基于多元自适应回归样条(multivariate adaptive regression splines, MARS)模型对航空器燃油流量进行建模与分析,旨在构建B737-800飞机在巡航阶段的油耗模型并分析飞行参数对燃油流量的非线性影响机制。通过对比MARS模型与随机森林回归(random forest regression, RFR)、BP神经网络模型以及多元线性回归(multiple linear regression, MLR)三种经典回归方法的拟合效果,结果显示MARS模型在训练集与测试集上的精度均较高,且具有良好的可解释性。在变量重要性分析中,飞行重量被识别为对燃油流量影响最显著的因素,其次为大气温度。
Abstract
In order to overcome the limitations of BADA-based fuel consumption models in terms of low prediction accuracy, and to address the lack of interpretability in machine learning approaches such as back propagation Neural Network (BPNN), an aircraft fuel consumption model based on the multivariate adaptive regression splines (MARS) method was developed and analyzed. The model was applied to the B737-800 aircraft during the cruise phase to investigate the nonlinear relationships between fuel flow and relevant flight parameters. Comparative analyses were conducted with three classical regression models—random forest regression (RFR), BPNN, and multiple linear regression (MLR). The results show that the MARS model achieves superior accuracy in both training and testing datasets, satisfying the requirements for predictive performance. Moreover, the MARS model provides high interpretability. The variable importance analysis indicates that aircraft weight is the most influential factor affecting fuel consumption, followed by ambient temperature. It is concluded that the MARS-based modeling approach effectively balances prediction accuracy and model interpretability, offering a promising tool for aircraft performance analysis and fuel optimization.
LIU Fei, HE Mu, WEI Zhi-qiang.
Fuel Consumption Analysis of Civil Aircraft during Cruise Phase Based on Multivariate Adaptive Regression Splines[J].
Science Technology and Engineering, 2026, 26(15): 6624-6632 DOI:10.12404/j.issn.1671-1815.2505340
NuicA. User manual for the base of aircraft data (BADA) revision 3. 10[J]. Atmosphere, 2010, 12(4): 10.
[2]
VillénB J. Improving the BADA 3 aerodynamic database for trajectory optimization[D]. Barcelona: Universitat Politècnica de Catalunya, 2015.
[3]
UzunM, DemirezenM U, InalhanG. Physics guided deep learning for data-driven aircraft fuel consumption modeling[J]. Aerospace, 2021, 8(2): 44.
[4]
TurgutE T, RosenM A. Relationship between fuel consumption and altitude for commercial aircraft during descent: preliminary assessment with a genetic algorithm[J]. Aerospace Science and Technology, 2012, 17(1): 65-73.
[5]
BaklaciogluT. Modeling the fuel flow-rate of transport aircraft during flight phases using genetic algorithm-optimized neural networks[J]. Aerospace Science and Technology, 2016, 49: 52-62.
[6]
KhanW A, MaH L, OuyangX, et al. Prediction of aircraft trajectory and the associated fuel consumption using covariance bidirectional extreme learning machines[J]. Transportation Research Part E: Logistics and Transportation Review, 2021, 145: 102189.
[7]
BaumannS, NeidhardtT, KlingaufU. Evaluation of the aircraft fuel economy using advanced statistics and machine learning[J]. CEAS Aeronautical Journal, 2021, 12(3): 669-681.
[8]
ZhuQ, PeiJ, LiuX, et al. Analyzing commercial aircraft fuel consumption during descent: a case study using an improved K-means clustering algorithm[J]. Journal of Cleaner Production, 2019, 223: 869-882.
XuDonglei, DingDongjin, XiaoGang, et al. Modeling and fuel consumption analysis of continuous climb operations based on aircraft performance[J]. Civil Aircraft Design and Research, 2022(4): 15-21.
ChenJingjie, CuiJincheng. Short-haul fuel consumption estimation based on different sampling rates[J]. Science Technology and Engineering, 2019, 19(24): 254-259.
ChenJingjie, LiangGuodong, LiuJiaxue, et al. Interval estimation of segment fuel consumption based on irregularly distributed samples[J]. Science Technology and Engineering, 2022, 22(14): 5935-5941.
LiuZhi, ChenJingjie. A piecewise linear estimation model of flight-segment fuel consumption based on statistically analyzed thresholds[J]. Science Technology and Engineering, 2022, 22(23): 10341-10346.
GuRunping, LaiJinghan, WeiZhiqiang. Prediction method for remaining fuel upon landing based on improved BP neural network[J]. Flight Dynamics, 2020, 38(4): 76-80, 86.
ChenMingqiang, ZhengWenhao, SunYanjun, et al. Fuel consumption prediction for general aviation aircraft based on regression tree and random forest[J]. Science Technology and Engineering, 2025, 25(7): 3026-3034.
ZhuYifei, SunYuan, ZhenZiyang, et al. Vertical profile optimization of civil aircraft climb segment based on an improved NSGA-II[J]. Flight Dynamics, 2025, 43(5): 88-94.
[27]
FriedmanJ H. Multivariate adaptive regression splines[J]. The Annals of Statistics, 1991, 19(1): 1-67.
[28]
HastieT, TibshiraniR, WainwrightM. Statistical learning with sparsity[M]. London: Chapman and Hall/CRC, 2015.
[29]
TatacharA V. Comparative assessment of regression models based on model evaluation metrics[J]. International Research Journal of Engineering and Technology (IRJET), 2021, 8(9): 2395-0056.
[30]
BreimanL. Random forests[J]. Machine Learning, 2001, 45: 5-32.
[31]
HanJ, PeiJ, TongH. Data mining: concepts and techniques[M]. San Mateo: Morgan Kaufmann, 2022.
[32]
RumelhartD E, HintonG E, WilliamsR J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088): 533-536.
[33]
LiL, YuanS, TengY, et al. A study on sustainable consumption of fuel—an estimation method of aircraft[J]. Energies, 2021, 14(22): 7559.