1 Aviation Accident Investigation Center of China Academy of Civil Aviation Science and Technology, Beijing 100028, China
2 School of Aeronautical Engineering, Civil Aviation University of China, Tianjin 300300, China
Show less
文章历史+
Received
Published
Revised
2025-07-01
2026-04-18
2025-11-26
Online Date
2026-06-08
PDF (4305K)
摘要
针对小型航空公司安全风险预测中数据分散、动态变化复杂的问题,提出一种基于改进黑翅鸢优化算法(modified black kite algorithm, MBKA)的时间卷积网络-双向长短期记忆网络(temporal convolutional network-bidirectional long short-term memory, TCN-BiLSTM)混合模型。首先,构建动态赋权的民航安全风险指标体系,通过事件等级、事件类型、事件原因和事发阶段的多维度量化映射,解决了传统方法指标冗余、计算复杂度高的问题;其次,提出时序自适应攻击策略、线性锁优策略和可变缩放因子差分进化机制,显著提升了MBKA的全局搜索能力与收敛效率,使其能够高效优化TCN-BiLSTM的超参数。实验结果表明,与传统TCN-BiLSTM模型、基于粒子群优化(particle swarm optimization, PSO)的PSO-TCN-BiLSTM模型以及BKA-TCN-BiLSTM模型相比,MBKA-TCN-BiLSTM模型的平均绝对误差(mean absolute error, MAE)分别降低18.44%、5.81%、3.95%,均方根误差(root mean square error, RMSE)分别降低17.21%、7.77%、8.72%,决定系数 R2分别提升4.59%、1.72%、1.90%。研究结果不仅为小型航空公司提供了轻量化、高精度的风险预测工具,还为复杂时序数据的建模与优化提供了普适性框架,可推广至交通、能源等领域的动态风险预测任务。
Abstract
To address the problem of data dispersion and complex dynamic changes in safety risk prediction for small airlines, a hybrid model of temporal convolutional network bidirectional long short term memory (TCN-BiLSTM) based on the modified black kite algorithm (MBKA)was proposed. Firstly, a dynamic weighting civil aviation safety risk indicator system was constructed, which solved the problems of redundant indicators and high computational complexity in traditional methods through multi-dimensional quantitative mapping of event levels, types, causes, and stages. Secondly, the proposed timing adaptive attack strategy, linear lock optimization strategy, and variable scaling factor differential evolution mechanism significantly enhance the global search capability and convergence efficiency of MBKA, enabling it to efficiently optimize the hyperparameters of TCN-BiLSTM. The experimental results show that compared with traditional TCN-BiLSTM, PSO-TCN-BiLSTM model based on particle swarm optimization (PSO) and BKA-TCN- BiLSTM models, the MBKA-TCN-BiLSTM model reduces mean absolute error (MAE) by 18.44%, 5.81%, and 3.95%, and root mean square error (RMSE) by 17.21%, 7.77%, and 8.72%, respectively. The coefficient of determination (R2) increases by 4.59%, 1.72%, and 1.90%, respectively. The results not only provides lightweight and high-precision risk prediction tools for small airlines, but also provides a universal framework for modeling and optimizing complex time-series data, which can be extended to dynamic risk prediction in fields such as transportation and energy.
WANG Zhan-hai, TIAN Bin, WANG Xiang-jie, WANG Xuan, WU Tao.
Risk Prediction for Small Airlines Based on the MBKA-TCN-BiLSTM Model[J].
Science Technology and Engineering, 2026, 26(11): 4894-4902 DOI:10.12404/j.issn.1671-1815.2504890
针对上述问题,提出一种基于MBKA-TCN-BiLSTM混合模型,该模型融合时间卷积网络(temporal convolutional network,TCN)强大的局部特征提取能力和双向长短期记忆网络(bidirectional long short-term memory,BiLSTM)出色的双向时序建模能力,并借助改进黑翅鸢优化算法(modified black kite algorithm,MBKA)对超参数进行优化,旨在实现对民航安全风险的高精度预测,模型主要框架结构如图3所示,主要包括以下细节。
MengQingsen, HanHao, LiYi. Research on secondary accident prediction of expressway based on Bayesian optimization random forest[J]. Journal of Safety Science and Technology, 2023, 19(7): 205-210.
YuSixuan, WangHuawei. Risk forecasting in general aviation based on sparse de-noisingauto-encoder neural network[J]. Journal of Systems Engineering and Electronics, 2019, 41(1): 112-117.
SunRuishan, ZhangGuanchao, GaoLuping. Improved relation analysis and the triangular fuzzy number prediction model for the civil aviation incidents[J]. Journal of Safety and Environment, 2019, 19(5): 1662-1668.
[7]
袁良运. A航空公司飞行安全风险研究[D]. 成都: 西南交通大学, 2022.
[8]
YuanLiangyun. A study on airline flight safety risk[D]. Chengdu: Southwest Jiaotong University, 2022
XiongHaitao. Research on safety risk management methods for unsafe events in airline operations[J]. Technology Innovation and Application, 2016(25): 262-263.
XuYi, WangHuawei, XiongMinglan. Civil aviation risk evaluation system and forecast model based on PCA-GA-BP[J]. Ship Electronic Engineering, 2021, 41(2): 77-81.
WangYantao, LiJingliang, GuRunping. Flight operation risk prediction model based on the multivariate chaotic time series[J]. Chinese Journal of Engineering, 2020, 42(12): 1664-1673.
ChengJie, ChenDing, LiChun, et al. Ultra-short-term wind power prediction based on GWO-CNN-BiLSTM[J]. Science Technology and Engineering, 2023, 23(35): 15091-15099.
CaiZhiyong, LüHui, ZhouRui, et al. Comprehensive fuzzy assessment on risk and combat plan of forest aviation fire fighting[J]. Journal of Safety Science and Technology, 2022, 18(7): 201-206.
YuanHua, ChenZehao. Short-term traffic flow prediction based on temporal convolutional networks[J]. Journal of South China University of Technology (Natural Science Edition), 2020, 48(11): 107-113, 122.
WangXingwang, ZhangQingyang, JiangShouyong, et al. Dynamic UAV path planning based on improved black winged kite optimization algorithm[J]. Application Research of Computers, 2025, 42(5): 1401-1408.