基于T-S模糊神经网络的飞行学员飞行技能评价模型构建研究  

Construction of Flight Skills Evaluation Model for Flight Cadets Based on T⁃S Fuzzy Neural Network

作　　者：李根 汪海波[1] 司海青[1] 潘亭 刘海波 LI Gen;WANG Haibo;SI Haiqing;PAN Ting;LIU Haibo(College of General Aviation and Flight,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学通用航空与飞行学院,南京210016

出　　处：《载人航天》2023年第5期616-623,共8页Manned Spaceflight

基　　金：国家自然科学基金委民航联合基金重点项目(U2033202);中央高校基本科研业务费专项资金资助(NS2022094);江苏省高等教育教改研究课题(2021JSJG206);南京航空航天大学实验技术研究与开发项目(SYJS202207Y)。

摘　　要：为准确评价飞行学员飞行技能的优劣,在分析飞行学员起落航线各飞行阶段任务的基础上,参照飞行训练手册并结合与教员的访谈,建立了飞行学员飞行技能评价指标体系。运用T-S模糊神经网络搭建的飞行技能评价模型实现对飞行学员飞行技能评价,采集118名飞行学员飞行数据(有效数据110组),80组数据用于训练模型,30组数据用于测试,以验证模型评价的适用性和精确度。结果表明:T-S模糊神经网络具有很好的学习效率,评价飞行学员飞行技能准确度为976%,该方法构建出的评价模型应用于飞行学员飞行技能评价有效可行。To better evaluate the flight skills of flight cadets,based on the tasks of flight cadets in each flight phase of the takeoff and landing route,an evaluation index system of flight skills of flight cadets was constructed by referring to the flight training manual and interviewing with the instructor.The flight skill evaluation model built by T-S fuzzy neural network could realize the evaluation of the flight cadets’flight skills.A total of 118 flight cadets’flight data were collected,with 110 sets of valid data.80 sets of data were used for model training,and 30 sets of data were used for testing to validate the applicability and accuracy of the model evaluation.The results showed that T-S fuzzy neural network was equipped with good learning efficiency.The evaluation precision reached 976%.The evaluation model constructed by this method was effective and feasible for the evalua-tion of flight cadets’flight skills.

关 键 词：T-S模糊神经网络 起落航线 飞行数据 飞行技能评价 

分 类 号：X949[环境科学与工程—安全科学] V19[航空宇航科学与技术—人机与环境工程]