模糊神经网络的侧滑角校正  

Calibration of angle of sideslip using fuzzy neural network

作　　者：孙绍山[1] 杨朝旭[1] 杨伟[1] 

机构地区：[1]成都飞机设计研究所飞控研究室,四川成都610041

出　　处：《飞行力学》2001年第2期52-56,共5页Flight Dynamics

摘　　要：叙述了一种利用模糊神经网络模型来校正侧滑角的方法。侧滑角的校正基于以下七个输入变量 :动静压之比、动压、迎角、滚转速率、偏航速率、飞机重心处侧向过载、基准侧滑角。在模糊神经网络模型的结构辨识中 ,采用了前向搜寻算法。样本数据的一部分用于训练模糊神经网络 ,另一部分用于测试和评价所得模型的性能 ;参数辨识中 ,采用了反向传播学习算法 (即 BP算法 )。某型飞机的六自由度仿真软件仿真验证表明 ,用这种方法校正后的侧滑角具有较高的精确度。This paper describes a method of angle of sideslip calibration using a fuzzy neural network model. The followed seven input variables are used for the calibration of angle of sideslip: ratio of impact pressure to static pressure ( q p ), impact pressure ( q c ), angle of attack ( α ), roll rate ( p ), yaw rate ( r ), lateral acceleration at the center of aircraft gravity ( n y ), basic angle of sideslip ( β bas ). A search forward algorithm has been employed for the structure identification of the fuzzy neural network model, part of the sample data are used to train the fuzzy neural network, the other for testing and evaluating the performance of the models; BP (back propagation) algorithm has been employed for the parameter identification of the fuzzy neural network. The six degrees of freedom simulation result of civil certain model aircraft is shown that the angle of sideslip calibrated by this method has good precision.

关 键 词：侧滑角 模糊神经网络模型 反向传播学习算法 校正 前向搜寻算法 模型性能 BP算法 精确度 战斗机 仿真验证 输入变量 横航稳定性 

分 类 号：V249.122.4[航空宇航科学与技术—飞行器设计] TP183[自动化与计算机技术—控制理论与控制工程]