基于神经网络的飞机襟翼机构磨损失效判据确定方法  

The Method to Determinethe Failure Criterion of Aircraft Flap Mechanism's Wear Based on the Neural Networks

作　　者：张割[1] 吕德峰[2] 

机构地区：[1]北京航空技术研究中心 [2]中国人民解放军94637部队

出　　处：《飞机设计》2011年第5期36-39,63,共5页Aircraft Design

摘　　要：飞机襟翼使用过程中存在失效判据不易确定的问题,目前襟翼机构滑轨滑轮磨损间隙许用量主要是通过试验及使用统计数据分析确定。针对新研机构缺乏试验及使用数据的问题,通过对影响磨损间隙最大许用量的因素分析,提出应用神经网络学习确定襟翼滑轨滑轮间隙最大许用量的预测方法。算例验证表明,运用神经网络方法预测得到的滑轨滑轮架的间隙最大许用量误差小于10%,说明方法有效。该方法对当前新研机构的失效判据确定具有重要的工程意义。In the operational process of the aircraft flap it is not easy to determine the failure criterion. The largest clearance wear loss allowed, that is, the clearance threshold determination is difficult but according to the experimental and statistical data. The new mechanism has no statistical data and the experimental data is a little. Through the analysis of influencing factors to the largest clearance wear loss allowed, the Neural Networks （NN） learning algorithm was proposed. And the examples authentication＇s results showed that the prediction error of the maximum allowable amount of the gap between the slide and the pulley＇s shelf is less thanl0%, the method was feasible and accurate for the prediction of wear loss threshold. The method to threshold determinations has actual significance to new mechanisms.

关 键 词：飞机 襟翼 机构 神经网络 失效判据 

分 类 号：E237[军事—军事理论]