摘要
针对飞机操纵面故障趋势预测问题,结合系统可测状态参量能间接表现操纵面故障情况的特点,提出一种基于多因子高阶模糊变动的方法。将传统的模糊时间序列预测方法进行扩展,利用多元时间序列的变动值构建多辅助因子模糊逻辑关系。采用自组织映射(SOM)方法将整个论域划分为不等长度的多个论域区间,并重新设定所属论域区间的隶属度。根据时间序列的周期性特点,建立多因子高阶模糊变动预测模型。为了验证算法的有效性,针对飞机转弯时左副翼损伤故障趋势进行预测和分析,并与传统模糊时间序列预测方法进行对比,用两种方法得到的预测结果通过建立好的故障映射模型进行故障类型和故障程度的判别,仿真结果充分表明了多因子高阶模糊变动预测模型具有更好的故障趋势预测能力。
The trends of aircraft control system state parameters which can be measured are indirect manifestations of surface damage.In order to improve prediction accuracy,a method based on multi-factor high-order fuzzy variation is proposed.The proposed method constructs multi-cofactor fuzzy logical relationship with the variation of multivariate time series and extends the conventional fuzzy prediction method.The universe of discourse is divided into unequal interval length using self-organizing map(SOM) and its membership is reset.According to the periodic feature,a multi-factor high-order fuzzy variation model is built.In order to verify the validity of the method,the prediction and analysis of aileron fault trend was performed.The identification of fault type and fault degree is obtained by fault mapping model with the prediction results.Compared with the traditional method,the simulation result demonstrates the proposed prediction model has a better predictive ability.
出处
《中南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2013年第S1期232-237,共6页
Journal of Central South University:Science and Technology
基金
国家自然科学基金资助项目(60874117
61101004)
关键词
模糊时间序列
故障预测
自组织映射
区间长度
操纵面
fuzzy time series
fault prediction
self-organizing map
interval length
control surface
