测控技术2011，Vol.30Issue(1)：112-116,5.
基于混合粒子群算法的小波神经网络故障诊断
Faults Diagnosis Using Wavelet Neural Networks Based on Hybrid Particle Swarm Algorithm
宋玉琴 1章卫国1
作者信息
- 1. 西北工业大学,自动化学院,陕西,西安,710072
- 折叠
摘要
Abstract
Aimed at complex sensors faults of aircraft control system, the fault diagnosis models are set up and various fault data are acquired. Three wavelet neural networks are constructed. An improved particle swarm algorithm called hybrid particle swarm algorithm optimization is proposed. The wavelet neural networks are trained by this optimization. The wavelet neural networks structures are simplified by discrete particle swarm algorithm. The wavelet neural networks weighs are optimized by based particle swarm algorithm. This method is used in sensors faults diagnosis of flight control system. The experimental results show that the method identifies all sensors faults of aircraft control system effectively and eliminates the influences on the fault diagnosis capabilities because of the redundant structures in the wavelet neural networks.关键词
小波/神经网络/粒子群/故障诊断Key words
wavelet/ neural networks/ particle swarm/ fault diagnosis分类
信息技术与安全科学引用本文复制引用
宋玉琴,章卫国..基于混合粒子群算法的小波神经网络故障诊断[J].测控技术,2011,30(1):112-116,5.
