摘要
故障诊断对于事故后快速恢复具有重要意义。针对飞机系统的故障诊断研究,本文提出一种基于改进遗传算法优化的BP神经网络智能诊断技术。采用基于实数编码的遗传算法优化神经网路的权值和阀值,克服了BP神经网络收敛速度慢、易陷入局部极小的缺点,并通过Matlab仿真,用实例验证了经过优化的BP神经网络的训练步数得到大大减少,准确性有所提高,泛化能力也得到提升,增强了网络的学习能力,具有一定的实用性。
Fault diagnosis is of great importance to quick recovery. For aircraft system fault diagnosis,this paper presents an intelligent diagnosis method based on genetic algorithm （GA） optimized BP Neural Networks. The method is based on real-coded genetic algorithm optimization neural network and the weight threshold to overcome the BP neural network convergence slow and easy to fall into the shortcomings of the local minimum.Through Matlab simulation, using the example of the optimized BP neural network .Training steps have been greatly reduced, ability of generalization has also been improved ,and enhanced the learning ability of the network, so it has practical use to a certain extent.
出处
《微计算机信息》
2009年第19期110-112,共3页
Control & Automation
关键词
飞机故障诊断
遗传算法
BP网络
civil aviation aircraft fault diagnosis
genetic algorithm
BP net
