基于GA-BP的水电机组轴心轨迹识别的研究  

Identification of shaft centerline orbit of hydropower units based on neural network improved by genetic algorithm

作　　者：贾嵘[1] 张佩刚[1] 张欣伟[1] 武桦[1] 路瑶[1] 

机构地区：[1]西安理工大学水利水电学院,西安710048

出　　处：《水力发电学报》2013年第2期277-281,共5页Journal of Hydroelectric Engineering

基　　金：国家自然科学基金项目资助(51279161)

摘　　要：在水电机组故障诊断中,轴心轨迹识别是一种重要的方法。本文应用遗传算法优化BP神经网络的权值和阈值以构建诊断模型,利用离散Hu不变矩提取轴心轨迹的特征向量,并将其作为神经网络的特征参数,然后利用该模型对几种典型的轴心轨迹进行了识别。仿真结果表明这种方法识别精度高、速度快,具有较高的实用价值。Identification of shaft centerline orbit is a key measure in fault diagnosis of hydro units.To develop a diagnosis model,this study adopted a genetic algorithm(GA) to improve the weights and threshold of a BP neural network,and used discrete affine invariant moments to calculate the characters of the orbit.With these orbit data inputted into the neural network as its characteristic parameters,several typical orbits were identified.Results show that this identification method is accurate and fast with a highly practical value.

关 键 词：水电机组 轴心轨迹 HU不变矩 遗传算法 BP神经网络 

分 类 号：TM312[电气工程—电机]