基于混沌神经网络的旋转机械状态趋势预测  

Condition development prediction of rotating machine based on chaos-neural networks

作　　者：朱春梅[1,2] 徐小力[1,2] 张建民[2] 

机构地区：[1]北京信息科技大学机电工程学院,北京100192 [2]北京理工大学机械与车辆工程学院,北京100081

出　　处：《北京机械工业学院学报》2009年第1期8-11,共4页Journal of Beijing Institute of Machinery

基　　金：国家自然科学基金(50375017);北京市自然科学基金(3062008)

摘　　要：旋转机械的特征信号常具有非线性和非平稳性,传统的平稳信号分析方法对这类信号不适用,将混沌理论引入旋转机械的故障预测领域,详述了混沌预测方法和混沌神经网络的预测原理,并以工业现场大型烟气轮机为研究对象,完成了基于混沌神经网络的预测,与灰色预测方法进行了比较,实验结果表明基于混沌神经网络的预测精度更高,更有效。Condition development prediction is indispensable to the safe operation of rotating machines. The typical signal of rotating machine is nolinear and non-stationary. So the traditional signal analysis methods are not suitable for these signals, introducing chaos theory into condition development prediction of rotating machine and aiming at the industrial smokes and gas turbine, the method of chaos forecasting and the forecasting theory based on chaos-neural networks are elaborated. The paper finishes a prediction based on the chaos-neural networks and compares it with the gray predicting method. The result shows that the prediction based on the chaos-neural networks has a higher accuracy and effectiveness.

关 键 词：旋转机械 状态趋势预测 混沌 相空间重构 混沌神经网络 

分 类 号：TH206.3[机械工程—机械制造及自动化]