欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出了一种基于总体平均经验模态分解和GRNN神经网络的滚动轴承故障诊断方法。首先通过EEMD方法将非平稳、非线性的滚动轴承振动信号分解为若干个平稳的固有模态函数（IMF）之和,提取前8个IMF分量作为频域特征,同其他14个时频域特征指标组成特征集输入到GRNN神经网络中,建立起GRNN网络模型,对滚动轴承三种故障状态进行模式识别。通过分析比较BP和GRNN两种网络模型对故障的诊断结果,验证了GRNN网络的优越性和可行性。
英文摘要：
Propose a fault diagnosis methods for rolling bearing based on an ensemble empirical mode decomposition and GRNN neural network. First of all,through the EEMD method,non- stationary and nonlinear properties of rolling bearing vibration signal is decomposed into several stationary intrinsic mode function（ IMF）,the sum of first eight IMF component as a frequency domain feature extracting,with 14 other time- frequency domain characteristic index of characteristic collection input into the general regression neural network（ GRNN）,establish the GRNN network model,for the three kinds of rolling bearings fault state for pattern recognition. Through analyzing and comparing the BP and GRNN two network model for fault diagnosis,verified the superiority and feasibility of GRNN network.
同期刊论文项目
同项目期刊论文
期刊信息
位置：