摘要
针对齿轮箱故障特征重叠难以有效分离问题，提出基于局部切空间排列与多核支持向量机的齿轮箱故障诊断模型。在由振动信号时域统计指标及内禀模态分量能量构造的多元特征空间中，据局部切空间排列算法对多元特征进行非线性降维处理，得到初始低维流形结构，获取最优敏感特征向量；将该特征向量输入至多核支持向量机进行学习训练与故障辨识。局部切空间排列能克服传统降维方法的不足，多核支持向量机可实现复杂故障高精度、自动化智能诊断。通过齿轮箱故障模拟实验验证该方法的有效性。
Abstract
Aimed at the overlapping of the gearbox fault features and these features are difficult to distinguish, a fault diagnosis model based on local target space alignment and multi-kernel support vector machine for gearbox is proposed. In the vibration feature space constructed by time domain statistics and intrinsic mode energy value, the nonlinear multi-dimensionality reduction based on local target space alignment to get the initial low-dimensional manifold feature value is firstly executed, then the low-dimensional feature vector which have the fault characteristics are regarded as the input feature vector of the multi-kernel support vector machine for gearbox fault classification. Local target space alignment overcome the shortcoming of the traditional reduction method, multi-kernel support vector machine realized the high-precision, automated intelligent diagnosis for gearbox. The gearbox fault diagnosis experiment shows the effectiveness of this novel model.
关键词
局部切空间排列 /
多核学习 /
支持向量机 /
齿轮箱 /
故障诊断
{{custom_keyword}} /
Key words
local target space alignment /
multiple kernel learning /
support vector machine /
gearbox /
fault diagnosis
{{custom_keyword}} /
陈法法;汤宝平;苏祖强.
基于局部切空间排列与MSVM的齿轮箱故障诊断[J]. 振动与冲击, 2013, 32(5): 38-42
CHEN Fa-fa;TANG Bao-ping;SONG Tao.
Gearbox fault diagnosis based on local target space alignment and multi-kernel support vector machine[J]. Journal of Vibration and Shock, 2013, 32(5): 38-42
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1589 KB)
