摘要
碰摩是旋转机械中常见的故障,碰摩故障位置识别是一个有待深入研究的故障诊断问题。对球结构支持向量机进行改进,充分考虑分类球的大小对分类的影响,经过理论分析和仿真实验得到新的分类规则。把转轴上不同位置的碰摩当作不同的故障,转轴碰摩故障位置识别就是个较大规模的多类别故障诊断问题,运用改进的球结构支持向量机进行转轴碰摩故障位置识别。实验结果表明,和其他同类算法相比,改进的球结构支持向量机具有识别率高、速度快、计算量少、数据处理容量大等优点,适合于较大规模的多类别故障诊断。
Sphere support vector machines, a kind of multi-classes support vector machines, were improved taking into account the influence of sphere size on classification and an effective and simple rule of classification was derived through rational analysis and simulation experiment. The rub-impacts in different parts of shaft were regarded as different faults, and their position detection becomes a large-scale multi-classes faults diagnosis and can be solved by the improved sphere support vector machines. The experiment of rub-impact position detection of simulated shaft shows that the improved sphere support vector machines distinctly improve the fault recognition accuracy, the diagnosis rapidity and the data processing capability, and are more suitable for practical application in large-scale multi-classes faults diagnosis.
出处
《振动与冲击》
EI
CSCD
北大核心
2009年第8期70-73,77,共5页
Journal of Vibration and Shock
基金
教育部新教师基金项目(200805041079)
国家杰出青年科学基金(50425516)
关键词
故障诊断
球结构支持向量机
多类分类算法
碰摩
faults diagnosis
sphere support vector machines
multi-classes support vector machines
rub-impact fault
