论文部分内容阅读
论述了关联维数在大型旋转机械故障诊断中的应用 ,并针对故障诊断的实际情况 ,对关联维数的 G- P算法作了一定的改进 .传统的 G- P算法使用欧几里德范式计算点间距 ,包含了较多的重复运算 ,使用改进的点间距计算公式不仅简单 ,而且可借助于递推公式以节省计算时间 ;引入最小动态关联时间以便消除流数据的动态关联特性 ;建议采用局部斜率曲线以判断标度区间的范围 .对数据长度和噪声干扰对于关联维数计算结果的影响进行了仿真分析 .研究结果表明 :由于不同故障的动力学产生机制不同 ,通常也具有不同的关联维数 ,故关联维数可用于故障的特征提取
This paper discusses the application of the correlation dimension to the fault diagnosis of large rotating machinery, and improves the G-P algorithm of the correlation dimension according to the actual situation of fault diagnosis.The traditional G-P algorithm uses the Euclidean norm Computation of point spacing, including more repetitive calculations, the use of improved point spacing calculation formula is not only simple, and can be recursive formula to save the calculation time; the introduction of the minimum dynamic association time in order to eliminate the dynamic correlation of stream data; recommended The local slope curve is used to judge the range of scale interval.Analyzed the effect of data length and noise on the correlation dimension calculation results.The results show that due to the different mechanism of dynamic generation of different faults, Dimension, so the correlation dimension can be used for fault feature extraction