关联维数在汽轮机故障诊断中的应用

Application of correlative fractal dimension to the vibration faults in steam turbine shaft systems

在汽轮机轴系典型故障模拟试验的基础上,对故障信号的关联维数进行了计算研究.采用G-P算法分别计算了故障信号的时域波形及小波高频重构波形的关联维数,采用自相关函数法确定延迟时间.计算结果表明,对原始信号进行高频重构后再计算关联维数,可以达到突出故障信息,增大关联维数对故障的区分度的目的,此关联维数可以作为汽轮机故障诊断的定量指标.

Time-serials waveforms of the faulty signals and the correlative fractal dimension of reconstructed high frequencies wavelets were calculated by G-P arithmetic and their delay-times were evaluted by self-correlation function, after typical faulty signals from the rotor test rigs are simulated. The analysis shows that correlative dimension can be used as a quantitative index for fault diagnosis. Compared with correlative fractal dimension calculated from the time-serials of faulty signal directly, the correlative dimension calculated from the high frequency reconstruction signal has better identification ability than that from time-serials.