基于DSP的柴油机振动信号小波降噪实时性研究

Real-Time Research of Diesel Engine Vibration Signal Denoising by Wavelet Based on DSP

针对柴油机实时监测系统采集的振动信号采样频率高、实时降噪困难,而传统的实时性评价方法难以准确描述数字信号处理器(DSP)中所需降噪时间的问题,提出基于DSP的小波变换指令周期模型作为实时性的评价依据,对比不同小波算法的实时性优劣;根据采集的柴油机缸盖振动信号特征,提出基于变异系数定权法的综合评价指标,从降噪效果和实时性两个角度优选了小波基函数.结果表明:该模型与DSP中小波算法的实时性吻合较好,且优选的小波降噪参数能满足采样频率为25 kHz下的实时降噪需求.

In view of the high sampling frequency of vibration signals collected by the diesel engine real-time monitoring system and the difficulty of real-time noise reduction,the traditional real-time evaluation method is difficult to accurately describe the noise reduction time needed in digital signal processor(DSP).A wavelet transform instruction period model based on DSP was proposed and the real-time performance of different wavelet algorithms was analyzed and compared.According to characteristics of the vibration signals collected from the cylinder head of diesel engine,a comprehensive evaluation index based on variation coefficient was proposed,and the number of wavelet decomposition layers and wavelet basis function were optimized from the aspects of noise reduction effect and real-time.The results show that the model is in good agreement with the real-time performance of wavelet algorithm in DSP,and the optimized wavelet denoising parameters can meet the real-time noise reduction requirements at 25 kHz sampling frequency.