基于独立分量分析的高压油管压力信号消噪研究

分析了用夹持式油压传感器采集高压油管压力信号的噪声特点和组成,针对传统的消噪方法无法较好地消除油压信号中混叠的油管振动信号的问题,提出基于独立分量分析的消噪方法。实测数据分析结果表明,基于负熵的快速固定点算法成功实现了油压信号和油管振动信号分离,提高了油压信号的信噪比。

Study on Signal Denoising of High-pressure Fuel Pipe Pressure Based on ICA

The characteristic and composition of noise for high-pressure fuel pipe pressure signals collected by clamping pressure sensors were analyzed.For the problem that the traditional denoising method could not effectively eliminate the pipe vibration signal that was overlapped with the fuel pressure signal,the denoising method based on independent component analysis(ICA) was put forward.Analysis results showed that the fast fixed-point algorithm based on negentropy could separate of the pipe vibration signal from fuel pressure signal and improve the SNR of fuel pressure signal.