气液两相流压差波动信号小波去噪中阈值规则的确定

利用小波分解、滤波、重构对压差波动信号进行去噪处理,消除实验中由于各种因素产生的噪声,得到了较为满意的压差波动实验数据,该方法与其他处理方法相比,得到的处理信号不失真且去噪效果明显。比较发现小波去噪的结果与阈值选取相关,通过分析确定合适的阈值选取规则,使得处理后的信号能满足以后分析的需要。研究表明选择的阈值可广泛应用于不同流型和不同取压间距的压差波动信号。

Confirmation of Threshold Rule in Signal Denosing of Gas-Liquid Two-Phase Flow Differential Pressure Fluctuation Using Wavelet Transform

By using wavelet decomposition, filtration and reconstruction, noise signal could be eliminated from differential pressure fluctuation signal and better experimental data of differential pressure fluctuation can be obtained. Compared with other methods, the method can obtain non-anamorphic and well-denoised signal and it can be found that denoise effect is related to the choice of threshold. The signal which be processed by proper rule of threshold can be fit for the necessary of analysis. Result shows that the corresponding threshold value can be widely applied to the differential pressure signal of various flow patterns and pressure measure intervals.