基于EEMD的改进小波阈值算法在超声水表中的应用

Application of improved wavelet threshold algorithm based on EEMD in ultrasonic water meter

针对超声水表在实际工作环境中容易受到噪声干扰从而导致计量精度下降的问题,提出了基于集合经验模态分解(EEMD)的改进小波阈值降噪算法。为了提高降噪效果,对小波阈值降噪算法进行了改进,构造了非线性阈值函数取代传统阈值函数,同时给出了一种分解尺度选择的方法。利用EEMD将流速信号分解为一系列的本征模态函数,通过改进小波阈值降噪算法对本征模态函数进行降噪处理,去除其中的噪声分量,为了验证该算法的适用性,将其与小波阈值降噪算法和时空滤波分析方法进行了比较。试验结果表明,以超声水表流速信号为降噪对象时,基于EEMD的改进小波阈值降噪算法具有较好的降噪效果。

Here, aiming at the problem of ultrasonic water meter being easy to be disturbed by noise in the actual working environment to cause decline of measurement accuracy, an improved wavelet threshold denoising algorithm based on ensemble empirical mode decomposition(EEMD) was proposed. In order to improve denoising effect, the wavelet threshold denoising algorithm was improved, a nonlinear threshold function was constructed to replace the traditional threshold function, and a method of decomposition scale selection was given. Flow rate signal was decomposed into a series of eigenmode functions using EEMD, and eigenmode functions were denoised by improving the wavelet threshold denoising algorithm to remove noise components. In order to verify the applicability of the proposed algorithm, it was compared with the wavelet threshold denoising algorithm and the space-time filtering analysis method. The test results showed that the improved wavelet threshold denoising algorithm based on EEMD has better denoising effect when taking flow rate signal of ultrasonic water meter as the object to be denoised.