To address the sanding monitor issue in wells during high water cut period,wavelet packet analysis method is used to identify sand vibration signals at the pipe wall.A sanding monitoring laboratory test was designed to obtain sand vibration signal,which is decomposed by db4 wavelet packet and three layers.The energy of each component is normalized.The signal component containing the sanding characteristics is analyzed and reconstructed.The high frequency band fine decomposition method is used to highlight the frequency domain localization characteristics of the sanding signal.The frequency domain feature of the reconstructed signal indicates that the sanding signal frequency range is 19 ～25 kHz.The power spectrum characteristics of the reconstructed signal indicate that the power spectrum amplitude of the sanding signal has a quadratic increase relationship with the sand content.The wavelet packet analysis method can effectively identify the sanding vibration signal.The conclusions provide a reference for identifying the sanding information of wells under complex fluid conditions.%针对高含水期的油井出砂监测问题,基于小波包分析方法对管壁处出砂振动信号进行识别研究.通过设计出砂监测室内试验,获取出砂振动信号,对该信号进行db4小波包3层分解,归一化每组分量的能量,分析含有出砂特征的信号分量,并进行重构.采用高频段精细分解方法,突出了出砂信号的频域局部化特征.重构信号的频域特征表明,出砂信号的频率区间为19 ～ 25kHz.重构信号的功率谱特征表明,出砂信号功率谱幅值与含砂量之间呈二次增长关系,采用小波包分析方法可以有效地识别油井出砂振动信号.所得结论可为识别复杂流体条件下的油井出砂信息提供参考.
展开▼
