Application of Wavelet Threshold De-noising Method in Processing sanding signal of Oil Wells
- 作者：Song NanDang Rui-RongXu Jing-Jing
- 会议时间：2013-09-16
- 关键词：wavelet transform ; threshold de-noising ; sanding signal ; piezoelectric ultrasonic sensor
- 作者单位：School of Electronic Engineering, Xi'an Shiyou University, Xi'an, 710065, P.R.China
- 母体文献：2013油气田监测与管理联合大会论文集
- 会议名称：2013油气田监测与管理联合大会
- 会议地点：西安
- 主办单位：中国石油学会
- 语种：chi
摘要
Oil and gas contained in sandstone reservoirs are likely to encounter sand production problem due to reservoir pressure depletion, water production, poor operation philosophy, inadequate completion practices, etc.The sand in the high-speed fluid will collide the metal wall when it moves to the elbow of the pipeline.A pulse vibration signal will be produced randomly.Sanding signal is regarded as vibration signal.It is a non-stationary weak random signal.The signal belongs to ultrasonic signal.The external acoustic detecting technique is used widely.In this method the piezoelectric ultrasonic sensor will be installed on the outer wall of the bottom of the pipe bend in order to sample the vibration signal.The quantity of sand produced has been determined using a ultrasonic sensor.The signal contains a variety of random noise signal, such as the piping vibration, electromagnetic interference…Conventional processing methods alone do not give the true picture of sand produced in pipeline.It is necessary to take effective signal de-noising method.Wavelet de-noising is an important application field of wavelet transform.Especially threshold de-noising method has been developed rapidly to an efficient de-noising method.The peak and the mutation part in the original signal can be retained by wavelet threshold de-noising.The value of SNR can be seen that multiples of the SNR improved by wavelet de-noising is bigger than the Fourier transform de-noising.The signal is weaker;the improving effect is more obvious.In order to explain the combination de-noising ability of wavelet and threshold value, de-noising effect is contrasted by using different wavelet base, different decomposition layer and different threshold criterion Through analyzing and comparing feature of de-noised signal, the Sym8 wavelet of Sym wavelet series and the Rigrsure threshold rule are selected.Experiments show that better results can be achieved by the method of wavelet threshold de-noising.