A Study Of The Dynamic Level Signal Denoising Based On Improved Wavelet Threshold Method

In the field of petroleum industry, the well dynamic fluid level parameter is the key of conducting the oil production technology adaptability evaluation and optimization. If the measurement results were not accurate, then would lead to the pumping unit "air pumping" phenomenon. Currently, most of oilfields use echo method to measure the depth of the well dynamic fluid level. This approach sends acoustic signal into pumping unit wells by using echo instrument, surface acoustic signal occur reflection when it encounters the liquid surface, the echo instrument receives the reflected wave signal. Due to the complex environment in the pumping unit wells and many kinds of noise and a high degree of interference, it would lead to the received signal with noise, seriously affect the accuracy of measurement results.So the denoising processing of reflected wave signal is important. In recent years,signal denoising field had made rapid development and put forward many denoising algorithms, among these, due to wavelet transform is a local time and frequency conversion, so it can effectively extract information from the signal, and the signal is analyzed by multi-scale operations such as scaling and translation functions, so it has became the most widely used method in the field of signal denoising. This paper is based on wavelet transform, around the traditional wavelet thresholding defects, put forward an improved wavelet threshold method, and apply this method in the well fluid level signal denoising, then concluded by contrast.Firstly, this paper introduces the relevant concepts of well dynamic fluid level,common measurement methods, commonly used methods of wavelet transform denoising, focusing on the basic principle of wavelet thresholding and denoising steps,factors of affecting the denoising effect, Describes the shortcomings of the traditional soft and hard threshold method and garrote threshold method, provides a theoretical basis for improving the proposed threshold method.Secondly, for the existence defects of the traditional threshold methods, this paper gives several principles of constructing excellent denoising threshold function that should be followed. Then propose improved wavelet threshold method and proves its feasibility.Appling traditional threshold method and the improved threshold method for signal denoising using MATLAB simulation software, by contrasting visual evaluation criteria(simulation graphs) and objective evaluation criteria(signal tonoise ratio(SNR), the mean square error(MSE)) then proved the improved wavelet threshold method is better than the traditional methods.Finally, doing the well dynamic fluid level signal denoising processing by using improved wavelet threshold method, the traditional hard and soft threshold and Garrote threshold method, the simulation results and performance data prove the improved method in the well fluid level denoising has good results and introduced this threshold method application prospects.