小波变换除噪在油井出水机理诊断中的应用

为提高油藏调剖堵水的成功率，必须搞清油井的出水机理及主要影响因素，利用油井水油比对时间导数曲线的形状，根据理论分析结果，可以判断油井出水的原因。由于实际生产动态数据中存在噪音信号的污染，使得判 别曲线的跳跃性很大，甚至难以识别，从而降低了出水机理诊断的准确性和可靠性。因此，在进行出水类型判断之前首先需要对油井的生产动态数据进行预处理。采用小波变换方法对实际数据进行除噪，可以保证水油比 导数曲线具有很好的光滑性和逼近度。实例证明，该方法简单易用，结果可信。

APPLICATION OF WAVELET TRANSFORMATION DENOISING IN EXCESS WATER PRODUCTION MECHANISM DIAGNOSIS

According to theoretical analysis, the time derivative curves of water-oil ratio can be used to diagnose the reasons of water output in oil well. However, the interference of noise signals in production dynamic data reduced the accuracy and reliability of water output diagnosis. Consequently, the pre-treatment of production dynamic data is necessary before estimating the production water type. Using wavelet transformation method to denoise for real production data can simultaneously guarantee both the smoothness and the approximation of the derivative curves. The simplicity and reliability of this denoising method is verified by field examples.