基于同步挤压改进短时傅里叶变换的谱分解应用

随着油气勘探开发深度的增加以及地震数据采集受外界的干扰严重,使得地震资料处理解释人员对于含油气层的识别也变得更加困难。基于时频分析的地震谱分解技术已经广泛应用于油气储层预测中;但由于短时傅里叶变换、小波变换、S变换、Wigner-Ville分布等传统时频分析方法受自身窗函数的约束,使得它们的时频聚焦性不高或交叉项干扰,导致油气检测结果存在很大的误差。针对这一难题,为了实现准确的储层预测,通过对短时傅里叶窗函数进行拓展,并且对拓展后的短时傅里叶变换结果执行挤压,将挤压结果重排放置于信号的瞬时频率处,提出了同步挤压改进短时傅里叶变换。信号分析表明同步挤压改进短时傅里叶变换具有更高的时频聚焦能力。将同步挤压改进短时傅里叶变换与地震谱分解技术结合,并将其运用于实际地震资料,结果表明,该方法可以对含油气层进行精细刻画,频率异常特征十分显著,对于含油气性检测具有很强的实用性。

Application of the spectral decomposition based on the improved short-time Fourier transform by synchronous extrusion

With the rise of the depth of the oil and gas exploration and development, and due to the severe disturbance of the seismic data acquisition by the outside world, it is more difficult for the seismic data processing and interpreting personnel to identify the oil and gas layers. The seismic spectrum decomposing technique based on the time-frequency analysis has been widely applied in the prediction of the oil and gas reservoirs;however, because the traditional time-frequency analysis methods such as short-time Fourier transform, wavelet transform, S-transform, Wigner-Ville distribution and so on are constrained by their own window functions, their time-frequency focusing property is not high or the cross-terms are interfered, those result in a large error in the oil and gas detection results. For this problem, in order to achieve the accurate reservoir prediction, the short-time Fourier window function was expanded, and moreover the expanded short-time Fourier transform results were extruded, the extrusion result was rearranged at the instantaneous frequency of the signal, and furthermore the short-time improved Fourier transform by the synchronous extrusion was developed. The signal analysis shows that the improved short-time Fourier transform has pretty higher time-frequency focusing ability. The new transform was integrated with the seismic spectrum decomposing technique and then applied in the actual seismic data. The results show that the method can carry out the fine characterization of the oil-bearing layer, the frequency anomaly is very obvious, thus it has much stronger practicability for the oil and gas detection.