摘要
随着勘探难度的增加,小波变换时频分辨率的精度已经难以达到实际勘探目标的要求,这就需要探索分辨率更高的时频分析方法。同步挤压小波变换(Synchrosqueezing Wavelet Transform-SWT)通过对小波变换的复系数谱在频率方向上进行压缩重组,得到比小波变换更高的时频分辨率,同时还具有可逆性和一定的抗噪性。通过模拟信号的测试结果表明,同步挤压小波变换在刻画信号的时频特征方面精度更高、准确性更好;而在实际地震资料的分析当中,利用同步挤压小波变换对舍气薄储层的地震数据进行分频处理,其低频处出现明显异常,而且频率越低异常越明显,有效地预测出储层的存在。
With the increasing difficulty of exploration, the accuracy of the time-frequency resolution of wavelet transform has been difficult to meet the requirements of the actual exploration targets, which needs to explore a higher resolution time-frequency analysis method. The synchrosqueezing wavelet transform （SWT） can obtain time-frequency resolution better than the wavelet transform by squeezing and recon- structing complex coefficient spectra in frequency direction. And it also has reversibility and anti-noise property.The simulation results show that the synchrosqueezing wavelet transform has better accuracy in characterizing the time-frequency characteristics of the signal.And in the analysis of the actual seismic da- ta,the frequency division process of the synchrosqueezing wavelet transform is applied to process the thin gas-bearing reservoir seismic data.Where the low frequency appears obvious anomalies, and the lower frequency is,the more obvious anomalies are.So the reservoir is predicted effectively.
出处
《天然气地球科学》
EI
CAS
CSCD
北大核心
2017年第2期341-348,共8页
Natural Gas Geoscience
基金
国家重点研发计划重点专项项目"近海底高精度水合物探测技术"(编号:2016YFC0303900)资助
关键词
同步挤压小波变换
时频分辨率
分频处理
低频异常
储层预测
Synchrosqueezing wavelet transform
Time-frequency resolution
Frequency division processing
Low-frequency anomaly
Reservoir prediction
