Hydrocarbon detection method based on fractional adaptive superlet transform

The lack of time-frequency resolution of seismic signals in complex reservoirs limits the precision of hydrocarbon detection. The limitations of conventional spectral decomposition methods in frequency band selection and time-frequency focusing significantly affect the reliability of key hydrocarbon indicators, such as fluid mobility properties and high-frequency attenuation characteristics, and there is an urgent need for the development of high-resolution temporal-frequency analysis methods. A hydrocarbon detection method based on the fractional adaptive superlet transform (FASLT) is proposed. The FASLT optimizes the time-frequency resolution by combining wavelet sets with different bandwidths and using geometric averaging to mitigate the limitation of the uncertainty principle of the traditional method, and introduces the adaptive and fractional-order superlets to achieve the time-frequency super-resolution across frequencies. The algorithm dynamically extracts the dominant frequency band by combining the adaptive centroid frequency algorithm and constructs a two-parameter joint detection model of fluid mobility attributes and high-frequency energy attenuation gradient to improve the hydrocarbon detection precision. Based on the seismic data of an offshore area, the results show that compared with the conventional time-frequency analysis method, the accuracy of this method in identifying gas-bearing reservoirs is significantly improved, and the detection results are basically consistent with the logging data, which verifies the effectiveness and practicality of the method in improving the precision of hydrocarbon detection in complex reservoirs.