Application of Wavelet Transform-Based Seismic Frequency Division Geological Sweet Spots Prediction Technology in Tight Sandstone Gas Reservoirs

The Lower Jurassic Ahe Formation tight sandstone gas reservoir in the eastern Dibei area of the Kuqa Depression exhibits significant resource potential, serving as a key replacement target for increasing gas reserves and production in the Tarim Basin. Drilling data from the Kuqa Depression indicate that high-productivity wells commonly develop high-angle structural fractures, a phenomenon that aligns with the fracture development patterns observed in the Cenozoic fault propagation folds of the eastern Kuqa Depression. However, due to the complex formation mechanisms of fault-fracture systems influenced by multi-phase tectonic movements, exExisting seismic prediction techniques demonstrate limited consistency with actual drilling outcomes, significantly impeding exploration and development advancements. In response to the technical challenge, this study introduces an innovative approach: a multi-scale fracture prediction method that leverages wavelet transform for seismic frequency division. This method employs wavelet transform to decompose images into mid-low and high-frequency components, subsequently utilizing coherence attributes and most-positive curvature illumination to enhance the data volumes. attribute calculations, achieving spatial characterization and prediction of fault-fracture systems at different scales. Application results demonstrate that: (1) Based on mid-low frequency divided amplitude data, the imaging quality of Class Ⅲ faults in the high structural position of the Dibei area is significantly improved, with clear spatial distribution characteristics showing a near east-west trend; (2) Fracture prediction using high-frequency divided amplitude data reveals the development of high-angle structural fractures in the eastern flank of the structure, forming a fault-fracture zone favorable for future rolling development. The technology has been successfully applied in two new wells in the Ahe Formation of the Dibei area, achieving 100% accuracy in large-scale fault prediction and 85% accuracy in medium- to small-scale fracture prediction, while effectively guiding engineering risk warnings. The research results demonstrate that this method provides important practical guidance for well placement optimization and safe drilling in tight gas reservoirs within complex structural areas.