Research on in-Situ Stress Prediction Technology for Tight Sandy Conglomerate Gas Reservoirs in Faulted Basins: A Case Study of the Shahezi Formation in Anda Block, Songbei Region

Accurate prediction of crustal stress in deep rifted basins plays a pivotal role in the efficient exploration and development of tight sandy conglomerate gas reservoirs. However, the structural complexity and reservoir heterogeneity in the northern Anda Block of the Songliao Basin, targeting the Shahezi Formation, pose significant challenges to stress prediction, resulting in low precision. To address this issue, this study proposes a multi-source crustal stress prediction technology integrating well logging, seismic data, and field measurements. Dynamic-static mechanical parameter relationships are established through petrophysical core analysis, enabling the derivation of static mechanical parameters from logging data. By incorporating tectonic strain, formation pore pressure, and rock mechanical parameters into Huang’s model, a crustal stress prediction model is constructed. The model is calibrated using in-situ stress data from hydraulic fracturing tests, forming a multi-source data fusion optimization method that achieves an accuracy exceeding 93%. Finally, based on the prestack elastic parameter inversion results derived from the ‘Two-Wide One-High’ (wide-azimuth, broadband, and high-density) 3D seismic data, the spatial distribution characteristics of the in-situ stress field are characterized. Results indicate that the orientation of maximum horizontal principal stress (σH) in the Shahezi Formation predominantly trends NW, with horizontal stress differences ranging from 12 to 18 MPa. This technology provides a robust methodology for crustal stress prediction in complex rifted basins, supporting sweet spot identification, well pattern optimization, and hydraulic fracturing design. The approach is applicable to other heterogeneous tight gas reservoirs with analogous geological conditions.