This paper is the result of oil and gas exploration survey engineering. Objective Xiaoquan-Fenggu area in the western Sichuan Basin has huge potential for exploration and development of Xujiahe Formation gas reservoirs. However, due to the region's deep burial, complex structure, and multiple superimposed fault systems that cause frequent variation in stress orientation, effective well placement design and hydraulic fracturing practices have been limited. Therefore, it is necessary to evaluate the precise magnitude of in-situ stress in this area to provide recommendations for engineering development and increase production capacity. Methods Experimental methods such as rock mechanics, acoustic emission testing, and differential strain analysis combined with conventional logging, special logging, and hydraulic fracturing data were used to experimentally test the in-situ stress magnitude in deep heterogeneous blocky tight sandstone reservoirs. Based on accurate evaluation of single-point in-situ stress magnitude, a logging interpretation model was established for the subdivision of tectonic units in the study area, examining structural variation of in-situ stress magnitude and distribution along a single well. Results Our study showed that differential strain analysis provided the most accurate measurement of stress in heterogeneous tight sandstone reservoirs. Test results indicated that the Xujiahe Formation belongs to the type III in-situ stress category and exists in a strike-slip stress state with partial compression and thrust stress states. Based on single-point test data, we developed a technique to evaluate in-situ stress magnitude by utilizing borehole image inversion. Structural changes in in-situ stress magnitude were divided vertically into five types, whereby high positions of north-south (SN) faults with grades above three and folds in SN or northeast-trending (NEE) resulted predominantly in low-low-high (LLH) or low stress (LC) profiles. Meanwhile, small-scale faults or gentle deformation areas had high-low-high (HLH) or high-low-low (HLL) profiles. Conclusion A low-low-high (LLH) stress profile was suggested for engineering development to penetrate more gas layers vertically, while avoiding bottom water and preventing rapid water breakthrough during early production. Therefore, it is recommended to select the second to third order north-south(SN) trending faults and north-south(SN) or northeast trending(NE) longitudinal flexure zones located in the middle-upper part of the second layer of the second member of Xujiahe Formation.