Shallow resources are becoming increasingly depleted, deep resource exploration has become a global strategy. The design and testing of deep in-situ core samples are prerequisites for exploring deep resources; however, no in-situ condition-preserved coring and testing techniques and tools have been reported yet. Here, the first deep in-situ condition-preserved coring system (with the preservation of pressure, temperature, substance, light, and moisture) was developed that considers the effects of high water pressure and formation dynamic loads, along with an in-situ condition-preserved testing system. A pressure-preserved controller was designed, achieving the ultimate capacity of 140 MPa and 150 ℃. A temperature-preserved coring system combining active heating and passive insulation was constructed, realizing temperature preservation from room temperature to 150 ℃. Three generations of film-formation principles and methods were designed, achieving an excellent quality preserved rate, moisture preserved rate, and visible light barrier rate. Moreover, a deep in-situ condition-preserved coring system, and a simulated coring platform for large cores under in-situ environments was fabricated. A non-contact testing system was derived to cut and prepare specimens under in-situ environment and to perform non-contact non-destructive testing and true triaxial testing. The research findings can be successfully applied to deep coal and gas development, deep oil and gas resources assessment, and deep-sea sediment prospecting, achieving excellent application outcomes. This study provides important theoretical, technical and hardware support for deep in-situ rock physics and mechanics research and deep resource exploitation.

Multistage fracturing is a commonly-used method to improve gas production in tight hydrocarbon reservoirs. Stress shadowing effect among multi-fractures is crucial in effectively connecting the pre-existing natural fracture of reservoir and forming a complex fracture network that facilitates gas flow. Challenges remain in accurately characterizing the fracture structure and propogation patterns of naturally fractured reservoirs. In this study, we adopt an adaptive finite-discrete element method to simulate the multistage fracturing of a naturally fractured reservoir by improving the mesh auto-refinement and identification of multiple fracture propagation. The numerical model covers interactions among hydraulic fractures, pre-existing fractures, and microscale pores, while integrates the nonlinear Carter leak-off criterion to describe fluid leak-off and hydromechanical coupling effects during multistage fracturing. We introduce the proppant transport equation for idealised parallel plate flow in fractures, and Darcy's law is adopted to analyse the seepage flow in the fracture network and determine gas recovery. We then compare the fracture network and consequent fluid flow induced by the hydrofracturing of unfractured and naturally fractured models to assess the influence of pre-existing fractures on multistage fracturing behaviour and gas production. This study provides a new approach to determine and optimize fracturing cluster spacing in tight gas reservoirs.