This paper presented the development process and applicable conditions of the classical seepage mechanics theory-Darcy's law, and derived the mathematical expressions of Darcy's law in capillary seepage and fracture seepage in porous media from the Navier-Stokes (N-S) equations.The paper pointed out eight major problems in the current application of Darcy's law and comprehensively analyzed the main challenges of seepage mechanics theory in oil and gas field development.In response to these challenges, this paper proposed a series of countermeasures and considerations.The paper emphasized that constructing multiscale, multi-physics field coupling models and leveraging AI scientific computing is the only way to reveal the complex and real flow mechanisms of oil and gas reservoirs and fill the current theoretical gaps.It was suggested to further develop high-precision experimental techniques such as nuclear magnetic resonance, electron microscopy scanning, and intelligent data and image processing to visually demonstrate the behavior and processes of fluids in reservoirs.Finally, it was suggested to use experimental research, the establishment of new theoretical models and AI for Science to innovate and break through the current challenges in the theory of oil and gas seepage mechanics.It provided an important reference for universities, research institutions and researchers to carry out theoretical research and project establishment of petroleum science, and it can also provided strong technical support for the scientific and technological development strategic planning of oil and gas resources in China.

Artificial neural networks have been widely applied in reservoir engineering. As a powerful tool, it changes the way to find solutions in reservoir simulation profoundly. Deep learning networks exhibit robust learning capabilities, enabling them not only to detect patterns in data, but also uncover underlying physical principles, incorporate prior knowledge of physics, and solve complex partial differential equations. This work presents the latest research advancements in the field of petroleum reservoir engineering, covering three key research directions based on artificial neural networks: data-driven methods, physics driven artificial neural network partial differential equation solver, and data and physics jointly driven methods. In addition, a wide range of neural network architectures are reviewed, including fully connected neural networks, convolutional neural networks, recurrent neural networks, and so on. The basic principles of these methods and their limitations in practical applications are also outlined. The future trends of artificial intelligence methods for oil and gas reservoir development are further discussed. The large language models are the most advanced neural networks so far, it is expected to be applied in reservoir simulation to predict the development performance.