Rate of penetration (ROP) is crucial for evaluating drilling efficiency, with accurate prediction essential for enhancing performance and optimizing parameters. In practice, complex and variable downhole environments pose significant challenges for mechanistic ROP equations, resulting in prediction difficulties and low accuracy. Recently, data-driven machine learning models have been widely applied to ROP prediction. However, these models often lack mechanistic constraints, limiting their performance to specific conditions and reducing their real-world applicability. Additionally, geological variability across wells further hinders the transferability of conventional intelligent models. Thus, combining mechanistic knowledge with intelligent models and enhancing model stability and transferability are key challenges in ROP prediction research. To address these challenges, this paper proposes a Mechanism-Data Fusion and Transfer Learning method to construct an intelligent prediction model for ROP, achieving accurate ROP predictions. A multilayer perceptron (MLP) was selected as the base model, and training was performed using data from neighboring wells and partial data from the target well. The Two-stage TrAdaBoost.R2 algorithm was employed to enhance model transferability. Additionally, drilling mechanistic knowledge was incorporated into the model’s loss function as a constraint to achieve a fusion of mechanistic knowledge and data-driven approaches. Using MAPE as the measure of accuracy, compared with conventional intelligent models, the proposed ROP prediction model improved prediction accuracy on the target well by 64.51%. The model transfer method proposed in this paper has a field test accuracy of 89.71% in an oilfield in China. These results demonstrate the effectiveness and feasibility of the proposed transfer learning method and mechanistic–data integration approach.

Efficient proppant transport in conglomerate reservoirs is severely constrained by rough fracture surfaces and strong geometric heterogeneity, leading to premature near-wellbore deposition and insufficient distal support. To address this challenge, this study aims to clarify the transport and deposition mechanisms of proppants in rough-wall fractures representative of the Mahu conglomerate reservoir. A large-scale visualized physical simulation system with an artificial rough fracture (20 m length × 4.5 m height) was developed based on dynamic similarity principles, enabling long-distance proppant transport observation under controlled pumping rate, fluid viscosity, proppant size, and sand concentration. Ten systematic experiments were conducted, and real-time particle motion and sand ridge evolution were captured using high-speed imaging and pressure monitoring. The results show that proppants form longitudinally layered sand ridges that evolve through three stages: leading-edge initiation, equilibrium-height growth, and distal extension. Viscosity and sand concentration primarily control propped-area continuity, while pumping rate governs transport distance and particle size affects structural stability. Rough fracture surfaces significantly intensify near-wellbore accumulation by enhancing energy dissipation and local flow heterogeneity. These findings provide mechanistic insights into proppant transport in rough fractures and offer quantitative guidance for optimizing fracturing parameters in conglomerate reservoirs.