In high-pressure formations, gas influx often leads to high shut-in standpipe and casing pressures. Due to complex geological conditions and uncertainties in the pressure system, the variable-density well killing method is often used for safe well control. In this paper, a general matrix boundary of killing fluid density and volume was proposed, and a dynamic variable-density well killing model suitable for complex high-pressure formations was developed. The accuracy and reliability of the model were validated using the field measurement data and the recognized well control simulator-Drillbench, with good agreements. The developed model can impose any sequence of killing fluid densities and dynamically update subsequent well killing parameters when control strategies change. It overcomes the limitations of previous models, which required predetermined killing fluid densities before well killing and lacked the capability to adjust parameters in real time. The calculated results indicate that, under high shut-in standpipe and casing pressures, if the formation has enough pressure-bearing capacity, implementing aggressive followed by gradual killing fluid density increments, increasing large-density fluid volume, and reducing killing fluid density adjustment frequency enables rapid standpipe and casing pressures reduction. For formations with difficult-to-estimate pore pressure and low pressure-bearing capacity, the control strategies are opposite, and the lost circulation materials should be added to the killing fluid during early well killing circulation to improve the formation's pressure-bearing capacity. Additionally, before the killing fluid reaches the drill bit, increasing the flow rate is the only means of reducing high casing pressure. Moreover, during the annulus gas removal, the change in casing pressure is consistent with that in pit gain. So, a simplified real-time casing pressure calculation method was proposed with the standpipe pressure as the primary control target, which overcomes the empiricism and blindness of conventional casing pressure control under the influence of manual control errors, pressure wave propagation delays, and secondary kicks. By integrating variable-density well killing parameter predesign with real-time decision, this research provides a more effective theoretical guidance for safe well control in complex high-pressure formations.

High-speed sand-laden fluids can cause severe erosion of the packer mandrel in downhole packer systems. However, predictive models for packer systems under coupled thermo-mechanical stresses remain underdeveloped, and no research has been conducted on the multiphase flow erosion of the packer mandrel with temperature considerations. This study applies a coupled CFD–DPM framework to systematically investigate multiphase flow–particle interactions and erosion behavior in a retrievable test-treat-squeeze (RTTS) packer mandrel. The computational framework employs Lagrangian particle tracking coupled with turbulent flow modeling, enabling precise quantification of erosion patterns influenced by five key operational parameters: fluid velocity (5-35m/s), temperature (160-220°C), silica particle size (0.2-4.0mm), solids loading (0.5-5.0kg/s), and diameter ratio (0.35-0.80). The results show that the maximum erosion rate increases by up to 25.5 -fold as flow velocity increases, and by 8.27-fold as solids loading increases. Particle size exhibited a non-monotonic effect on erosion. Temperature exhibits a limited influence on erosion, contributing less than 2% variation under the assumption of constant fluid properties. Flow field analysis reveals that vortex intensity and particle impact localization are highly sensitive to flow acceleration and geometric contraction. The research results can provide a reference for the structural optimization design and failure position prediction of packer, in order to improve the service life and reliability of packer.

Summary Deepwell drilling is strongly affected by drillstring vibrations, especially stick/slip and whirl, which threaten safety and efficiency. Despite the extensive development of downhole measurement tools, systematic quantitative characterization and mechanistic analysis of vibration features remain insufficient due to the unpredictability of complex downhole conditions. For this study, we utilized high-frequency near-bit measurements of engineering parameters and triaxial acceleration data, combined with surface logging data, to systematically analyze the dynamic characteristics of three typical operating conditions—normal drilling, pure stick/slip vibration, and coupled whirl/stick/slip vibration. Time-domain statistical analysis, fast Fourier transform (FFT), and short-time Fourier transform (STFT) methods are used to establish quantitative dynamic feature signatures for each condition. Results reveal that normal drilling exhibits a quasistatic equilibrium with a dominant lateral vibration frequency at the rotation frequency and an acceleration amplitude of 0.0145 g. In contrast, pure stick/slip conditions show extreme downhole rotational speed (revolutions per minute or RPM) oscillations [coefficient of variation (CV) up to 350%], with the lateral vibration frequency shifting to the stick/slip characteristic frequency and amplitude reaching 0.259 g. Coupled whirl/stick/slip conditions demonstrate anomalous downhole torque CV peaks (up to 100%) and a lateral vibration amplitude surging to 0.511 g, reflecting strong nonlinear torsional/lateral coupling instability. Based on these findings, three quantitative evaluation indices are proposed as follows: the stick/slip index (SSI) for torsional vibration severity, the whirl index (WI) based on the triaxial vibration energy ratio, and the whirl impact evaluation parameter (WIEP), which utilizes the kurtosis features of engineering parameters for low-cost whirl detection. Validation using continuous drilling data from two representative wells with distinct well trajectories, bottomhole assembly (BHA) configurations, and measurement-sub positions shows that the proposed indices respond rapidly to condition transitions. In particular, the WIEP preserves stable whirl-event contrast in both tested wells, providing initial field evidence for the engineering applicability of the index system. This research provides a theoretical and feature-engineering basis for real-time monitoring and intelligent recognition systems in deepwell drilling.

To reveal the dynamic damage and instability mechanisms of coal–gangue composites in deep coalbed methane horizontal wells under drill string–wellbore collision disturbance, on-site transient collision was converted into SHPB incident-wave loading parameters based on impulse equivalence. A calibration relationship between impact disturbance intensity and laboratory impact pressure was established, and uniaxial and true-triaxial SHPB tests were conducted under different interface inclinations, impact pressures, and prescribed confining pressure levels. The results show that impact pressure, confining pressure, and interface inclination jointly control the dynamic response of the composite. Increasing impact pressure raises the incident-wave amplitude, input energy, and loading strain rate, resulting in higher dynamic strength, peak strain, and dynamic stiffness. True-triaxial confining pressure promotes fracture closure, strengthens interface normal constraint, and inhibits lateral expansion, thereby improving load-bearing capacity and delaying damage evolution. The three-dimensional stress response generally follows σx > σy > σz, indicating that the X direction is the main load-bearing direction, whereas the Y and Z directions reflect lateral stress redistribution and deformation feedback. Interface inclination shows a weak-plane control effect. The 0° and 90° specimens are mainly governed by axial compression and interface compaction, while the 60° specimen is more prone to interface shear sliding, tensile–shear crack propagation, and weak-plane coalescence, making it a relatively unfavorable orientation for impact resistance. Energy dissipation shifts from tensile-dominated to compression–shear coupled dissipation under the joint regulation of the three factors. These results provide an experimental basis for impact-damage analysis and wellbore-stability evaluation in deep coalbed methane horizontal wells containing gangue interlayers.

To address the industry challenge that the formation breathing effect in fractured formations narrows the safe mud weight window and significantly increases well control difficulty, this study employs two approaches—a self-designed experimental apparatus for the formation breathing effect and a combined finite-discrete element method (FDEM) numerical model—to systematically reveal the characteristic behavior and underlying mechanism of this effect, and to establish a prediction method for near-wellbore lost-circulation pressure that accounts for the breathing effect. The numerical simulation achieves high quantitative accuracy, with errors of less than 2.1% during the loss stage and less than 4.1% during the flowback stage. The results show that the typical signature of the breathing effect in fractured formations is a sustained loss of drilling fluid followed by rapid flowback, resulting in a pronounced reversible volume change in the wellbore. The intrinsic mechanism lies in the switching between fracture opening and closure triggered by the shift in the pressure differential between the wellbore and the formation. Parametric sensitivity analysis indicates that increasing wellbore pressure intensifies the breathing effect; formations with low fracture opening pressure, high porosity, and high permeability are more prone to severe breathing effects. Increasing the plastic viscosity and yield point of the drilling fluid can suppress the breathing effect, but careful management of the resulting increase in circulating friction and equivalent circulating density (ECD), which raises bottomhole pressure, is required. Field case calculations for a well in the Cameroon block show that, after improving the lost-circulation pressure calculation method to incorporate the breathing effect, the safe mud weight window can narrow by up to 0.03 g/cm3. This study advances the understanding of breathing effects in fractured formations and provides theoretical support for safe drilling within the narrow mud weight windows commonly encountered in such formations.

To address the challenges in predicting leakage rate and location during drilling in deep and complex formations, this study proposes a quantitative interpretation method for leakage conditions based on Physics-Informed Neural Networks (PINN). By utilizing automatic differentiation techniques, the method employs neural networks to solve the wellbore hydraulics model under leakage conditions, enabling intelligent and accurate predictions of both leakage rate and location. The neural network takes time and well depth as inputs, while its outputs include annulus flow velocity, pressure, and leakage rate, with the leakage location treated as a trainable variable. The total loss function is composed of several components: the residuals of the mass and momentum conservation equations, prediction errors of dual-point pressure data obtained from a downhole dualmeasurement tool, prediction errors of wellhead pressure data, and errors derived from expert knowledge constraints. The Gradnorm algorithm is employed to assign weights to each loss term adaptively, and the neural network is trained by minimizing the total loss. Test results demonstrate that the neural network model trained with this approach can efficiently and reliably solve the wellbore hydraulics model under leakage conditions. Physical constraints are satisfied throughout the input-output process, achieving a mean relative error (MRE) of less than 10 % for leakage rate predictions and an absolute error (AE) within 20 m for leakage location predictions. Compared with methods such as the Unscented Kalman Filter (UKF) and Genetic Algorithm (GA), this approach, which leverages a global optimization strategy and intrinsic physical constraints, exhibits superior stability and accuracy across different noise levels. When integrated with the downhole dual-measurement tool, this approach provides critical guidance for leakage mitigation operations during drilling processes in deep and ultra-deep wells.

Well control is essential for safe offshore drilling. Traditional kill models design pre-operation parameters, lacking real-time data to correct deviations, reducing accuracy under sudden influxes or operational errors. To address this, this paper proposes a real-time wellbore flow state inversion framework for kill operations based on the H-infinity filter. Using a transient gas-liquid multiphase flow model within a game-theoretic structure, the framework performs closed-loop correction of three real-time observations: casing pressure, standpipe pressure, and pit gain. These corrected observations act as updated boundary conditions to reconstruct the wellbore pressure field and gas holdup distribution. Combined with safety constraints like bottomhole pressure balance and surface equipment limits, a real-time kill plan optimisation method was established. Validated with wait-and-weight method observation data from Well Y, the framework achieved mean absolute errors of 0.028 MPa, 0.118 MPa, and 0.029 m3 for inverted casing pressure, standpipe pressure, and pit gain, respectively. Hold-out, sensitivity, and variability analyses confirmed robust reproducibility. Applying the optimised plan reduced peak pit gain by up to 20.3% and improved gas discharge efficiency by up to 76.8% during critical stages. This framework supports dynamic decision-making in complex offshore well control operations, helping prevent uncontrolled blowouts and protect marine ecosystems.

CO2 injection alters the evolution of the subsurface pressure regime, and overpressure within the caprock compromises caprock integrity, thereby diminishing storage performance and potentially posing environmental risks. Based on the dimensions of the GTZ secondary anticline within the Changyuan Anticline, a three-dimensional saline-aquifer storage model with an explicit anticline geometry was constructed. The analysis evaluated the effects of engineering parameters and anticline size on the spatiotemporal evolution of caprock overpressure and on the timing of hydraulic sealing failure, and examined the effectiveness of coupled operational schemes for controlling caprock overpressure. Results show that sustained CO2 injection increases pore pressure; buoyancy drives upward migration, focusing pressure at the caprock base and producing overpressure. A strategy combining a horizontal well, an extended injection interval, reservoir stimulation, and intermittent injection effectively reduces caprock overpressure. For anticline settings, selecting saline aquifers with a shorter major-axis length and a larger closure height further lowers overpressure and increases safety margins. Under an optimized scheme, the annual injection rate can be increased from 0.60 Mt/a to 0.75 Mt/a. Injection design should be tailored to the demonstrated caprock integrity in geological storage projects.

The proper proppant placement in complex fractures is significant in improving the production of unconventional reservoirs. In the past, many studies assumed that the fractures have smooth surfaces, without considering the effect of the surface roughness on the proppant transportation and deposition in complex fracture models. In this work, a realistic complex fracture model with rough surface is considered. The fracture model with rough surface consists of a primary fracture and a secondary fracture with an obtuse-angle bend of 120 degrees. Meanwhile, the differences of proppant transport laws between rough and smooth fractures are studied in depth. The experimental results show that for the low pump rate (below 83 mL/s), the roughness does not affect the proppant distribution much in the fracture. However, higher pump rate (e.g., more than 83 mL/s) can result in large empty areas in the primary fracture. To solve this problem, larger proppant with a low pump rate should be injected to effectively fill in the empty area. In the secondary fracture, the roughness can increase the proppant transportation efficiency. The experimental findings are useful to understanding the proppant transportation and deposition efficiency in complex fractures with rough surfaces.

In managed pressure drilling (MPD), wellbore flow states cannot be obtained in real time, so kick intervention decisions rely on the empirical judgment of engineers, which introduces a significant lag. The central hypothesis of this study is that fusing a physics-constrained transient two-phase flow model with real-time surface measurements through data assimilation can reconstruct the unobservable downhole flow state and, on this basis, enable quantitative and earlier prediction of the safe intervention timing than empirical judgment alone. To this end, this paper proposes a method for real-time inversion of wellbore flow states and safety intervention timing prediction based on the Ensemble Kalman Filter (EnKF). Using a transient wellbore gas-liquid two-phase flow model as the EnKF model operator, the method continuously assimilates real-time casing pressure, standpipe pressure (SPP), and pit gain data. This process dynamically corrects model prediction bias while maintaining multiphase flow physical constraints. Thus, the method achieves high-precision dynamic inversion of wellbore pressure profiles and gas holdup distributions. On this basis, the authors use the inverted states as initial conditions to calculate safety casing pressure with the multiphase flow model. The method then predicts intervention timing by combining three trigger conditions: safety casing pressure, pit gain, and the density difference between the inlet and outlet. The authors validated the method using kick scenarios from Well L and Well Z in the Shunbei block. The results showed that the mean absolute errors (MAEs) for casing pressure inversion were 0.113 MPa and 0.135 MPa, respectively. The MAEs for SPP were 1.324 MPa and 0.954 MPa. The MAEs for pit gain were 0.174 m3 and 0.114 m3. The inverted spatiotemporal distribution of gas holdup reflected the entire process of gas migration and expansion in the wellbore. Prediction results for intervention timing showed that the method issued early warning signals approximately 53 min and 29 min earlier than actual field operations. This method provides a quantitative decision-making basis with safety redundancy for MPD field operations.

Slickwater is widely used in shale gas fracturing owing to its low cost and proppant-transport capacity, but large volumes of fluid retained in rough fractures can cause persistent conductivity loss. The contributions of "polymer clogging" and "hydrodynamic mixing" to this degradation are still unclear. This study investigates rough fractures from Longmaxi shale using core-flooding experiments combined with a "no-clogging" numerical model. Slickwater concentration, fracture roughness, proppant size, and confining pressure are varied, and conductivity evolution during formation-water flushing is evaluated using residual permeability. A three-dimensional laser-scanned fracture geometry is incorporated into a coupled Darcy-flow and dilute-species transport model including only roughness-induced hydrodynamic mixing. Breakthrough and residence-time distribution curves are used to obtain mean residence time, variance, and skewness. Experimentally, increasing slickwater mass fraction from 0.1% to 0.7% reduces final residual permeability by over 40% and lowers permeability recovery from similar to 85% to <45%; increasing joint roughness coefficient from 3.2 to 19.8 further decreases residual permeability by similar to 25%, whereas coarse proppant increases it by similar to 35%, and high confining pressure amplifies damage. In contrast, in the no-clogging model, higher injection rate and roughness enhance mixing, shorten mean residence time, and intensify long-tail behavior. The opposite responses demonstrate that fracture conductivity degradation in actual rough fractures is dominated by polymer clogging and aperture constriction, while potentially beneficial mixing is largely suppressed. For highly rough, strongly confined shale reservoirs, fracture cleanup should therefore focus on reducing polymer loading, selecting relatively coarse proppants, and optimally matching flowback drawdown and rate to mitigate irreversible conductivity loss.

Abstract To elucidate the generation mechanism of acoustic emission signals from PE gas pipeline leaks and advance long-distance detection technology for such leaks based on acoustic emission, this study established a three-dimensional model of PE gas pipeline leaks using COMSOL based on energy conservation theory. The fluid-solid coupling interface was adopted to simulate the stress-strain field and energy changes at the leakage orifice under different pressure and leakage aperture parameters. The results show significant stress concentration around the leakage hole. Elastic strain energy accounts for more than 95% of the energy storage density. Energy dissipation mainly occurs as elastic strain energy, which fluctuates with the input, dissipation, and transfer of energy during the initial stage of leakage. It is the release of this energy that forms acoustic emission signals. The shape of the leakage hole under different parameters affects the stress and energy distribution. This study reveals the generation mechanism of acoustic emission signals during leakage. It provides a theoretical basis for optimizing signal recognition and improving detection accuracy.

Cement plugging materials for offshore abandoned wells suffer severe cracking and poor hydro-mechanical sealing performance. Sn58Bi low-melting alloy is a promising alternative owing to its low melting point, slight solidification expansion and superior fluidity, but intrinsic brittleness and segregated brittle Bi-rich phases weaken its long-term downhole reliability. Herein, Cu (0.5–2.0 wt.%) was added to modify Sn58Bi alloy. Uniaxial tensile, static bearing and gas tightness tests were conducted at 30, 60 and 90 °C, coupled with optical microscopy and SEM to observe microstructures, fractures and interfaces. Based on SEM observations and previous reports on Sn-Bi-Cu systems, Cu addition was considered to promote the formation of Cu₆Sn₅ intermetallic compounds during solidification.Moderate Cu₆Sn₅ grains refine Bi-rich phases, break brittle phase continuity and block crack propagation. At 90 °C, Sn58Bi-1.5Cu achieves 15.11% elongation; at 30 °C, Sn58Bi-1.0Cu has 281.5% higher bearing capacity than pure Sn58Bi, while Sn58Bi-0.5Cu exhibits the best gas tightness with 47% higher breakthrough pressure. Excessive Cu causes agglomerated coarse Cu₆Sn₅-related particles, inducing interfacial stress concentration and deteriorating overall performance. This study verifies that proper Cu microalloying eliminates brittle failure of Sn58Bi and imsuggests its applicability for offshore abandoned well casing plugging.

Accurate prediction of drilling fluid rheological parameters under high-temperature and high-pressure (HTHP) conditions is critical for reliable drilling hydraulics and wellbore pressure control in deep and ultra-deep wells. However, most existing empirical and semi-empirical rheological models are developed for limited temperature-pressure ranges and specific fluid formulations, which restrict their applicability and accuracy under HTHP conditions. In this study, systematic rheological experiments were conducted on multiple drilling fluid systems over wide temperature-pressure ranges (20-200 degrees C and 0.1-200 MPa). Based on the experimental data, a unified predictive model for key rheological parameters was developed using a symbolic regression (SR) algorithm. The model performance was evaluated using standard statistical metrics and compared with commonly used conventional models. Compared with conventional models, the proposed model shows stronger applicability for predicting the rheological parameters of the investigated oil-based and water-based drilling fluids over a wider temperature-pressure range. It effectively overcomes the limitations of existing models under HTHP conditions (150-200 degrees C and 80-200 MPa) and demonstrates improved prediction accuracy and robustness for both high- and low-density drilling fluids. The overall prediction errors are generally within approximately 10%. The results indicate that the proposed unified model provides a reliable and computationally efficient tool for predicting drilling fluid rheological parameters under HTHP conditions, facilitating its integration into wellbore hydraulics, wellbore pressure, and equivalent circulating density calculations in deep and ultra-deep well applications.

This paper aims to solve the problem of Sn-xBi alloy plugging in perforations at different ambient temperatures associated with varying formation depths in oil and gas wells. A method is proposed to investigate the influence of ambient temperature on the properties of Sn-xBi alloys. A dedicated experimental device is developed to test the mechanical pressure-bearing capacity and liquid sealing performance of Sn-xBi alloys with different component ratios in rock perforations. Mechanical sealing tests and hydraulic sealing tests are performed on rock perforations plugged with Sn-xBi alloys. The effects of alloy composition on the sealing performance of rock perforations at different temperatures are studied, and the underlying microstructural mechanisms responsible for these effects are analyzed. Experimental results show that at ambient temperatures of 30 ℃, 60 ℃, and 90 ℃, both the mechanical bearing strength and hydraulic sealing strength decrease with increasing temperature. The plugging performance of Sn-xBi alloys in rock perforations is affected by both the Sn/Bi content ratio and the ambient temperature. Among all tested compositions, the Sn-58Bi alloy exhibits the optimal mechanical pressure-bearing performance and hydraulic sealing performance. At 30 ℃, the maximum hydraulic sealing pressure of pure Sn is 0.79 MPa, while that of pure Bi is 1.04MPa, representing a 25% improvement in hydraulic sealing over pure Sn. Microstructural analysis reveals that a higher Bi content in the Sn-xBi alloy enhances its expansion behavior, which exerts a positive effect on plugging performance. However, excessive micro-expansion also leads to surface roughness, which can adversely affect the final sealing quality.

The wellbore temperature field under the co-existence of kick and loss (CKL) conditions is influenced by kick influx, fluid loss, and wellbore-formation coupled heat transfer, resulting in a complex heat exchange mechanism. Existing studies primarily focus on single operating conditions, lacking a systematic analysis. This study develops a transient heat transfer model, incorporating convective heat transfer and variable mass flow. The measurement temperature data from the Qiongdongnan Basin is used to validate the developed model, with an error within 3

Severe drill string vibrations, particularly stick–slip, significantly compromise drilling efficiency and tool longevity in deep hard formations. Compound percussive drilling (CPD) has emerged as a promising technique to mitigate these vibrations and enhance the rate of penetration (ROP). However, the complex coupling mechanisms between impact loads and bit dynamics remain insufficiently understood. This study aims to elucidate the axial–torsional vibration characteristics of the drill bit and the underlying vibration reduction mechanisms under CPD conditions. A multi-degree-of-freedom (MDOF) dynamic model was first established, integrating both the dynamics of the CPD tool and the regenerative cutting effects inherent in bit–rock interactions. The governing equations were then solved numerically using the fourth-order Runge–Kutta method, followed by a systematic parametric sensitivity analysis to quantify the influence of impact parameters on vibration mitigation. The results show that while CPD induces detrimental axial–torsional vibrations in soft rock formations, it effectively suppresses stick–slip and enhances ROP in hard rock formations. Notably, coupled axial–torsional impact loading exhibits superior vibration suppression capabilities compared to singular axial or torsional impacts. A critical proportional relationship for parameter optimization was identified; specifically, maximizing vibration mitigation requires scaling the axial impact load proportionally with the torsional impact load. For example, when the axial impact load amplitudes are 5 kN and 10 kN, the corresponding optimal torsional impact load amplitudes are approximately 500 N·m and 1000 N·m, respectively. Furthermore, maintaining the impact frequency within the range of 10–30 Hz yields optimal vibration reduction effects. The benefits of CPD become increasingly pronounced with higher rock strength and longer drill strings. These findings confirm the suitability of CPD technology for deep hard rock environments and provide theoretical guidelines for the optimal selection of impact parameters in engineering applications.

Proppant flowback during the flowback phase after hydraulic fracturing in coal reservoirs critically impacts fracture conductivity and wellbore integrity. However, experimental studies on its critical conditions and controlling mechanisms within coal's complex fracture networks are scarce compared to sandstone or shale. This study conducted physical simulation experiments using outcrop coal samples from the XD block in China and a modified fracture conductivity system. By establishing a determination method for the critical backflow rate (Q(c)), the dynamic evolution process of proppant backflow-characterized by the stages of initial stability, critical instability, severe backflow, and re-equilibration-was revealed. The influences of proppant size, flowback fluid viscosity, proppant concentration, and effective stress on Q(c) were systematically analyzed, and the relative weight of each influencing factor was quantified through orthogonal experimental design. Results show that proppant backflow initiates and concentrates preferentially at the fracture outlet region, implying a higher risk of proppant failure in the near-wellbore fracture section. The Q(c) decreases with reducing proppant size, increasing flowback fluid viscosity, increasing proppant concentration, and decreasing effective stress, among which effective stress is identified as the dominant controlling factor. Furthermore, no necessary correlation is observed between Q(c) and the critical backflow ratio, suggesting that the initiation threshold and post-instability flowback intensity are governed by different mechanisms. This work provides experimental data and a quantitative basis for optimizing flowback strategies in coal reservoir fracturing operations.

Hydraulic fracturing, as a critical technology for reservoir stimulation, has functioned as a central stimulation technique in the large-scale development of shale gas and coalbed methane. During hydraulic fracturing operations, the movement and distribution of proppants within fractures serve as the key determinant of both fracture conductivity created by fracturing and long-term sustained production performance. Meanwhile, the settling and packing mechanisms of proppants greatly affects proppant transportation efficiency and distribution within fractures, which in turn determines the level of oil and gas production. However, current research on settling and packing mechanisms of proppants has primarily focused on spherical proppants, while some aspects of the non-spherical shaped proppants with different shapes and structures have not yet been fully considered. In this study, 3D printing technology was utilized to fabricate various non-spherical proppants, including cube, rhombohedron, tetrahedron, cuboid, and cylinder shapes. Compared to spherical proppants, these non-spherical counterparts were examined in terms of their settling, transport, and packing mechanisms using visualized proppant settling and transport setup. The result show that proppants with irregular shapes, such as rhombohedrons and tetrahedrons, experienced greater turbulent effects in the fluid, while proppants with regular shapes, such as cuboids, cubes, cylinders, and spheres, generated much less turbulence. For instance, under a low viscosity of 1 mPa·s, the settling velocity of spherical proppants was measured at 0.086 m/s, significantly higher than the 0.046 m/s for rhombohedron and 0.052 m/s for tetrahedron proppants. More importantly, the angle variation during the settlement process was much larger for the irregular-shaped proppants than for the regular ones. Moreover, the results also indicate that under low-viscosity conditions (< 3 mPa·s), non-spherical shape has a more significant effect on settling velocity. However, this effect becomes less apparent when the viscosity increases to the range of 6–9 mPa·s. Porosity measurements further revealed that rhombohedron and tetrahedron proppants achieved higher porosities (approximately 40–45%) compared to spheres and cubes (~ 35%), suggesting potential for enhanced fracture permeability. Finally, six new mathematical models were developed to predict the terminal settling velocity of different types of non-spherical shape proppants. The study of non-spherical proppants might provide a new idea for proppants application with new structures in the future.