Chen Lianjun1，2， Gao Kang1，3， Sun Zhenjiao1， Ma Hui4，5， Cai Guanghui1
1.College of Safety and Environmental Engineering， Shandong University of Science and Technology， Qingdao 266590， China； 2.School of Civil Engineering， Qingdao University of Technology， Qingdao 266520， China； 3.School of Civil and Environmental Engineering， Nanyang Technological University， Singapore 639798， Singapore； 4.School of Safety Engineering， China University of Mining and Technology， Xuzhou 221116， China； 5.Department of Mechanical Engineering， National University of Singapore， Singapore 117575， Singapore
Abstract
Significance Fresh concrete is a highly concentrated multiphase granular suspension composed of aggregates， cement paste， admixtures， and entrapped air. Its flow behavior directly affects mixing uniformity， formwork filling， passing ability， pumping stability， shotcrete transport， extrusion quality， and the buildability of three-dimensional concrete printing. Conventional workability and rheological tests provide useful macroscopic indicators， but they cannot fully reveal particle migration， force chain formation， paste film lubrication， aggregate bridging， fiber orientation， or time-dependent structural build-up. Therefore， numerical simulation offers an important approach for linking material composition， mesoscopic particle interaction， and engineering-scale flow performance. Among existing methods， discrete element method （DEM） is particularly valuable because it explicitly tracks particle motion， collision， friction， cohesion， and rearrangement. With advances in contact mechanics， computing power， and multiphysics coupling， DEM has gradually evolved from a qualitative tool for reproducing simple workability tests into a physically interpretable framework for predicting complex fresh concrete construction processes. This review summarizes the model system， physical assumptions， calibration logic， application boundaries， and future development trends of DEM-based simulation of fresh concrete flow.
Progress This study reviews fresh concrete DEM from the perspective of powder and granular mechanics. Basic contact laws， including linear spring-dashpot， Hertz-Mindlin， cohesive damping， and JKR adhesion models are first discussed in relation to macroscopic rheological parameters. These models can describe collision， damping， frictional sliding， and equivalent cohesion， and are efficient for slump， flow spread， L-box， and small-scale filling simulations. However， since fresh concrete contains paste-coated aggregates rather than dry particles， direct use of dry granular contact laws may underestimate adhesion， yield resistance， and lubrication effects. Therefore， paste bridge and liquid bridge models introduce finite tensile strength， rupture distance， viscous dissipation， and paste film thickness between adjacent aggregates. Hard-core soft-shell models further regard aggregates as rigid cores wrapped by deformable mortar shells， thereby representing mortar film lubrication， blocking， and passing behavior. Kneading contact models introduce moisture or paste distribution as internal variables and are suitable for simulating water addition， liquid migration， mixing， and homogenization. Discrete fresh concrete models embed mortar constitutive behavior into the contact zone between aggregates， allowing contact forces to be calculated from the strain， strain rate， and time-dependent response of the equivalent mortar region. These developments show that DEM modeling is shifting from empirical fitting of contact parameters toward measurable， rheology-related， and history-dependent interaction laws. This study also discusses the effects of particle morphology， fiber orientation， and thixotropy. Real aggregates are angular， elongated， or flaky， and their interlocking can increase dilatancy， shear resistance， and local blocking probability. Multi-sphere clumps， random polyhedra， CT-reconstructed particles， and shape-factor corrections provide different compromises between geometric fidelity and computational efficiency. For fiber-reinforced mixtures， DEM can track fiber orientation and alignment during pumping， spraying， and printing， while flexible fiber models are required when bending deformation is significant. Time-dependent hydration， flocculation， and structural build-up further require evolving contact parameters or constitutive models， especially for printable concrete and shotcrete. In terms of application， DEM and its coupled methods have been used to analyze mixing uniformity， self-compacting concrete flow， reinforcement passing， local blockage， pumping pipe flow， pneumatic conveying， shotcrete transport， and concrete 3D printing. In mixing simulations， DEM can quantify particle trajectories， collision frequency， mixing index， and torque evolution. In self-compacting concrete， DEM reveals that blockage is controlled not only by average fluidity， but also by local aggregate concentration， arching， force chains， and insufficient paste lubrication near constrained openings. For pumping and shotcrete conveying， pure DEM is insufficient to resolve the continuous paste or air pressure field， so coupled computational fluid dynamics （CFD）-DEM is important for predicting pressure loss， particle migration， lubrication layer formation， wall collision， and bend wear. Coupled smoothed particle hydrodynamics （SPH）-DEM is more suitable for free surface flow， nozzle extrusion， interlayer deposition， and large deformation. Coupled DEM-finite element method （FEM） is useful when particle-scale deposition or rearrangement must be transferred to early-age structural deformation， formwork pressure， or printed layer stability. These methods are complementary， and their selection should depend on the dominant physical field， boundary condition， target output， and available validation data.
Conclusions and prospects DEM-based simulation of fresh concrete has gradually shifted from empirical reproduction of single tests to predictive modeling constrained by particle mechanics， paste rheology， and multisource data. Its major value lies in connecting microscale interaction mechanisms with engineering indicators such as workability， passing ability， pumping pressure， blockage risk， wear intensity， extrudability， and buildability. However， several challenges remain to be addressed. Contact parameters are often back-calculated from macroscopic tests， causing nonuniqueness and limited transferability. Particle scaling may change the Stokes number， Froude number， contact network， and cohesion scale. Spherical particles may underestimate angular interlocking. Implicit paste treatment may lose local pressure fields and lubrication layer details. Hydration， thixotropy， temperature， admixture effects， wall roughness， and experimental variability also introduce uncertainty， while engineering-scale simulation remains computationally demanding. Future research should establish a measurable parameter-driven calibration framework combining rheometry， paste bridge tests， particle-paste shear tests， standard workability tests， and process-scale validation. Coarse graining， GPU acceleration， realistic particle morphology， flexible fibers， time-dependent contact laws， machine learning inversion， and field monitoring data should be integrated to build a virtual concrete laboratory for material design， process optimization， and intelligent construction control.
Keywords: fresh concrete； numerical simulation； discrete element method； rheology； three-dimensional concrete printing； pumping
Get Citation: Chen Lianjun， Gao Kang， Sun Zhenjiao， et al. Research progress on numerical simulation of fresh concrete flow［J］. China Powder Science and Technology， 2027， 33（3）： 1-18.
Received: 2026-05-16， Revised： 2026-07-23， Online： 2026-10-21。
Funding: The research was supported by the National Natural Science Foundation of China (Grant No. 52574253); Postdoctoral Innovation Project of Shandong Province (Grant No. SDCX-ZG-202502047); China Scholarship Council Fund (Grant No. 202508370048).
DOI：10.13732/j.issn.1008-5548.2027.03.009
CLC No.: TU528； TB4
Type Code: A
Serial No.: 1008-5548（2027）03-0001-18
