The authors proposed two distinct prefabricated joint configurations for underground silos: a steel cavity joint and a concrete-filled cavity joint. They subjected full-scale specimens incorporating these joints to axial compression tests to quantify load-bearing capacity, deformation patterns, yielding behavior, and failure modes. The study team complemented physical testing with three-dimensional nonlinear finite-element simulations using ABAQUS to clarify internal load-transfer mechanisms that are difficult to measure directly in the laboratory.
Test protocols included controlled axial loading of full-scale specimens. The source reported that the experiments were intended to observe the progressive response from the elastic stage through yielding to ultimate failure; specific loading rates, boundary conditions, and specimen dimensions are described in the original manuscript (the abstract did not reproduce those numerical details). The combined experimental-and-numerical approach enabled cross-validation of observed failure phenomena and offered a means to explore parametric effects beyond the tested specimens.
Testing revealed distinct failure processes tied to joint configuration. The specimen with the steel cavity joint primarily failed because the external waterproof steel plate became unstable under axial compression. During the elastic stage, investigators reported that the steel plate contributed approximately 41% of the applied load. Deformation and instability of that plate governed the subsequent loss of capacity and the observed failure sequence.
The concrete-filled cavity joint behaved differently. Data indicated the cavity concrete carried roughly 75% of the applied load during the elastic stage, and the specimen ultimately failed by crushing of prefabricated concrete blocks. The authors observed a notably higher initial stiffness in the concrete-filled configuration; measured peak load capacity increased by more than 28% relative to the steel cavity joint. These contrasts define practical tradeoffs between relying on a metallic waterproofing liner for early-stage stiffness versus using a structural concrete infill to share axial demand.
The study team used nonlinear finite-element analysis in ABAQUS to reproduce test responses and to probe internal force paths. Numerical simulations aimed to mirror the experimental load–deformation curves and to visualize stresses and instabilities that appeared during testing. Data indicated good qualitative agreement between tests and models in terms of failure location and sequence; the paper provides fuller model parameterization and validation plots not reproduced in the abstract.
Modeling clarified how load was partitioned between the steel plate, cavity infill, and the prefabricated concrete blockwork at different stages of loading. For the steel cavity joint, simulations showed the steel plate carrying a substantial fraction of load in the elastic stage and then losing contribution as instability developed. For the concrete-filled joint, the cavity concrete remained a dominant load-carrying element until crushing of prefabricated concrete elements initiated the final failure.
Investigators examined how key joint design parameters influenced axial compressive performance. The primary comparative findings were: the concrete-filled cavity produced higher initial stiffness and greater ultimate capacity than the steel cavity; the steel plate dominated early-stage load share in the steel-cavity design; and the cavity concrete was the main load-bearing material in the concrete-filled design. The source reported that the concrete strength grade used inside the cavity can be reduced appropriately to optimize material utilization efficiency, although the abstract did not list specific recommended grades or stepwise optimization results.
These parameter observations point to design strategies: if early stiffness and higher peak capacity are priorities, the concrete-filled cavity offers advantages; if the design targets lighter infill or relies on metallic waterproofing systems, the steel cavity configuration will behave differently and may require additional stiffening measures to avoid plate instability.
The authors documented technical limitations and proposed areas for future research. The source indicates that further work is needed to refine joint designs, to expand parametric studies beyond the cases tested, and to translate laboratory and numerical findings into standardized design guidance for prefabricated underground silos. The abstract and manuscript note limitations of the present tests and models; specific unresolved questions and recommended next steps are outlined in the paper's Technical limitations and future research section. The article ends with these concrete research needs rather than a broad, synthetic wrap-up.