The ribbed steel truss composite slab forms the floor system by casting a composite concrete layer atop a lightweight precast ribbed base slab. The ribbed base slab improves the distance between the top surface of the base slab and the upper chord reinforcement by reducing the thickness of the precast base slab, thereby facilitating rebar penetration and pipeline installation. Meanwhile, concrete ribs are arranged along the upper chord reinforcement to enhance the stiffness of the precast base slab and compensate for the stiffness reduction caused by the decreased slab thickness [21]. The construction details of the ribbed base slab are illustrated in Fig. 1.

The experiment was conducted using the thickness of the base slab, truss spacing, and presence or absence of concrete ribs as design variables. Two precast reinforced truss concrete composite base slab specimens (YZB1 and YZB2) and one precast ribbed reinforced truss concrete composite base slab specimen (DLB1) were fabricated. The specimen dimensions were 3000 mm × 600 mm, which are commonly used in practice. The top chord reinforcement had a diameter of 10 mm, while the bottom chord reinforcement and the longitudinal reinforcement in the base slab had diameters of 8 mm. The truss web reinforcement and the transverse reinforcement in the base slab had diameters of 6 mm, with a transverse spacing of 600 mm. The top and bottom chord reinforcement, as well as the transverse and longitudinal reinforcement, in the base slab were all HRB400 hot-rolled ribbed bars, while the web reinforcement was HPB300 hot-rolled plain bar. The concrete strength grade for all specimens was C30. The main parameters of each specimen are presented in Table 1, and the schematic diagram of the reinforcement layout is shown in Fig. 2.

The composite base slab for the experiment was formed in a single pour of fine aggregate concrete. After curing was completed, three standard cubic specimens were cast during the pouring process to measure their compressive strength. Three reinforcement samples of different diameters were selected to test the mechanical properties of the reinforcement used in the specimens. Table 2 lists the compressive strength of the concrete cube specimens, and Table 3 lists the mechanical properties of the reinforcement, all of which were obtained from tests.

According to scholars’ research [13,28], the experiment was carried out using a heavy load stack to achieve a homogeneous loading process on the specimen. The experiment was loaded incrementally until the specimen was damaged, and a single concrete block with a mass of about 25 kg was chosen for the stacking weight. Considering that the upper surface of the base slab specimen has exposed steel trusses and concrete ribs, making direct loading inconvenient, wooden blocks and boards were used to adjust the height, as shown in Figs. 3 and 4. When the following signs appear, the specimen is considered to have reached its maximum load capacity, and loading should be terminated: 1) the width of the cracks in the bottom slab reached 1.5 mm; 2) the maximum deflection reached 1/50 of the support span; 3) the concrete in the pressure zone was crushed during loading; 4) the rebar in the top chord of the steel truss was bent. The experimental site is shown in Fig. 5.

This experiment monitored the specimens’ deflection changes, crack development, and load-carrying capacity. The number of displacement sensors was five and were placed at the two end supports, at 1/4, 1/2, and 3/4 spans of the test slab, as shown in Fig. 3 below.

In the initial stage of the experiment loading, the deformation characteristics of the three specimens were consistent. With the gradual increase of external loading, the concrete of the bottom slab developed cracks, and the damage process and bearing capacity performance of different specimens exhibited distinct behaviors.

Specimen YZB1 (one reinforcement truss, base slab thickness of 60 mm) was observed with the first crack at the bottom of the slab when the external load was increased to 2.5 kN/m², near the center of the span, extending from the center of the slab width to a height of 40 mm at the side of the slab, with a crack width of 0.1 mm. When the load continued to be applied, no new cracks appeared in the base slab, but the existing cracks kept widening and propagating through the slab surface. When the load was increased to 3.89 kN/m², the specimen suddenly sank, and part of the weld joint between the top chord reinforcement and web reinforcement of the truss broke and buckled (Fig. 6a), and the experiment was stopped. The distribution of cracks at the bottom of specimen YZB1 is shown in Fig. 7a.

The first crack in specimen YZB2 (two reinforcement trusses, base slab thickness of 30 mm) was observed near the span of the bottom of the slab when the load was increased to 5 kN/m², extending from the center of the bottom of the slab to the edges on both sides, with a crack width of 0.05 mm. Continue loading, the number of cracks within the span increased, and they developed in an interlacing pattern from the bottom of the base slab to the side of the base slab. The crack width and length increased. When the external load increased to 8.47 kN/m², the top chord reinforcement of the truss buckled (Fig. 6b), and the test was terminated. The distribution of cracks at the bottom of specimen YZB2 is shown in Fig. 7b.

The first crack in the ribbed composite slab base slab specimen DLB1 (one reinforcement truss, base slab thickness of 30 mm) was observed at the bottom of the slab at mid-span and along the slab width when the external load was increased to 6.39 kN/m², with a crack width of 0.1 mm. With continued loading, new cracks appeared in the span, and the existing cracks widened and ran up the side of the slab through the surface (Fig. 6c). When the load was applied to 12.92 kN/m², the upper concrete ribs were crushed (Fig. 6d), and the experiment was terminated. The distribution of cracks at the bottom of specimen DLB1 is shown in Fig. 7c.

The load-deflection curve shown in Fig. 8 reflects the mid-span displacement changes of each specimen as the load increases, along with their respective variation values. Table 4 lists the calculated short-term stiffness and cracking load for three specimens. Short-term stiffness means the bending stiffness of the base slab during the construction of cast-in-place concrete for composite slabs, while the cracking load is defined as the load value at which the first crack is observed in the base slab during experimental loading. As shown in the figure, the bearing capacity performance and span deflection development trend of the three specimens are not the same. Loading commenced, revealing markedly different initial stiffnesses among the three specimens: DLB1 exhibited the highest stiffness, followed by YZB2 and YZB1. As the load continued, cracks appeared in the concrete at the bottom of the specimen base slabs. The contribution from the concrete in the bottom slab’s tension zone gradually diminished, causing a sudden drop in the section’s bending stiffness. The slope of the curve becomes smaller, and the inflection point appears after the cracking of the typical single-truss 60 mm thick base slab YZB1 and the ribbed concrete base slab DLB1.

In contrast, the double truss base slab YZB2 shows a steady increase in load and deflection. The increase in load and deflection is uniform. Specimen YZB1 developed cracks in the base slab, the concrete lost its contribution to the work, and the reinforcement was subjected to the upper stacked load. Finally, the reinforcement was bent and damaged, resulting in poor ductility. Before cracking, specimen DLB1, the role played by the concrete rib is minimal, and at this stage, the base slab primarily bears the compressive stress. When the base slab is subjected to tension and cracks, the slope of the curve becomes smaller, with the neutral axis moving upward. The load from the base slab is transferred to the ribs via the diagonal rebars of the truss. At this point, both the ribs and the base slab bear the load simultaneously. This results in an increased slope on the curve, though it remains smaller than the initial slope.

From Table 4, the cracking load of both specimen YZB2 and specimen DLB1 is greater than the design load (4.56 kN/m²), which meets the construction requirements. By comparison, the short-term stiffness of specimen YZB2 is 1.64 times the short-term stiffness of specimen YZB1. Its cracking load is twice that of YZB1. So the loss of stiffness due to reduced slab thickness can be compensated by decreasing the truss spacing (increasing the number of truss groups), while also enhancing the ductility of the base slab. The short-term stiffness of specimen DLB1 is 1.94 times that of specimen YZB2 and 3.19 times the short-term stiffness of YZB1, and its cracking load is 1.28 times that of the cracking load of YZB2 and 2.56 times that of the cracking load of YZB1. This is due to the setting of concrete ribs, which increases the concrete area in the compression zone of the base slab. This enhances stiffness and load-bearing capacity, effectively addressing the issue of excessive deflection in precast slabs during construction. The concrete ribs form a whole with the base slab through the reinforcement truss, and the joint force effect is remarkable.

According to the above experimental study, it can be seen that the ribbed base slab demonstrates optimal force performance, with the cracking load being higher than the design load, thus meeting the construction requirements. The additional concrete ribs effectively improve the flexural stiffness of the precast base slab. Considering the differences between a ribbed base slab and an ordinary precast base slab, combined with the experimental results in the paper and with reference to the design method of ordinary concrete members [29], the following three assumptions are made when carrying out the calculation of the positive section bearing capacity: (1) the section strain conforms to the flat section assumption; (2) the contribution of the tensile concrete to the section stiffness is considered before the base slab cracks; (3) the web reinforcement only plays the role of transferring force, and its contribution to the section stiffness is ignored.

Cracking Bending Moment

The experimental results show that the ultra-thin ribbed base slab (DLB1) is superior to the ordinary single truss base slab (YZB1) and the ultra-thin double truss base slab (YZB2). Additionally, the beneficial effect of ribbed concrete on a precast base slab can be supplemented by the calculation of the cracking moment, which is calculated separately for three base slab specimens (taking the mid-span moment) for comparison below. Referring to the relevant provisions in the Code for the Design of Concrete Structures (GB 50010-2010) [29], the cracking moment of reinforced concrete flexural members was calculated by Eqs. (1) and (2). (1)Mcr=ftkW0 (2)W0=I0y0

In the above equation, Mcr is the cracking moment of reinforced concrete bending members; ftk is the standard value of concrete tensile strength; W0 is the cross-sectional resistance moment at the tensile edge of the converted section of the member; I0 is the moment of inertia of the converted section of the member; y0 is the height of the neutral axis of the member.

The reinforcement in the specimen is converted into the equivalent cross-sectional area A0, according to the ratio of the elastic modulus of reinforcement to concrete (∂E=ES/EC). The base slab is regarded as a homogeneous elastic material to calculate the height of the compressive zone x0, the moment of inertia of the converted cross-section I0, and the height of the neutral axis y0.

The calculated cracking moments were compared with those obtained from the experiments, and the results are listed in Table 5. The reason why the experimental value is larger than the calculated value is that the first crack observed in the experiment does not appear during the loading holding phase but during the application of the load, so the actual time when the crack is observed lags behind the cracking time of the base slab, making the recorded load slightly higher. As can be seen from the table, the calculated values are closer to the experimental values and do not affect the analysis of the results. It is obvious that the cracking moment of the ribbed base slab specimen DLB1 is substantially higher compared to that of the ordinary precast base slab specimen YZB1.

Establish a numerical analysis model for the parameters of experimental specimen DLB1. The model of the ribbed base slab established in the ABAQUS software is shown in Fig. 9. The numerical analysis primarily references the relevant recommendations in GB 50010-2010 [29], and related research works [30,31]. For the model of concrete, the concrete damaged plasticity (CDP) model C3D8R solid element is selected (see Fig. 10). For reinforced structures, the double-line model T3D2 truss element is used. The concrete and the rebar have a strong bonding force, so the structural rebar of the base slab and the lower chord rebar of the steel truss are embedded in the concrete base slab using “Embedded” restraint, and the upper chord reinforcement is also embedded in the concrete ribs using “Embedded” restraint. Within the model, it is assumed that no slip deformation occurs between the rebars and the concrete. In order to avoid stress concentration, RP reference points are used at both ends of the original precast base slab to make coupling contact with the support positions. The boundary conditions are set at the RP points, simply supported by constraints. The model is calculated using a “Standard” solver, and the model load-deflection curve is extracted and compared with the experimental data, as shown in Fig. 11.

It can be seen from Fig. 11 that the curve fits more closely in the early stage, the deformation trend is the same, the initial stiffness is similar, the later damage phase also aligns well, and there is a significant difference in the elastic-plastic deformation in the middle. This is due to the ABAQUS simulation, in which the uniform load is applied continuously and uninterrupted, without a holding phase. Therefore, the deformation performance is more continuous, while the test loading process involves holding the load for observation. Consequently, the experimental value curve is a stepped change. The difference between the principal structural relationship and the interaction between concrete and reinforcement, compared with the actual one, is one of the reasons for the error. However, the short-term stiffness of the ribbed base slab under the construction load (4.56 kN/m²) differed by less than 10% when comparing the short-term stiffness, cracking load, and ultimate load of the software simulation and test, as shown in Table 6. The data obtained from numerical analysis are used to perform parameter analysis on the base slab.

Based on the verification of the model accuracy, ten models were created to investigate the effects of three parameters: the top chord reinforcement diameter, concrete rib height, and rib width of the truss, on the short-term stiffness of the base slab. The detailed parameters of the models are shown in Table 7.