Capacity analysis of advanced bolt shear connectors in composite beams with finite element method using MIDAS FEA software

. Prefabricated construction is now familiar and widely used in various parts of the world. The important thing in the composite structure is the shear connection. Advanced bolt shear connectors are one of the innovations for prefabricated construction because their use is more efficient. The shear capacity of the advanced bolt shear connectors was carried out with the nonlinear finite element method using MIDAS FEA software and then compared with the results of laboratory testing and theoretical equations. There were 8 models to be tested with 2 variations of bolt diameters (12 mm and 16 mm) and 3 variations of bolt pretension (without pretension, 0.2P c pretension, 0.5P c pretension, and 0.8P c pretension). The results of the analysis show that the diameter of the bolts has a large effect on the shear capacity of the shear connectors where with the same bolt pretension, the shear capacity increases 71.4286% - 73.0769% on the test models with 16 mm bolts. Bolt pretension also affects the shear capacity of the shear connectors although the impact is small, 1.9231% - 7.6923% (12 mm diameter bolts) and 1.1111% - 6.6667% (16 mm diameter bolts) from the value of the shear capacity of the model without pretension. The most optimal pretension is 0.5P c for 2 bolt diameter variations. In the results of the analysis of the finite element method and laboratory tests, the deflection is directly proportional to the shear capacity of the models.


Introduction
For decades, steel-concrete composite beams (SCCBs) have been widely used in building and bridge structures.A composite beam is a type of beam that combines 2 or more materials with different characteristics therefore they can work together to become a single unit to withstand loads on a structure.This merger was carried out to take advantage of the advantages of each material to create a stronger and more efficient beam.Steel is a material that is strong in tension while concrete is a material that is strong in compression [1].Thus, if the two materials are combined into a composite unit, a new material will be obtained which has properties consistent with its constituent materials, strength in tension and compression [2].
However, composite structures will cause horizontal shear force during loading.This shear force will be resisted by several shear connectors to prevent slip during its lifetime.The shear connectors must be rigid enough to withstand the shear force that occurs.Shear connectors affect the steel beam elements by counteracting the shear forces that occur between the steel beams and the concrete slab [3].
Conventional shear connectors such as weldedheaded studs can fulfill the mechanical structural performance requirements for composite beams.
*Corresponding author: nicholekurniawan@gmail.comHowever, these connectors are not only welded to the steel but also fully embedded within the concrete block, resulting in structural composite beam elements that cannot be knocked down and cannot be efficiently reused.This is very influential, especially in prefabricated construction which has been built recently because the construction is more environmentally friendly due to the knockdown system.Therefore, to overcome this problem and increase the sustainability of construction, various types of bolted shear connectors have been developed and used as alternative methods in SCCBs to replace conventional connectors [4].
Advanced bolt shear connectors have been developed because they have several advantages over conventional bolt shear connectors.One of the advantages is using HPG (high-strength pouring material) as a grouting material to produce a stronger structural bond and using corrugated pipes that can strengthen the bonding mechanism between the postgrouted material and the precast slab.Pictures of several shear connectors can be seen in Fig. 1.
Therefore, to find out the capacity of advanced bolt shear connectors, the authors analyzed these connectors with different diameters and pretensions using the finite element method which will be carried out using the MIDAS FEA software.2 Methods and materials

Research methods
The following are the stages and research methods used by the author: 1. Conduct literature studies from various sources of articles, scientific journals, books, and other reliable sources related to this research.2. Collect and summarize the results of tests that have been carried out by Wang, et al. [4].3. Create and analyze computational models using FEM with the help of MIDAS FEA software.4. Comparing the results of the MIDAS FEA with the results of Wang et al. [4] laboratory tests and theoretical equations.5. Make conclusions to answer the objectives and formulation of research problems based on the results obtained in the analysis using MIDAS FEA.

Finite element model
This study consisted of eight models that will be developed using MIDAS FEA software to simulate the behavior of advanced bolt shear connectors.The dimensions, size, materials, and geometry of each model are the same as Wang, et al. [4] laboratory test as shown in Table 1 and Fig. 2.  Table 2 shows all the main parameters of the models.This study used two bolt diameters, 12 mm and 16 mm, each of which was distinguished by 4 variations of pretension (without pretension, 0.2Pc pretension, 0.5Pc pretension, and 0.8Pc pretension).Pc denotes the standard bolt pretension in GB 50017-2017 (50 kN and 90 kN for bolt diameters 12 mm and 16 mm).All model models studied by Wang, et al. [4] will also be analyzed in this study.However, there will be 3 additional models (SP-12, SP-16, and SP-16 0.8Pc) to complete the variation of bolt diameter and pretension.

Structural geometry modeling
Modeling of concrete slabs, HPG (high-strength pouring material), and steel beams were made with 3dimensional solid elements as shown in Fig. 3.

Meshing/discretization
After the modeling is complete, discretization is performed on the model.Discretization is done by dividing a structure into small elements so that it can be analyzed numerically.The smaller the mesh is made, the finer the results will be obtained.This modeling used a 10 mm mesh and the results are presented in Fig. 5.

Interface modeling
The interface model is carried out at the meeting of 2 different materials, as happened in the connection between the HPG and the concrete slab.The interface uses rigid parameters.

Bolt modeling
To model bolts with 1-dimensional elements, first link the elements to the bolt holes on the inside of the WF, between the WF and the HPG, and on the inside of the HPG.The master connector is made in the middle of the hole.After that, create a line element that is connected by the three master connectors.The results of bolt modeling can be seen in Fig. 6.

Pretensioning, loading, and boundary conditions
The bolt pretension is made by inserting a load in the form of force around the bolt hole as shown in Fig. 7, then in the middle of the bolt, a load in the form of pretension is used.The central load is given to the middle of the steel beam and rigid links are made throughout the top of the steel beam with a master connector in the middle.Pinned boundary condition was given at the bottom of the two concrete slabs.

Non-Linear Static Loading
Static non-linear loading is carried out by providing a load whose magnitude continues to increase until the structure reaches failure.This is done to determine the capacity of the structure to withstand the load.In MIDAS FEA software, the addition of non-linear loads will be carried out automatically during analysis by creating an analysis case.

Analysis and optimization
If the analysis results obtained are still not optimal in achieving the research objectives and expected, then optimization needs to be done.Optimization is done by remodeling to obtain optimal results.

Interpretation of results
From the analysis that has been carried out, the shear capacity will be obtained for each model.The results of this analysis will be compared with the results of laboratory tests and theoretical equations.Then conclusions can be drawn up based on the interpretations that have been made.

Failure modes and analysis result
The failure begins with a crack in HPG and spread to the concrete slab below it.All models have the same crack pattern as shown in Fig. 8.The results of the MIDAS FEA are summarized in Tables 3-4.Fu represents the shear capacity, Su represents the deflection of each model, and σ represents the tensile stress of the bolt.As shown in the tables, the results of the analysis with two different bolt diameters have the same pattern.Maximum shear capacity is obtained when the bolt pretension is 0.5 Pc.The deflection also has the same pattern as the shear capacity.This indicates that the increase in shear capacity effect the increase in deflection that the model can withstand.

Theoretical equation and laboratory test result
There are no specific design codes to estimate the bolt shear capacity of steel-concrete composite beams.However, there are theoretical equations that can be used as a design reference.
-In GB 50017-2017 [7], the shear strength of stud shear connectors is determined by Equation 1.
-Kwon et al. [13] proposed that the ultimate strength of the post-installed shear connector under static load can be found by the Equation 6. where

Comparison
A comparison of the results of the shear capacity analysis between MIDAS FEA, theoretical equations, and laboratory tests can be seen in Table 6 and Table 7.
There are 2 shear capacity values from laboratory tests and 6 shear capacity values from the theoretical equations.The results of theoretical equations vary due to different formulas and theoretical equations do not involve bolt pretension.As shown in the tables, the theoretical equation results of ACI (Equation 3), Eurocode (Equation 4), and Yang et al. (Equation 5) are the same and overall these results produce the smallest percentage difference from the MIDAS FEA analysis results.
The percentage of difference in the results of laboratory tests and MIDAS FEA analysis on models with 12 mm bolts is high.Meanwhile, models with 16 mm bolts are quite low.
Comparison graphs of the results of the MIDAS FEA analysis, laboratory tests, and theoretical equations can be seen in Fig. 9    As shown in Fig. 9, the pattern of the results between MIDAS FEA and laboratory tests are the same but the results values are much different.The closest value to MIDAS FEA results in models with 12 mm bolts is the theoretical results of GB 50017-2017 (Equation 1) and after that followed by theoretical results of ACI (Equation 3), Eurocode (Equation 4), and Yang et al. (Equation 5).
As shown in Fig. 10, the pattern of the results between test 1 and test 2 of the laboratory test is different.The pattern of the results of MIDAS FEA analysis is the same as laboratory test 1 but the results values are quite different.The closest value to MIDAS FEA results in models with 16 mm bolts is the theoretical results of ACI (Equation 3), Eurocode (Equation 4), and Yang et al. (Equation 5).
In Fig. 11, the pattern of the results of MIDAS FEA analysis between 12 mm and 16 mm bolts is the same.Even so, the difference is significant between the values of the 2 diameters.

Conclusion
Based on the results of the analysis and discussion, the conclusions obtained are as follows: 1. Bolt diameter is very influential on the shear capacity of advanced bolt shear connectors where which can be seen from the comparison of the Fu models with 12 mm and 16 mm bolts.With the same bolt pretension, the shear capacity increased 71.4286% -73.0769% on models with 16 mm bolts.2. Bolt pretension also increases shear capacity even though the impact is not too large, namely 1.9231% -7.6923% (12 mm diameter bolt) and 1.1111% -6.6667% (16 mm diameter bolt) of the shear capacity value of the models without pretension.3. The magnitude of the bolt pretension exerted on the bolt is not always directly proportional to the shear capacity as happened to the bolt with 0.8Pc pretension where there is a decrease in shear capacity from the bolt with 0.5Pc pretension.The most optimal bolt pretension is 0.5Pc for 2 bolt diameters.4. The value of deformation/deflection (Su) of the test models is directly proportional to the shear capacity of the test models.5.The bolt stress (σ) results of all models are still below the bolt tensile stress therefore failure did not occur in the bolts.The largest bolt stress was in model SP-16 0.8Pc, which was 8.7269% of the bolt tensile stress (fu).6.Failure did not occur in the concrete slab but in the HPG, so that when the composite structure was used again, HPG (grouting material) can be refilled without having to replace the concrete slab.
Structural materials include density, elastic modulus, fy, fu, and others.Concrete material modeling uses the total strain crack function with the brittle function as the tensile property and the Thorenfeldt function as the compressive property.The functions graph can be seen in Fig. 4. Steel materials use Von Mises failure model and bolt materials use the strain hardening function.
(12 mm bolt)  and Fig.10(16 mm bolt).Fig.11.Present a comparison graph of the shear capacity results of models with 12 mm and 16 mm bolts.

Fig. 9 .
Fig. 9. Graph of comparison of MIDAS FEA shear capacity results, laboratory tests, and theoretical equations in test models with 12 mm bolt connections.

Fig. 10 .
Fig. 10.Graph of comparison of MIDAS FEA shear capacity results, laboratory tests, and theoretical equations in test models with 16 mm bolt connections.

Fig. 11 .
Fig. 11.Graph of comparison of shear capacity in test models with bolt diameters 12 mm and 16 mm.

Table 1 .
Material properties of the test models.

Table 2 .
Variation of test models.

Table 3 .
Results of test models analysis with 12 mm bolts using MIDAS FEA.

Table 4 .
Results of test models analysis with 16 mm bolts using MIDAS FEA.

Table 5 .
Results of laboratory test.

Table 6 .
Percentage of differences in the results of MIDAS FEA analysis, theoretical equations, and laboratory tests on test models with 12 mm bolts (%).

Table 7 .
Percentage of differences in the results of MIDAS FEA analysis, theoretical equations, and laboratory tests on test models with 16 mm bolts (%).