PSI - Issue 78

Gaetano Della Corte et al. / Procedia Structural Integrity 78 (2026) 199–206

205

3. Case studies 3.1. Geometry and finite element models

The geometry of two connection samples is illustrated in Fig. 3. The same member is supposed to be connected to a concrete substrate by means of two plates with different sizes and stiffener arrangements. The plate steel grade was supposed to be S 275, the bolt grade = 10.9, the concrete grade = C25/30. The length of the anchors indicated in the figures was calculated as the elongation length specified by EN 1993-1-8 (or the “stretch length” defined by EN 1992-4), i.e., 8 d b with d b = diameter of the anchor. Using the plastic resistance calculation method outlined in section 2.4, the moment resistance was calculated with the axial force varying from zero to the axial force capacity in compression. The consequent N - M interaction diagrams helped in selecting values of the applied axial forces, which are specified in the following section. The finite element models (FEM) were built using Ansys ®. The numerical models were developed consistently with the assumptions made for the analytical calculations (i.e., large distance of the connecting plate from the concrete edges and no concrete failure on the tension side). An elastic perfectly plastic material model was adopted for the plate steel; the anchors were represented with a bilinear model; the concrete was modelled using the Menetrey-Willam model (1995). Hexahedral second-order elements were used for all parts of the models. No sliding was assumed between the steel plate and the concrete, whereas frictionless contact was assumed for all the remaining interfaces. Welds were not included in the FE models. The large deflection option was activated for the analysis. Detailed information about the FE models is provided by Della Corte and Cantisani (2023).

a)

b)

Fig. 3. (a) Sample connection #1; (b) Sample connection #2.

3.2. Results from the applications

Numerical moment-rotation curves from the FE models are provided in Fig. 4, where also comparisons with the analytical predictions are illustrated. The results in Fig. 4a and Fig. 4b are for the same level of the applied axial force (600 kN) and for the first and second connection sample respectively (Fig. 3). It is seen that a very good agreement was obtained with a slightly varying value of the coefficient (0.60 or 0.67). Fig. 4c shows results for the sample #1, with an axial force sufficiently large (3600 kN) to produce yielding of the connected member prior to failure of the connection. Fig. 4d shows results for the sample #2 with an axial force two times larger than the value used for the results in Fig. 4b, but with failure of the connection still occurring. Overall, the proposed analytical model seems very promising for the applications. Conclusions A method to design steel-to-concrete connections with stiffened plates was developed and it is presented in this paper. The method uses the “component” approach of EN 1993-1-8, while also adopting most of the relevant rules. The main difference with codified rules consists in the method used to locate the neutral axis, considering the plate stiffeners. A slightly different approach to include the effect of prying forces was also explored, to provide agreement with FE model simulations of two connection examples. The proposed design method is also respectful of the main criteria provided by EN 1992-4 for considering concrete failure modes in tension.