**6. Final Remarks**

The main objective of this paper was to investigate the beam-to-beam structural connections effect (rigid, semi-rigid and flexible) and the influence of steel-concrete interaction degree (from total to various levels of partial interaction) over the non-linear dynamic behaviour of composite floors when subjected to human rhythmic activities. This way, an extensive para‐ metric analysis was developed focusing in the determination quantitative aspects of the composite floors dynamic response.

The investigated structural model was based on a steel-concrete composite floor spanning 40m by 40m, with a total area of 1600m2 . The structural system consisted of a typical compo‐ site floor of a commercial building. The composite floor studied in this work is supported by steel columns and is currently submitted to human rhythmic loads. The structural system is constituted of composite girders and a 100mm thick concrete slab.

The proposed computational model adopted the usual mesh refinement techniques present in finite element method simulations, based on the ANSYS program. The numerical model enabled a complete dynamic evaluation of the investigated steel-concrete composite floor especially in terms of human comfort and its associated vibration serviceability limit states.

The influence of the investigated connectors (Stud Bolts: 13mm and 19mm) on the compo‐ site floor natural frequencies was very small, when the steel-concrete interaction degree (from total to partial) was considered in the analysis. The largest difference was approxi‐ mately equal to 5% to 7%.

On the other hand, when the joints flexibility (rigid to flexible) and steel-concrete interac‐ tion degree (from total to partial) decreases the composite floor natural frequencies be‐ come smaller. This fact is very relevant because the system becomes more susceptible to excessive vibrations.

The composite floor vibration modes didn't present significant modifications when the con‐ nections flexibility and steel-concrete interaction was changed. The investigated structure presented vibration modes with predominance of flexural effects. The results have indicated that when the joints flexibility (rigid to flexible) and steel-concrete interaction degree (total to partial) decreases the composite floor peak accelerations become larger.

The maximum acceleration value found in this work was equal to 0.80m/s2 (ap = 0.80 m/s2 : flexi‐ ble model) and 0.63m/s2 (ap = 0.63 m/s2 : semi-rigid model), while the maximum accepted peak acceleration value is equal to 0.50m/s2 (alim = 0.50m/s2 ) [6,9]. The structural system peak acceler‐ ations were compared to the limiting values proposed by several authors and design standard [6,9]. The current investigation indicated that human rhythmic activities could induce the steel-concrete composite floors to reach unacceptable vibration levels and, in these situations, lead to a violation of the current human comfort criteria for these specific structures.
