Analysis On Interface Debonding Failure Of Rc Beams Strengthened With FRP Based On A Beam Segment Model

Fiber reinforced composites strengthened RC structure is a current hot topic in civil engineering. FRP–to-concrete interface bond quality is the key issue for a successful strengthening, so far two difficulties yet have not been solved successfully: (1) analysis method of interface debonding in the end and mid-span of FRP strengthened RC beams; (2) testing method of bond-slip constitutive law of FRP-to-concrete interface. Existing studies of these two issues above were systematically reviewed in this thesis firstly, and then targeted researches were carried out.(1) Analysis method of interface debonding in the end and mid-span of FRP strengthened RC beamsIt is difficult and cumbersome to take a whole strengthened beam as a finite element model model and consequently to carry out a parameter study of interface debonding. Noticing these, various simple "beam segment" models of beam end, shear-flexure zone or pure-flexure zone were proposed respectively in this thesis, in which concrete cracks were taken as the boundaries for each segment. The new simple―beam segment model‖was validated to be rational and reliable respectively by using Saint-Venant Principle and a numerical study. The difficulties of the whole beam calculation, thus, were resolved.Based on these "beam segment" models, parameter analysis of debonding was performed for the first time, using general finite element software ANSYS. The result shows that FRP tensile stress and the layers are the most important influence factors of debonding. FRP-concrete bond-slip constitutive law has an important impect on debonding, which is the response of bonding quality. Distance from critical diagonal crack to the bearing of beam has a great influence on debonding of beam end. The influence of shear span ratioλand the crack spacing on debonding of shear-flexure zone is relatively small. Because of symmetry, no entire interface debonding occurs in pure-flexure zone.Comparison with a large number of experimental results collected shows the various "beam segement" debonding models proposed by this thesis are in good agreement with existing experimental results.As no finite element model of the entire beam which can analyse deonding in both beam end and mid-span has been reported yet. Based on "beam segement" model, a simple and reliable new analysis method named as segmentation-synthesization method was proposed by this thesis, which can be used to analyse debonding in different zones at the same time. Calculation examples show in addition to tensile stress and the FRP layers, the distance from critical diagonal crack to the bearing of beam is most sensitive on debonding failure. In order to further investigate the fluence of the critical diagonal crack location on debonding failure, a new beam specimen desigh with an opening was proposed by this thesis.(2) Test method of FRP-to-concrete interface bond-slip constitutive curveIt is impossible to measure descending constitutive curve of FRP-to-concrete interface bond-slip by using existed specimens loaded from just one side of FRP, because of the sudden destruction of the entire FRP-to-concrete interface when FRP-concrete slip is relatively small. In light of "beam segement" model, a new test method of FRP-to-concrete interface bond-slip constitutive curve was proposed. An improved "double-pull" specimen and loading equipment were designed to load from both two sides of FRP. Complete bond-slip constitutive curve is obtained by each of 6 specimens’tests, indicating the success of the new experimental method.By "beam segement" model and "double-pull" specimen, a new beam specimen was inspired for testing FRP-to-concrete interface bond-slip constitutive curve. Preset cracks in pure-flexure section were designed to form several "double-pull beam segment" specimens, which might improve test efficiency.To sum up the main innovation and contribution of this paper are:(1) Reliable, simple debonding calculation method of FRP strengthened RC beams and analysis of influencing factors.a) Reliable, simple "beam segement" models in both beam end and mid-span were proposed, which make the calculated results more realistic.b) For the first time, a relatively comprehensive and detailed analysis for influencing factors of FRP-concrete debonding was carried out. Finite element results agree well with a large number of experimental results.c) A new method was proposed for analyzing debonding in different zons at the same time.d) An experimental method using beam with an opening was presented for the study of beam end debonding.(2) Experimental method by which complete FRP-to-concrete interface bond-slip constitutive curve can be measured.a) "Double-pull" specimen and test method are improved, by which complete FRP-to-concrete interface bond-slip constitutive curve can be measured.b) Tentative idea of improved beam specimen might improve FRP-to-concrete interface bond-slip constitutive curve test efficiency.