Evaluation of real performance of reinforced concrete cantilevered balcony structures using forced-vibration technique

Abstract

In this thesis, the term 'real' is used as the opposite of 'ideal'. 'Ideal performance' can be described as behavior of a structure that can be expected to perform strictly in accordance with the assumed physical and material properties of the structure based on the theories and knowledge known. However, engineering design/analysis theories are inherently incomplete description and inevitably simplifications of the reality due to a whole host of factors: complexities of material behaviours, presence of flaws, possibilities of manufacturing or construction errors etc. Hence the 'real' performance of structures rarely fits or agrees with the 'ideal' performance. Hence, the 'real' structural responses/performance of structures cannot be obtained from design and theoretical analyses alone but via real life full-scale testing only. In structural design, large factors of safety (FOS) are allowed in order to provide greater safety margins to cover these imperfections of knowledge and understanding and between the real and ideal performance of structures. As a result, the traditional engineering design theories/approaches adopt appropriate FOS material factors (regarding materials strengths and ultimate resistance/stresses) and FOS loading factors (regarding loading and their combinations) to balance economy and risk of failure. This thesis reports an experimental investigation/evaluation to enrich our existing understanding of the real performance of reinforced concrete cantilevered balcony structures (RCCBS) through investigating the inter-relationships amongst the so called 'balcony parameters' (expressed in terms of dynamic, structural and material properties). It details an experimental approach of investigating/evaluating the 'real' performance of RCCBS and the efforts in the establishment and formulation of a performance-based phenomenological models, which is heretheto conveniently coined the terms 'unified structural performance' (USP) model, using forced-vibration measurement technique. To reveal the 'real' performance of the RCCBS, the contribution and the constitutive effects to the global stiffness of structure arising from tensile capacity of concrete (TCC) has been a subject of intensive investigation in this study. This research is an attempt, as far as the author is aware, the first study of its kind (i.e. in terms of the technique of evaluating the real performance of RCCBS based on an investigation of the inter-relationships amongst dynamic, structural and material properties of structures with due consideration of TCC effect). An important contribution of this research is that, by simply charting the resonant frequency measured experimentally into the prescribed USP model, the instantaneous/inherent values/changes of structural and material properties of the tested RCCBS at varying degrees of damage (the loss of tensile steel rebar area) can be ascertained successfully based on their respective trajectories. To obtain these useful and informative trajectories (i.e. empirical 'causal' relationship between the relevant properties charted according to the different balcony parameters of the USP model), a series of well-coordinated systematic experimental programmes have been designed and implemented using three different RCCBS models (containing one scaled-down and two full-scale physical models) each differs one another by the structural characteristics (in terms of steel reinforcement ratio, physical size, span etc,) and loading conditions (high/low loading condition) so that variations in structural characteristics and loading conditions have been taken care. These RCCBS models were made to deteriorate artificially and progressively by a novel damage strategy (in which the tensile steel rebars were cut-off by rotary cutting machine in order to reduce the effective area of steel rebar at predetermined step sizes). With this technique, the severity of damage to the structure could be measured/quantified accurately by quantifying/measuring only the loss of steel rebar area and the corresponding changes of structural and material properties at varying degrees of damage without the need of imposing additional loads on the tested structure. The effect of tensile capacity of concrete (TCC) to the global stiffness of the structure is a subject of concern in this research. The effective I-value of the slab section is one of the critical parameters governing the global stiffness of the structure. For a structure with high TCC effect, the effective concrete area in tension zone is considerably larger (reflected by much lower effective area ratio of steel-to-concrete), whereas the TCC effect becomes lower as the reduction of effective concrete area is lowered due to cracking (as the effective concrete area decrease). However, the determination of TCC effect not only depends on the effective concrete area alone, but also the instantaneous tensile stress/strain of concrete (ITSScon) which is another critical parameter affecting the extent of participation of tensile steel rebar (in terms of the effectiveness of anchorage and composite action by inflicting stress on the tensile steel rebars in concrete). Arising from the experiments on the scale-down RCCBS models, the performance regimes of the RCCBS were found to be divided into two regions by a so-called 'turning point' (TP). Before the TP, where the TCC effect was higher and the ITSScon was lower, the resonant frequency was found to be less sensitive to the damage introduced. But the situation changes after passing the TP, where the TCC effect was lower and the ITSScon was higher, the resonant frequency was more sensitive to the damage on the structure. These observations and findings were further validated by the experiments on two full-scale RCCBS models under two separate performance regimes (high/low ITSScon). These observed phenomena are considered to exhibit profound influences on the understanding of the role of TCC on the real structural behaviour of RCCBS. Also, the findings provide evidence to explain why most researchers claimed that the resonant frequency is found less sensitive to the damage especially when damage occurred at sections of low stresses as the TCC effect have not been taken into consideration. By constructing and assembling a set of trajectories for the individual balcony performance not only in terms of dynamic properties (i.e. resonant frequency) but also structural and material properties (i.e. remaining tensile steel area, neutral axis position and stress distribution of materials) of the two full-scale RCCBS models, a performance-based phenomenological USP model, was established/formulated to cover the two different performance regimes of RCCBS. With this USP model, the real performance of RCCBS can be tracked and evaluated by charting the resonant frequency measured from the tested structure to their rightful positions corresponding to the two performance regimes governed by different ranges of instantaneous tensile stress/strain of concrete (ITSScon). Verification of the validity and usefulness of the USP model was further undertaken using data measured from a number of real-life balconies with similar structural characteristics as those of the two full-scale RCCBS models. However, since limited numbers of performance regimes were investigated due to the limitations on scope and time, the resonant frequencies of the tested real-life balconies are found not to fit completely into the prescribed frequency range covered by the USP model and thus over-interpretation of the structural condition of the inspected real-life balconies should be cautioned. Therefore, further investigation of more performance regimes is recommended in order to establish and formulate a more complete USP model, which allows even better understanding of real performance of these real-life balconies. It is also believed that the USP model and the experimental technique proposed and advocated in this research have already provided fundamental and indispensable understanding of the real performance of RCCBS and the role of TCC in these kinds of structure. Thus it will enable us to apply such understanding to other RC structures/elements. Indeed, this research is believed to have advanced the frontiers of both research on damage detection and structural assessment.