Fire resistance is one of the essential requirements that has to be considered in the design process of any structure. Methods for ensuring fire resistance of new structures are therefore, relatively well developed. Thus, loss of bearing capacity of a concrete structure during fire is relatively well known and documented, while determination of remaining bearing capacity of concrete structure after fire is less known. This can mainly be attributed to the lack of adequate knowledge of mechanical properties of concrete after fire, duration of fire exposure, heating rate, and maximum temperatures reached in the structure during fire. These factors have to be taken into account when making a decision on the rationality of the structure’s restoration. Estimation of bearing capacity of concrete structures after fire depends on precise but difficult determination of temperature dependent changes in material. Bearing capacity of structures after fire is in general, often assessed with simplified procedures and experimental investigation. These mainly covers visual inspection of the damaged structure and the execution of some standard, mostly destructive tests. Such tests are precise, but often difficult to implement, expensive, time-consuming, limited to certain pre-selected locations and can somehow weaken the individual structural element. Therefore, in order to determine bearing capacity of concrete structure after fire, various non-destructive methods are being developed. Among these, different ultrasonic (US) tests are frequently used and well described in the literature. These results indicate good connection between US pulse velocity and concrete compressive strength after exposure to elevated temperatures.This article presents the first part of the ongoing research. Main goal of the research is to develop a numerical method based on artificial neural network approach to estimate concrete properties after fire using non-destructive tests. In this paper, results of destructive and non-destructive tests are presented and analyzed to determine changes in basic mechanical properties of standard concrete mixture after exposure to elevated temperatures, namely 200 °C, 400 °C, 600 °C, or 800 °C. With the aim of separating ‘material’ and ‘structural’ effects, temperature gradients within the concrete specimens, were minimized, as possible. Prior experimental testing, the damage to the concrete was roughly detected by observing the specimen’s surface. Compressive strength, tensile strength, surface hardness, dynamic elastic modulus, and shear modulus are determined. It is noticed that crack formation at 400 °C has an influence on reduction of the US pulse velocity, tensile strength, dynamic elastic, and shear modulus. Analysis of variance is used for assessing the experimental results sensitivity on temperature elevation. It is noticed that US method detects statistically significant impact of elevated temperatures on US pulse velocity. On the other hand, rebound hammer technique indicates statistically significant impact of elevated temperatures on concrete surface hardness between 400 °C and 600 °C.
Read full abstract