Jerzy Hoła

Application of artificial neural networks to determine concrete compressive strength based on non‐destructive tests

Abstract The paper deals with the neural identification of the compressive strength of concrete on the basis of non‐destructively determined parameters. Basic information on artificial neural networks and the types of artificial neural networks most suitable for the analysis of experimental results are given. A set of experimental data for the training and testing of neural networks is described. The data set covers a concrete compressive strength ranging from 24 to 105 MPa. The methodology of the neural identification of compressive strength is presented. Results of such identification are reported. The results show that artificial neural networks are highly suitable for assessing the compressive strength of concrete. The neural identification of the compressive strength of concrete has been verified in situ.

Identification of moisture content in brick walls by means of impedance tomography

Purpose – This paper aims to present a new non‐destructive method of brick wall dampness testing in real building structures. The electrical impedance tomography (EIT) method makes it possible to obtain a distribution of wall dampness. The paper aims to give basic information about the measuring system, including prototype equipment. The setup was used to determine the dampness of test brick walls on a specially built laboratory test rig. The paper seeks to compare test results obtained by the non‐destructive impedance tomography method with the results obtained by the conventional destructive dry‐weight method.Design/methodology/approach – The paper adopts a tomography approach to control humidity inside of the brick walls. In case of brick walls other nondestructive methods fields, for example, the ultrasonic tomography is useless. On the other hand the most popular dry‐weight method is strictly forbidden for historical buildings. As a forward solver, functionally graded material boundary element method...

New nondestructive way of identifying the values of pull-off adhesion between concrete layers in floors

AbstractThis paper presents a new nondestructive way of identifying the values of pull-off adhesion between the concrete layers in concrete floors. It based on the roughness parameters of the base layer surface, using the nondestructive optical technique, and on the floor surface, using the nondestructive acoustic techniques and employing artificial neural networks (ANNs) for this purpose. The new way has a potential for being widely used in practice, whereby it may become possible to employ previously trained ANNs to identify the pull-off adhesion, without impairing the surface of the tested concrete floor.

Methodology of nondestructive identification of defective concrete zones in unilaterally accessible massive members

AbstractThe paper deals with the nondestructive identification of defective concrete zones in unilaterally accessible massive members, for example, access galleries in hydroelectric power plants. The concrete in such zones is, for various reasons, excessively porous. The authors propose to use state-of-the-art acoustic testing techniques, including ultrasonic tomography, integratively to detect and identify defective zones. An original methodology for such tests has been developed. The methodology is illustrated with an example of its practical application to a real civil engineering structure.

Methodology of Neural Identification of Strength of Concrete

In this article, a new technique of the neural identification of the compression strength of concrete on the basis of nondestructive tests is described. The neural training and testing processes were carried out on a data set created on the basis of nondestructive tests. The data set covers a concrete compression strength range from 24 to 105 MPa after 28 days of curing. The methodology of the neural identification of the compression strength of concrete is presented. The results presented herein demonstrate that the identification of compressive strength of concrete by neural networks on the basis of such nondestructive methods as ultrasonic, sclerometric, and pull-out is a viable technique.

ANN modeling of pull-off adhesion of concrete layers

When making and repairing concrete floors it is vital to properly prepare the interlayer bonding surface. The measure of the bond is the value of pull-off adhesion fb experimentally determined in building practice by the semi-destructive (SDT) pull-off method. In this paper it is proposed to assess pull-off adhesion by jointly the optical laser triangulation method and the acoustic impulse response method, using artificial neural networks (ANN), on the basis of a few parameters (independent of top layer thickness) determined by these methods. The proposed non-destructive (NDT) pull-off adhesion assessment method is devoid of the drawbacks and inconveniences of the pull-off method and makes possible the reliable mapping of the adhesion level on the tested surface without local damage to the latter.

Stress failure of cement concretes under compression – synthesis of knowledge, conclusions

AbstractThis paper presents a synthetic review of the existing knowledge on the stress failure of cement concretes under compression, based on the available literature and the authors’ own research. It is pointed out that there is no knowledge on the stress failure of new-generation concretes, which needs to be acquired through proper research. The usefulness of the knowledge on the stress failure of concrete under compression for building practice is highlighted.

The effect of failure to comply with technological and technical requirements on the condition of newly built cement mortar floors

The article presents the results of “in situ” tests of two new cement mortar floors that were made defectively in an important public building with a large area. The performed results of organoleptic tests and extensive strength tests, which are helpful in determining the causes of numerous and various defects in floors, were presented. It was observed that the source of the aforementioned causes is not only the failure to comply with the technological and technical conditions and requirements for laying a cement mortar mix and the execution of floors but also the lack of appropriate supervision over the entire flooring process. The causes of defects were determined and should help to avoid similar situations in the future.

The Use of a 3D Scanner for Evaluating the Morphology of a Sandblasted Concrete Surface

The possibility of the non-destructive testing of concrete surface morphology by means of a new-generation 3D laser scanner is presented. The test results for a reference concrete substrate surface and a sandblasted concrete substrate surface are reported. The effect of sandblasting on the morphology of the concrete surfaces of the substrates is highlighted. The results are presented as three-dimensional images and the values of the morphology parameters are compared in a table.