Artificial intelligence methods and analysis of structure in evaluation of hardened concrete quality

Title: Artificial intelligence methods and analysis of structure in evaluation of hardened concrete quality
Author(s): J. Kasperkiewicz
Paper category : conference
Book title: 2nd International RILEM Workshop on Life Prediction and Aging Management of Concrete Structures
Editor(s): D.J. Naus
Print-ISBN: 2-912143-36-5
e-ISBN: 2912143780
Publisher: RILEM Publications SARL
Publication year: 2003
Pages: 185 - 194
Total Pages: 10
Nb references: 23
Language: English

Abstract: Evaluation of quality of concrete in elements of building or civil engineering structures is needed in procedures of their acceptance, repair, modification or forensic analysis. There are various specialised test methods dedicated to particular characteristics of concrete, but of real importance is evaluation of the general quality of the material. Conventional methods of checking the concrete in such structures are either approximate or laborious, sometimes really expensive. Testing of various properties other than compressive strength may need long time and specialized equipment. A procedure is suggested for evaluation of the current state of hardened concrete based on structural analysis of small samples of material taken from selected elements of the construction, which is supplemented by a general assessment of the entire element, formulated during inspection on site.
The main data taken into account in the presented experiments enclose quantitative characteristics resulting from image analysis of the microstructure, from microhardness and other mechanical tests. Certain components of the database concern also qualitative classification and/or subjective statements of experienced observers. The processing of the data is done applying methods originating in so called Artificial Intelligence, like Artificial Neural Networks, (ANNs), or Machine Learning, (ML).

Online publication: 2003-04-02
Publication type : full_text
Public price (Euros): 0.00
doi: 10.1617/2912143780.018