How to Control the Quality of Concrete Strength Estimates Derived from NDT Measurements and Conversion Models

Author(s): Denys Breysse, Maitham Alwash and Zoubir Mehdi Sbartai
Paper category: Proceedings
Book title: Proceeding of the 71st RILEM Annual Week & ICACMS 2017
Editor(s): Manu Santhanam, Ravindra Gettu, Radhakrishna G. Pillai, Sunitha K. Nayar
ISBN:978-2-35158-195-7, 978-2-35158-190-2 (Set)
e-ISBN: 978-2-35158-191-9
Publisher: RILEM Publications SARL
Publication year: 2017
Pages: 517-526
Total Pages: 10
Language : English

Abstract:  The non destructive (ND) assessment of concrete strength in existing structures is a complex challenge. Its quality depends on: the number of cores, the measurement uncertainties, the properties of concrete, especially its strength range, and variability, the quality of the conversion model which is used for deriving strength estimates from ND test results. When focusing on the quality of the conversion model one must also consider the mathematical shape of the model, how the location of cores is decided, and the calibration method of the model. The deterministic assessment problem is therefore changed into a more accurate and realistic probabilistic problem, which will be studied in this paper. The work will be based both on (a) already available datasets, from laboratory studies and on-site structures, (b) additional synthetic datasets covering a range of representative concrete strengths. Risk curves will be built which establish a direct relationship between the quality of the strength assessment and major influencing factors, namely the quality of measurement and the number of cores. Additional recommendations will guarantee an assessment quality compatible with the assessor’s objectives. This work will be used as a contribution for future RILEM recommendations which will be issued in few months.

Online publication :2017
Publication type :full_text
Public price (Euros) : 0.00

>> You must be connected to view the paper. You can register for free if you are not a member