A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Ultrasonic pulse velocity and artificial neural network prediction of high-temperature damaged concrete splitting strength
To examine the integrity of any structure following a fire, assessments of the impact of high temperatures on concrete are essential, particularly its decreased in tensile strength. Destructive examinations, such as the extraction of concrete cores, can pose significant cost and safety challenges, particularly when applied to structures that have already sustained damage. Consequently, for assessing damaged concrete, non-destructive in-situ tests are the favored approach. This study aims to develop an artificial neural network model utilizing data from ultrasonic pulse velocity measurements. The model's purpose is to assess the tensile splitting strength of concrete subjected to elevated temperatures, ranging from 200 to 800 °C. The splitting strength investigation showed that increasing the exposure temperature from 200 to 800°C results in splitting strength reduction of 15 to 75% respectively. Also, the ultrasonic pulse velocity experienced a reduction of 85% when the exposure temperature reaches 800 °C. In addition, the results of the artificial neural network model indicated that ultrasonic pulse velocity and temperature data were sufficient to reasonably forecast the tensile splitting strength of concrete. The developed artificial neural network model has a coefficient of determination (R2) of 0.943, a mean absolute relative error (MARE) of 5.028, and an average squared error (ASE) of 0.000907.
ANN modelling could predict concrete strength after fire exposure accurately.
Concrete splitting strength decreases significantly at higher temperatures (75% at 800 °C).
Non-destructive testing offers safe and accurate assessments for fire damaged concrete.
Ultrasonic pulse velocity and artificial neural network prediction of high-temperature damaged concrete splitting strength
To examine the integrity of any structure following a fire, assessments of the impact of high temperatures on concrete are essential, particularly its decreased in tensile strength. Destructive examinations, such as the extraction of concrete cores, can pose significant cost and safety challenges, particularly when applied to structures that have already sustained damage. Consequently, for assessing damaged concrete, non-destructive in-situ tests are the favored approach. This study aims to develop an artificial neural network model utilizing data from ultrasonic pulse velocity measurements. The model's purpose is to assess the tensile splitting strength of concrete subjected to elevated temperatures, ranging from 200 to 800 °C. The splitting strength investigation showed that increasing the exposure temperature from 200 to 800°C results in splitting strength reduction of 15 to 75% respectively. Also, the ultrasonic pulse velocity experienced a reduction of 85% when the exposure temperature reaches 800 °C. In addition, the results of the artificial neural network model indicated that ultrasonic pulse velocity and temperature data were sufficient to reasonably forecast the tensile splitting strength of concrete. The developed artificial neural network model has a coefficient of determination (R2) of 0.943, a mean absolute relative error (MARE) of 5.028, and an average squared error (ASE) of 0.000907.
ANN modelling could predict concrete strength after fire exposure accurately.
Concrete splitting strength decreases significantly at higher temperatures (75% at 800 °C).
Non-destructive testing offers safe and accurate assessments for fire damaged concrete.
Ultrasonic pulse velocity and artificial neural network prediction of high-temperature damaged concrete splitting strength
Discov Appl Sci
Almasaeid, Hatem (author)
2024-01-20
Article (Journal)
Electronic Resource
English
Artificial neural network , High temperature , Nondestructive test , Splitting tensile strength , Ultrasonic pulse velocity Engineering , Engineering, general , Materials Science, general , Earth Sciences, general , Applied and Technical Physics , Chemistry/Food Science, general , Environment, general
Prediction of concrete strength using ultrasonic pulse velocity and artificial neural networks
| BASE | 2015
| British Library Online Contents | 2006