Peerawut Watsaratiyanont  <br/><br/>
Data-driven numerical model updating of reinforced concrete structures using artificial neural networks 

 - Pathum Thani, Thailand :  Asian Institute of Technology,  2025 
 - 79 leaves :  ill.+  1 online resource 
 - Thesis ;  no. ST-25-12 .
 - Asian Institute of Technology.  Thesis ;  no. ST-25-12 .
<br/><br/> A thesis submitted in partial fulfillment of the requirements for the  degree of Master of Engineering in Structural Engineering 
<br/><br/>Thesis (M. Eng.) - Asian Institute of Technology, 2025 
<br/><br/> This research develops and evaluates a data-driven model updating framework for reinforced  concrete (RC) building structures using Artificial Neural Networks (ANN) and the 3D Applied  Element Method (AEM). The proposed method aims to improve the accuracy of gradient-based  model updating by providing more realistic initial estimates of material properties based on  measured modal parameters.Databases were generated by varying Youngs modulus in 3D AEM models and recording  corresponding modal responses. ANN models trained on these databases achieved high  predictive accuracy in low- and moderate-resolution configurations. Verification on a  controlled numerical model demonstrated improved convergence and robustness compared to  conventional random-initial-guess approaches. Validation on two experimental case studies  showed that the method produces updated models with plausible material properties and  improved agreement with measured modal data.The results confirm that the proposed framework effectively integrates data-driven prediction  with numerical updating, providing a practical and efficient approach for RC building model  updating. 
<br/><br/><label>Subjects--Topical Terms: </label><br/>Reinforced concrete<br/>Concrete construction<br/>Artificial Intelligence