Search

Data-driven models utilizing powerful artificial intelligence (AI) algorithms have been implemented over the past two decades in different fields of simulation-based engineering science. Most numerical procedures involve processing data sets developed from physical or numerical experiments to create closed-form formulae to predict the corresponding systems' mechanical response. Efficient AI methodologies that will allow the development and use of accurate predictive models for solving computational intensive engineering problems remain an open issue. In this research work, high-performance machine learning (ML) algorithms are proposed for modeling structural mechanics-related problems, which are implemented in parallel and distributed computing environments to address extremely computationally demanding problems. Four machine learning algorithms are proposed in this work and their performance is investigated in three different structural engineering problems. According to the parametric investigation of the prediction accuracy, the extreme gradient boosting with extended hyper-parameter optimization (XGBoost-HYT-CV) was found to be more efficient regarding the generalization errors deriving a 4.54% residual error for all test cases considered. Furthermore, a comprehensive statistical analysis of the residual errors and a sensitivity analysis of the predictors concerning the target variable are reported. Overall, the proposed models were found to outperform the existing ML methods, where in one case the residual error was decreased by 3-fold. Furthermore, the proposed algorithms demonstrated the generic characteristic of the proposed ML framework for structural mechanics problems. [ABSTRACT FROM AUTHOR]

• A Nonlinear Finite Element Analysis (NLFEA) modeling approach is developed for simulating FRP-reinforced deep beams without shear reinforcement. • Extensive database of experimental and NLFEA data is established for developing AI-algorithm. • AI-model is benchmarked against several design standards [EC, ACI 440.1R-15 and the modified ACI 440.1R-15 (for size effect)] for blind predictions. • The AI-model demonstrated superior generalization on the blind prediction dataset, in comparison to the design codes. The presented work utilizes Artificial Intelligence (AI) algorithms, to model and interpret the behavior of the fiber reinforced polymer (FRP)-reinforced concrete deep beams without stirrups. This is done by first running an extensive nonlinear finite element analysis (NLFEA) investigation, spanning across the practical ranges of the different input parameters. The FEA modeling is meticulously validated against published experimental results. A total of 93 different models representing a multitude of possible FRP-reinforced deep beam designs are rigorously analyzed. The results are then utilized in building an AI-model that describes the shear capacity for FRP-reinforced deep beams. The study investigates the effect of several factors on the shear capacity and identifies the vital parameters to be used for further model development. Additionally, the developed AI-model is benchmarked against several design standards for blind predictions on new unseen data and design codes, namely: the EC, ACI 440.1R-15, and the modified ACI 440.1R-15 (for size effect). The AI-model demonstrated superior generalization on the blind prediction dataset in comparison to the design codes. [ABSTRACT FROM AUTHOR]