A Master of Science thesis in Mechatronics Engineering by Ibrahim Javed Choudhary
entitled, "Artificial Intelligence-Based Modeling of the Nonlinear Behavior of Buckling-Restrained Braces," submitted in March 2013. Thesis advisor is Dr. Khaled Assaleh and co-advisor is Dr. Mohammad AlHamayde. Available are both soft and hard copies of the thesis.
The present research attempts to investigate and compare various artificial intelligence techniques to model the dynamic nonlinear behavior of Buckling Restrained Braces (BRBs). The various intelligent models are developed using normalized time-delayed inputs and outputs to predict normalized brace forces during load reversals. The values of brace forces are denormalized via an auxiliary intelligent (MLP) model. The training and testing of the proposed models are performed using experimental data from BRB specimens tested at the Pacific Earthquake Engineering Research (PEER) Center, University of California, Berkley. Experimental data from one specimen is used in the model development (training) stage. In addition, three sets of data are used to test the model's learning and generalizing abilities. Brace extensions are used as the network input to estimate the resulting brace forces in a longitudinal direction only. The network performance with different parameters is evaluated in order to arrive at an optimized architecture that best models the phenomenon. The nonlinear hysteretic behavior predicted by the majority of the employed models shows excellent agreement with the experimental results for the training sample. The generalization and prediction capability of the several proposed models is further demonstrated by predicting the hysteretic behavior of the testing samples with noticeable and vivid accuracy. The presented models represent a powerful tool for virtually testing BRB specimens. Such a tool supplements the traditionally available experimental tools for BRB performance investigation. A comparison on the basis of RMSE and Coefficient of Determination (R^2) is carried out to quantify and judge the performance of each implemented model. Further, the estimated peak response quantities and the energy dissipation during hysteretic cycles are also evaluated for precise comparison. The developed modeling techniques facilitate the BRB design and performance investigation processes by minimizing the need for, and extent of, experimental testing. Keywords: Buckling-Restrained Brace, Artificial Neural Network, TDNN, NARX, GMM, ANFIS, Polynomial Classifier.