In the practical design of thin-walled composite columns, component dimensions should be wisely designed to meet the buckling resistance and economic requirements. This paper provides a novel and useful investigation based on a numerical study of the effects of the section dimensions, thickness ratio, and slenderness ratio on the critical buckling load of a thin-walled composite strut under uniaxial compression. The strut was a channel-section-shaped strut and was made of glass fiber-reinforced polymer (GFRP) composite material by stacking symmetrical quasi-isotropic layups using the autoclave technique. For the purpose of this study, a numerical finite element model was developed for the investigation by using ABAQUS software. The linear and post-buckling behavior analysis was performed to verify the results of the numerical model with the obtained buckling load from the experiment. Then, the effects of the cross-section dimensions, thickness ratio, and slenderness ratio on the critical buckling load of the composite strut, which is determined using an eigenvalue buckling analysis, were investigated. The implementation results revealed an insightful interaction between cross-section dimensions and thickness ratio and the buckling load. Based on this result, a cost-effective design was recommended as a useful result of this study. Moreover, a demarcation point between global and local buckling of the composite strut was also determined. Especially, a new design curve for the channel-section GFRP strut, which is governed by the proposed constitutive equations, was introduced to estimate the critical buckling load based on the input component dimension.

No abstract

Due to the increase of missile performance, the safety design requirements of military and industrial reinforced concrete (RC) structures (i.e., bunkers, nuclear power plants, etc.) also increase. Estimating damage levels in the design stage becomes a crucial task and requires more accuracy. Thus, this study proposed a hybrid machine learning model which is based on k-nearest neighbors (KNN) and Bayesian optimization (BO), named as BO-KNN, for predicting the local damages of reinforced concrete (RC) panels under missile impact loading. In the proposed BO-KNN, the hyperparameters of the KNN were optimized by using the BO which is a well-established optimization algorithm. Accordingly, the KNN was trained on an experimental dataset that consists of 254 impact tests to predict four levels (or classes) of damages including perforation, scabbing, penetration, and no damage. Due to the unbalance of the number of tests in each damage class, an over-sampling technique called BorderlineSMOTE was employed as a balancing solution. The predictability of the proposed model was investigated by comparing with the benchmark models including non-optimized KNN, multilayer perceptron (MLP), and decision tree (DT). Accuracy, F1-score, and area under the receiver operating characteristic (ROC) curve (AUC) were utilized to evaluate the performance of these models. The implementation results showed that the proposed BO-KNN model outperformed the other benchmark models with the average class accuracy of 68.05%, F1-score = 0.641, and AUC = 85.8%. Thus, the proposed model can be introduced as a foundation for developing a tool for predicting the local damage of RC panels under the missile impact in the future. Keywords: impact damage; k-nearest neighbors; Bayesian optimization; oversampling; imbalanced data; RC panel.