New approach for developing soft computational prediction models for moment and rotation of boltless steel connections. (December 2018)

• Record Type: Journal Article
• Authors: Shah, S.N.R.; Ramli Sulong, N.H.; El-Shafie, Ahmed
• Is Part Of: Thin-walled structures. Volume 133(2018)
• Page Start: 206
• Page End: 215
• Publication Date: 2018-12
• Subjects: Boltless steel connections -- Moment-rotation -- LGP -- ANN -- ANFIS; Thin-walled structures -- Periodicals; 690.1
• DOI: 10.1016/j.tws.2018.09.032
• Languages: English
• ISSNs: 0263-8231

Abstract

Abstract: This study aims to minimize the expensive experimental testing of unique boltless steel connections using the prediction power of several computational techniques. Thirty-two tests were conducted on boltless steel connections using double-cantilever test method and their results were compared with developed models using Artificial Intelligence (AI) techniques. Linear Genetic Programming (LGP), Artificial Neural Networks (ANNs) and Adaptive Neuro Fuzzy Inference System (ANFIS) were applied to predict the moment-rotation (M-θ) behavior of boltless steel connections. The predictive performance of the models was assessed by comparing the values of co-efficient of determination (R 2 ), mean square error (MSE) and root-mean-square error (RMSE). The LGP model well predicted the M-θ behavior as compared to the other models. The robustness of the LGP model was further proved by performing different statistical tests. Highlights: Data set of thirty-two tests conducted on boltless steel connections was used. Computational models were developed using LGP, ANN and ANFIS. Computational models and experimental results were compared. Performance of the models was assessed by comparing the values of R 2, MSE and RMSE. LGP model outperformed the other developed models.