In this paper, three artificial neural networks are presented using the experimental results from bolted moment\r\nconnections among cold-formed steel members and the software MATLAB in order to predict the rotation at the\r\nconnections. A common neural network which has a multilayer perceptron along with back propagation learning\r\nalgorithm is applied in this research. Each of the networks consists of four layers including two hidden ones. The\r\nnumber of neurons in the first hidden layer is changed from 1 to 10 to achieve optimal results. The best results\r\nare obtained when the networks had 10, 10, and 9 neurons in the first hidden layer for column base and beam\r\ncolumn connections (in positive and negative rotations), and they had the performance of 0.0001371, 0.00044, and\r\n0.00047, respectively, after being trained in the software MATLAB. Thirty percent of the data from each test series\r\nwere omitted randomly in order to verify the networks. The Mannââ?¬â??Whitney P value tests are 0.9933, 0.9393, and\r\n0.9653 for column base and beam column connections (in positive and negative rotations), respectively.
Loading....