عنوان مقاله [English]
In this paper, three types of Artificial Neural Network (ANN) are employed to interpret pressuremeter test results. First, a multi layer perceptron neural network, one of the most applicable neural networks, is used. Then, a neuro-fuzzy network, a combination of neural-fuzzy networks, is employed and, finally, a radial basis function, successful in solving nonlinear problems, is applied. Of all neural network models, the multi layer perceptron neural network proved to be the most effective. Finally, different models have been compared and the network with the most outstanding performance in two stages is determined. For the purpose of assessment, the capability of the model generalization, and the performance of the mentioned network against inexperienced data has been compared with empirical results. Contrary to conventional behavioral models, models based on neural networks are unable to demonstrate the effect of input parameters on output parameters. This research is a response to this need, through conducting a sensitivity analysis on the optimal structure of proposed models. Also, derivation of a governing equation for the neural network model gives more assurance to the user to employ such models, and, consequently, facilitates the application of models in engineering practices.