Artificial neural network ability in evaluation of random wave-induced inline force on a vertical cylinder

An approach based on artificial neural network (ANN) is used to develop predictive relations between hydrodynamic inline force on a vertical cylinder and some effective parameters. The data used to calibrate and validate the ANN models are obtained from an experiment. Multilayer feed-forward neural networks that are trained with the back-propagation algorithm are constructed by use of three design parameters (i.e. wave surface height, horizontal and vertical velocities) as network inputs and the ultimate inline force as the only output. A sensitivity analysis is conducted on the ANN models to investigate the generalization ability (robustness) of the developed models, and predictions from the ANN models are compared to those obtained from Morison equation which is usually used to determine inline force as a computational method. With the existing data, it is found that least square method (LSM) gives less error in determining drag and inertia coefficients of Morison equation. With regard to the predicted results agreeing with calculations achieved from Morison equation that used LSM method, neural network has high efficiency considering its convenience, simplicity and promptitude. The outcome of this study can contribute to reducing the errors in predicting hydrodynamic inline force by use of ANN and to improve the reliability of that in comparison with the more practical state of Morison equation. Therefore, this method can be applied to relevant engineering projects with satisfactory results.     

Optimization of Gravity Concrete Dams Using the Grasshopper Algorithm (Case Study: Koyna Dam)

Geotechnical and Geological Engineering - The aim of this study is to optimize the geometric dimensions of the Koyna concrete weight dam with and without seismic forces using the grasshopper...