Application of swarm intelligence for dynamic properties of moored floating structures using two-dimensional fluid dynamic program

This paper presents a numerical investigation into interactions between oscillatory waves and moored floating structures by through applying Swarm Intelligence (SI). The effects of small and large incident waves placing a force on the obstacles were studied using two floating structures composed of a rectangle and a cage kind with taut-moored and slack-moored situations. The aforementioned problem was reduced to solve systems of sparse-nonlinear equations for surge, heave, and pitch responses, as well as mooring forces considering that there is a distorted wave motion due to the presence of the movable obstacle boundary. The particle swarm optimization (PSO) algorithm is proposed here to provide insight into computational intelligence factors affecting calculation results and efficiency. Besides that, a computational fluid dynamic (CFD) program is based on a full Navier-Stokes solver. A series of numerical examinations for selected wave heights and wavelengths including a set of optimization procedures have been taken into account in the evaluation of formulated objective function, F, where the best potential solution is computed at each computational step. Meanwhile, a collection of empirical benchmarks were also developed based on reported measurements in the literature to provide sound empirical support for the theoretical findings. The results show that the optimization method exhibits a favorable convergent behavior (F approximate to 0), where the computed results agree well with the measurements in terms of mean amplitudes of the surge, heave, and pitch responses and peak values of mooring-line tensions. The approach would firmly converge on benchmarks with improved computational steps of 18,321 running for 20.1 wave cycles in the case of the taut-moored rectangular structure. For the slack-moored cage, a satisfactory convergence was achieved using a series of data calibrations into an algorithm that took 27,976 iterations and 24.5 s to complete. Furthermore, CFD captures of wave deformations, dynamic forces, and streamlines were able to qualitatively corroborate the results.