A decision-making method for ship liquid tank transfer based on local optimization algorithms

To address the need for rapid decision-making in ship liquid tank transfer operations, a rapid transfer decision-making method based on local optimization algorithms is proposed. A nonlinear optimization model that includes mass movement and free surface correction effects is constructed, and tank level adjustment is thus transformed into a multi-constraint optimization problem. By introducing a criterion for minimizing the center of gravity transfer distance, efficient solutions are obtained using local optimization algorithms such as SLSQP (Sequential Least-Squares Quadratic Programming). Case studies show: (1) The accuracy of the SLSQP algorithm, which supports equality constraints, is higher than the BOBYQA (Bound Optimization by Quadratic Approximation) and COBYLA (Constrained Optimization by Linear Approximations) algorithms, which use penalty functions to handle equality constraints; (2) This decision-making method can generate transfer plans within 0.39 seconds under small-angle scenarios, meeting the rapidity requirements for adjusting the floating state of a ship.