Optimization Of Docking Blocks

This paper investigates fully into the issues of optimization of docking blocks.It describes a mathematical model for optimizing the placement and dimensions of docking blocks. A two-level solution strategy is proposed to solve for the optimal solution. In the first level, the genetic algorithm is employed to determine the optimal positions of docking blocks while the sequential quadratic programming is utilized to obtain the optimal dimensions of docking blocks in the second level. The effects of the numbers of blocks placed in the aft part, the stiffness of the supporting structures on which the blocks are placed, and the maximum thickness of blocks on the design are discussed in the study. Numerical examples show that increase in numbers of docking blocks in aft part may lead to decrease in total volume of blocks. Placing the blocks on the supporting structures with comparatively small stiffness benefits both the decrease in total volume of blocks and even distribution of loads applied to the ship structures. However, increase in maximum thickness of blocks gives rise to worse distribution of loads although it may decrease the total volume of blocks.The paper describes a mathematical model for multi-objective optimization of docking blocks. The total volume of the blocks and the strength reserve of ship structures at stern that is represented by the sum of some blocks' reactions in aft part are chosen to be the objective functions. The modified lexicographic method is employed to solve for the optimal solution. The numerical simulations show that the strength reserve obtained from multi-objective optimization increases although the total volume of blocks also increases compared to that of the single objective optimization. Compared to the design scheme resulting from the single objective optimization, more reasonable design scheme is obtained by using multi-objective optimization proposed.This paper also proposes an approach of robust design of docking blocks with probabilistic and non-probabilistic uncertainties. Emphasis is placed on the effects of non-probabilistic uncertainties in ship girder loads and probabilistic uncertainties in stiffnesses of docking blocks on optimal solution. Numerical results show that uncertainties considered affect the optimization of docking blocks and lead to increase in weights of blocks as the uncertainties increase. The effect of uncertain combined stiffnesses of blocks on design is larger than that of ship girder loads with uncertainty.