系泊缆初始配重参数优化方法

当前系泊缆大多为组合缆的形式，需要加装配重块，配重位置和重量通常根据经验进行设计。为减少计算量，尽快找到最优配重参数，借助粒子群优化反向传播(Particle Swarm Optimization-Back Propagation, PSO-BP)神经网络，采用正交试验设计法确定神经网络的样本数据，利用神经网络的泛化能力以尽可能地减小有限元计算工作量，然后将神经网络预测结果应用到遗传算法中。以所有系泊缆张力之和的最小值为优化目标，在给定范围内搜索最优解。优化后8根系泊缆安全因数之和提高36.07,且高于随机选择的配重方案的安全因数，该方法可为大计算量的优化问题提供思路。

Optimization Method of Initial Clump Parameters of Mooring Lines

Most of the current mooring lines are composite lines, and they need to be assembled with clumps. The position and weight of the clump are usually designed based on experience. In order to reduce the amount of calculation and find the optimal parameters, the Particle Swarm Optimization-Back Propagation (PSO-BP) neural network is built, and the sample data of neural network is determined by the orthogonal experiment. The workload of the finite element calculation is reduced by the generalization ability of the neural network as much as possible. The neural network prediction results are applied to the genetic algorithm. The minimum sum of the tension of all mooring lines is taken as the optimization goal to search for the optimal solution within a given range. The sum of the optimized safety factors of 8 mooring lines increases by 36.07, which is higher than that of the randomly selected clump weight scheme. This method can provide ideas for the optimization problems with a large amount of calculation.