基于粒子群算法的动力定位推力分配决策变量优选

针对群智能算法解决动力定位推力分配问题易陷入局部最优、计算时间长等不足，基于粒子群算法探索不同粒子决策变量对推力分配结果的影响。首先考虑推力分配目标力和力矩、推力限制、禁止角等约束条件，以推进器功率最优、磨损最小为优化目标建立了推力分配数学模型，构建了基于三种不同粒子决策变量的粒子群推力分配算法；其次以算法结果的适应度值、计算消耗时间的均值和方差量化算法的收敛性和实时性，对上述三种方法进行了仿真分析，仿真结果对比表明，基于本文提出的粒子决策变量搜索在收敛性和实时性上均达到最优，对粒子群算法解决推力分配问题有一定的参考价值。

Optimization of Decision Variables for Dynamic Positioning Thrust Distribution Based on Particle Swarm Optimization

Aiming at the fact that group intelligent algorithm for solving the dynamic positioning thrust distribution is easy to fall into the local optimum and the calculation time is long., the particle swarm optimization algorithm is used to explore the influence of different particle decision variables on the thrust distribution result. Firstly, the thrust force distribution target force and moment, thrust limit, prohibition angle and other constraints are considered. The thrust distribution mathematical model is established with the optimal thruster power and minimum wear. The particle swarm acceleration based on three different particle decision variables is constructed. The convergence and real-time performance of the algorithm are measured by the mean and variance of the fitness value and the calculated consumption time. The above three methods are simulated and analyzed. The comparison of the experimental results shows that the particle decision variables based on the proposed algorithm are based on the proposed method. The search is optimal in both convergence and real-time, and it has certain reference value for solving the thrust distribution problem by particle swarm optimization.