基于多策略改进蜣螂算法的推力分配应用设计  

Application Design of Thrust Allocation Based on Multi Strategy Improved Dung Beetle Algorithm

作　　者：刘明[1] 娄德成[1] 王晓飞 LIU Ming;LOU Decheng;WANG Xiaofei(Xinglin College,Nantong University,Nantong 226236,China;China Merchants Heavy Industry(Jiangsu)Co.,Ltd.,Nantong 226116,China)

机构地区：[1]南通大学杏林学院,南通226236 [2]招商局重工(江苏)有限公司,南通226116

出　　处：《自动化与仪表》2024年第8期117-121,136,共6页Automation & Instrumentation

基　　金：江苏省自然科学基金项目(BK20180953);南通市科技计划项目(JC22022085)。

摘　　要：推力分配求解是复杂的非线性约束优化问题。传统推力分配算法在处理该类问题时存在精度低及易陷入局部极值点等问题,而群智能算法能够较容易解决这些问题,但需要解决稳定性和快速收敛性问题。针对上述问题,提出一种改进蜣螂推力分配算法(IDBO),该算法通过选取解系数为个体变量、种群初始化采取拉丁超立方和反向学习法、蜣螂跳舞行为位置及最优蜣螂位置更新策略设计,提高算法收敛速度和稳定性。仿真结果表明,该算法收敛性速度优于所对比群智能算法,推力分配精度和能耗也明显优于所有对比算法。The solution of thrust allocation is a complex nonlinear constrained optimization problem.Traditional thrust allocation algorithms have low accuracy and are prone to falling into local extreme points when dealing with such problems.Although swarm intelligence algorithms can easily solve these problems,they need to solve stability and fast convergence problems.To solve them,an improved dung beetle thrust allocation algorithm(IDBO)is proposed.It improves the convergence speed and stability of the algorithm by selecting solution coefficients as individual variables,using Latin hypercube and reverse learning methods for population initialization,improving the position update strategy for dung beetle dancing behavior,and designing the optimal dung beetle position update strategy.The simulation results show that the convergence speed of it is superior to the compared swarm intelligence algorithms,and the thrust allocation accuracy and energy consumption are also significantly better than all the comparison algorithms.

关 键 词：推力分配 蜣螂算法 动力定位 拉丁超立方 

分 类 号：U664[交通运输工程—船舶及航道工程] TP273[交通运输工程—船舶与海洋工程]