基于改进海鸥优化算法的热连轧板带轧制规程优化

Optimization of hot continuous rolled strip rolling schedule based on improved seagull optimization algorithm

为提高热连轧板带产品质量及生产效率,提出一种改进的多目标海鸥算法来优化热连轧负荷分配。采用非线性衰减与余弦递减策略以改善海鸥算法中种群个体的越界问题;采用Sobol序列映射初始化种群来增加海鸥算法的全局搜索能力,从而使初始值尽可能均匀地分布在解空间内。将改进后的海鸥算法与NSGA-II、MOEA/D和MOSOA通过测试函数进行对比,结果表明,该算法的求解精度更高且具有良好的收敛性和多样性。最后,将该算法应用到热连轧负荷分配优化计算中,得出的解集在多目标空间中获得了较好的Pareto前沿,同时由该方法获得的负荷分配数据优于经验法,能更好地控制板形,为现场工程技术人员提供了多样化的选择。

In order to improve the product quality and production efficiency of hot continuous rolled strip,an improved multi-objective seagull optimization algorithm was proposed to optimize the load distribution of hot continuous rolling.The nonlinear decay and cosine decrement strategies were used to improve the boundary violation issue in the seagull optimization algorithm;the Sobol sequence mapping was used to initialize the population to enhance the global search capability,thus the initial values can be distributed evenly in the solution space.The improved seagull optimization algorithm was compared with NSGA-II,MOEA/D and MOSOA using test functions.The results show that the proposed algorithm achieves higher solution accuracy with good convergence and diversity.Finally,the algorithm was applied to optimization calculation of the load distribution of hot continuous rolling,and the obtained solution set obtains a better Pareto front in the multi-objective space.At the same time,the load distribution data obtained by this method outperforms empirical methods,allowing for better control of plate shape and providing engineering technicians with diverse choices in the field.