基于拉格朗日插值的教与学动态组自适应算法

Teaching and learning dynamic group adaptive algorithm based on Lagrange interpolation

为提高教与学动态分组优化算法的局部搜索能力,提出一种基于拉格朗日插值的教与学动态组自适应优化算法,通过引入拉格朗日插值作为局部搜索方法可处理求解多维度优化问题的加速收敛,使得求解精准度更高。为平衡算法的全局搜索能力和局部开发能力,引入自适应参数策略。通过引入这两种策略来提高局部搜索上的计算能力以及收敛速度,提高对全局的优化。选取6个单峰函数和4个多峰函数,将改进后的算法与另4个算法进行实验对比,研究结果表明,所提算法使计算结果更精准,收敛速度更快。

To improve the local search ability of teaching and learning dynamic group optimization algorithm,an adaptive optimization algorithm of teaching and learning dynamic group based on Lagrange interpolation was proposed.By introducing Lagrange interpolation as a local search method,the acccelerated convergence of solving multi-dimensional optimization problems was dealt with and the solution was more accurate.To balance the global search ability and local development ability of the algorithm,an adaptive parameter strategy was introduced.These two strategies were introduced to improve the computational power and convergence speed of local search,to improve the global optimization.Six unimodal functions and four multimodal functions were selected to compare the improved algorithm with the other four algorithms.Results show that the proposed algorithm makes the calculation results more accurate and the convergence speed higher.