| 摘要: |
| 通过在遗传算法中嵌入牛顿算子，并定义适当的适应度和数据结构，从而得到可结合遗传算法和牛顿法两者长处，既有较快收敛性，又能以较大概率求得非线性最小二乘全局解的混合计算智能算法．数值结果表明了该方法显著优于遗传算法和牛顿法． |
| 关键词: 武汉冶金科技大学自动化系 430081 |
| DOI： |
| 分类号: |
| 基金项目:本文研究得到武汉市科委“晨光计划”资助. |
|
| A HYBRID COMPUTATIONAL-INTELLIGENT ALGORITHM FOR THE NONLINEAR LEAST-SQUARES GLOBAL SOLUTION |
|
ZHAO Mingwang
|
| Abstract: |
| Through embedding a Newtonian operator into the genetic algorithm and defining a proper fitness and a numerical structure，a hybrid computational-intelligent algorithm for the global solution of the nonlinear least-squares problem，combined the advances of both of the genetic algorithm and the Newtonian algorithm，is got with the faster convergence and the greater probability for the global solution．The numerical results show that the method is distinctly superior to the genetic algorithm and the NeWtonian algorithm． |
| Key words: Computational intelligent genetic algorithm Newtonian algorithm nonlinear least-squares problem |
