一种高效的函数可微的全局优化模拟退火算法

An Efficient Simulated Annealing Algorithm for Global Optimization Problems with Differentiable Functions

针对函数可微的全局优化问题,将最速下降法,Newton法和罚函数法引入模拟退火算法中,提出了一种高效的模拟退火算法.该算法可以求得可微函数优化问题的全局最优解,且具有计算量小,效率高的特点.利用罚函数将约束优化问题转化为无约束优化问题后,可以利用提出的算法进行求解.数值算例表明,提出的算法能够高效地求解无约束及带约束的函数可微的全局优化问题.

This paper proposes an efficient simulated annealing algorithm with respect to the global optimization problem with differentiable function,which combines steepest decent method,Newton method and penalty function method into simulated annealing algorithm.This algorithm could acquire the global resolution of the global optimization problem with differentiable function and is characterized by small computation quality and high efficiency.After using penalty function to transform the constrained optimization problem into unconstrained optimization problem,this algorithm could solve the global optimization problem.Experiment shows that this algorithm could be highly efficient in solving the global optimization problem with unconstrained or even constrained differentiable function.