论文部分内容阅读
提出一种新的基于遗传算法的模拟退火罚函数方法求解地下水管理模型 .遗传算法是建立在自然遗传学和自然选择机理上的全局随机搜索和进化的过程 .与传统的基于梯度寻优方法相比 ,遗传算法寻优不须优化问题连续可导 ,同时通过模拟退火罚函数方法来处理约束条件 ,可以保证算法逐渐收敛于可行的最优解 ,克服一般遗传算法中罚因子选取的困难 .实例求解结果表明此方法在优化地下水管理模型中具有较好的效果 .
A new simulated annealing penalty function method based on genetic algorithm is proposed to solve the groundwater management model.Genetic algorithm is a global random search and evolutionary process based on natural genetics and natural selection mechanism.Compared with the traditional gradient based optimization method Compared with genetic algorithms, optimization problems need not be continuously derivable, and simulated annealing penalty functions can be used to deal with the constraints, which can ensure that the algorithm converges to the feasible optimal solution gradually and overcomes the difficulty of choosing penalty factors in general genetic algorithms. The results show that this method has a good effect in optimizing groundwater management model.