【摘要】 用进化规划对约束最优化问题提出了一种新的惩罚函数方法,该方法含有一个自适应惩罚参数校正方法,可以随个体的总数变化进行调整,它可以很快地脱离局部最优解而收敛于全局最优解。算法中只有很少的几个参数需要校正,故其比较简单且计算便捷。给出的仿真例子表明算法具有很好的收敛速度和很高的精确性。更多还原

【Abstract】 A novel penalty function method for constrained optimization problems using evolutionary programming is proposed. This method incorporates an adaptive penalty parameter -tuning algorithm that adjusts according to the landscape of population of individuals. It allows fast escape from local optima and fast convergence towards global optima. It is simple and computationally effective in the sense that only few penalty parameters are needed for tuning. Simulation results are presented to show that the proposed method performs well in terms of convergence speed and accuracy.更多还原