摘要
遗传算法(GA)是模拟生物在自然环境中的遗传和进化过程而形成的一种自适应全局优化概率算法,然而在GA求解过程耗时较长,易出现早熟现象导致结果准确度低。根据GA传统算法与结合最速下降法和惩罚函数方法,提出求解非线性优化问题的混合遗传算法(HGA)。在无约束优化问题和约束优化两类问题中分别使用基于最速下降法的SHGA、基于惩罚函数法的(PHGA)进行求解。通过数值算例验证,表明HGA在非线性优化问题中比GA传统算法具有更快的收敛速度以及更好的最优解。
Genetic algorithm(GA)is an adaptive global optimization probability algorithm formed by simulating the genetic and evolution process of living beings in the natural environment.However,the GA solution process takes a long time and is prone to premature phenomena leading to low accuracy of the results.According to the traditional GA algorithm combined with the steepest descent method and the penalty function method,a hybrid genetic algorithm(HGA)for solving nonlinear optimization problems is proposed.In unconstrained optimization problems and constrained optimization problems,SHGA based on the steepest descent method and PHGA based on the penalty function method are used to solve the problems.Verification by numerical examples shows that HGA has faster convergence speed and better optimal solutions than traditional GA algorithms in nonlinear optimization problems.
作者
丁李
崇金凤
DING Li;CHONG Jin-feng(Huaibei Vocational and Technical College,Huaibei 235000,China;School of Mathematical Sciences,Huaibei Normal University,Huaibei 235000,China)
出处
《宜春学院学报》
2023年第3期45-48,共4页
Journal of Yichun University
基金
淮北职业技术学院院级重点科研项目“随机死亡率与利率下的生存年金长寿风险分析”(编号:2019-A-1)。
