计算机工程与应用
計算機工程與應用
계산궤공정여응용
COMPUTER ENGINEERING AND APPLICATIONS
2015年
9期
62-67
,共6页
刘云连%伍铁斌%王俊年%周桃云%成运
劉雲連%伍鐵斌%王俊年%週桃雲%成運
류운련%오철빈%왕준년%주도운%성운
蝙蝠算法%自适应罚函数%约束优化问题
蝙蝠算法%自適應罰函數%約束優化問題
편복산법%자괄응벌함수%약속우화문제
bats algorithm%adaptive penalty function%constrained optimization problem
设计了一种基于自适应罚函数法和改进蝙蝠算法的约束优化问题求解方法。提出了一种自适应罚函数法，该处理方法综合考虑了约束违反的情况和进化过程的特点，如果某个约束违反的次数越多，则证明该约束越强，赋予惩罚系数越大；种群中的不可行解的数量越多，为保持种群的多样性，则约束应该取较小的值，即惩罚系数取较小的值。提出了一种改进的蝙蝠算法，利用混沌的遍历性特点产生初始种群，增强了初始种群的多样性和种群的质量；在考虑了脉冲响度的蝙蝠算法局部搜索中，融入了交叉操作；为防止算法在后期陷入局部最优解，引进了变异操作，保证了群体的多样性。将自适应罚函数法与改进的蝙蝠算法融合起来求解约束优化问题，4个复杂的标准测试函数和2个工程实际问题证明了该约束优化求解方法的可行性和有效性。
設計瞭一種基于自適應罰函數法和改進蝙蝠算法的約束優化問題求解方法。提齣瞭一種自適應罰函數法，該處理方法綜閤攷慮瞭約束違反的情況和進化過程的特點，如果某箇約束違反的次數越多，則證明該約束越彊，賦予懲罰繫數越大；種群中的不可行解的數量越多，為保持種群的多樣性，則約束應該取較小的值，即懲罰繫數取較小的值。提齣瞭一種改進的蝙蝠算法，利用混沌的遍歷性特點產生初始種群，增彊瞭初始種群的多樣性和種群的質量；在攷慮瞭脈遲響度的蝙蝠算法跼部搜索中，融入瞭交扠操作；為防止算法在後期陷入跼部最優解，引進瞭變異操作，保證瞭群體的多樣性。將自適應罰函數法與改進的蝙蝠算法融閤起來求解約束優化問題，4箇複雜的標準測試函數和2箇工程實際問題證明瞭該約束優化求解方法的可行性和有效性。
설계료일충기우자괄응벌함수법화개진편복산법적약속우화문제구해방법。제출료일충자괄응벌함수법，해처리방법종합고필료약속위반적정황화진화과정적특점，여과모개약속위반적차수월다，칙증명해약속월강，부여징벌계수월대；충군중적불가행해적수량월다，위보지충군적다양성，칙약속응해취교소적치，즉징벌계수취교소적치。제출료일충개진적편복산법，이용혼돈적편력성특점산생초시충군，증강료초시충군적다양성화충군적질량；재고필료맥충향도적편복산법국부수색중，융입료교차조작；위방지산법재후기함입국부최우해，인진료변이조작，보증료군체적다양성。장자괄응벌함수법여개진적편복산법융합기래구해약속우화문제，4개복잡적표준측시함수화2개공정실제문제증명료해약속우화구해방법적가행성화유효성。
A solving method for constrained optimization problem based on adaptive penalty function and improved bats algorithm is designed. An adaptive penalty function method is proposed, which both takes the circumstances of constraint violations and characteristics of evolutionary process into consideration. The more frequently a constraint is violated, the more powerful it is, the larger penalty coefficient is given to it. The more infeasible solutions in the population, the smaller the constrain should be, in other words, the smaller the penalty coefficient should be, in order to keep the diversity of the population. An improved bats algorithm is proposed, which generates the initial population by using the ergodicity of chaos, and enhances the quality of the initial population and diversity of population. In the local search of bats algorithm which takes the pulse loudness into consideration, crossover operation is added. In order to prevent the algorithm from falling into local optimal solution in the late, variation operation is added, which ensures the diversity of the population. Then adaptive penalty function and improved bats algorithm are mixed to solve constrained optimization problem, and 4 complex stan-dard test functions and 2 practical engineering problems prove the feasibility and effectiveness of the solving methods for constrained optimization problem.
