改进的退火遗传优化策略应用研究

英文篇名：Research and application of improved simulated annealing genetic strategy
中文刊名：计算机工程与应用
英文刊名：Computer Engineering and Applications
作者：李政伟 ; 谭国俊
英文作者：LI Zheng-wei ; TAN Guo-jun.1.School of Computer Science and Technology ; China University of Mining and Technology ; Xuzhou ; Jiangsu 221116 ; China 2.School of Information and Electrical Engineering ; China University of Mining and Technology ; Xuzhou ; Jiangsu 221116 ; China
中文关键词：地震参数反演 ; 模拟退火遗传算法 ; 模拟退火算法 ; 遗传算法 ; 混合优化
英文关键词：seismic parameters inversion ; simulated annealing genetic algorithm ; simulated annealing algorithm ; genetic algorithm ; hybrid optimization
出版日期：2010-02-01
机构：中国矿业大学计算机科学与技术学院;中国矿业大学信息与电气工程学院;
年：2010
期：04
出版单位：计算机工程与应用

摘要

地震参数反演属于典型的非线性优化问题。针对遗传算法和模拟退火算法各自的优缺点,将改进的遗传算法与模拟退火算法相结合,提出了改进的退火遗传算法(ISAGA)。该方法通过筛选和修复进行初始种群的选择,采用允许父代参与竞争的退火选择机制,并根据模拟退火思想对交叉和变异概率进行自适应的调整,从而增加了种群的多样性并提高了收敛速度。该方法既具备了遗传算法强大的全局搜索能力,也拥有模拟退火算法强大的局部搜索能力。经理论模型试算结果表明,该方法不仅收敛速度快,优化精度高,抗干扰能力强,而且避免了局部收敛和依赖初始模型等问题,计算所得反演参数更接近于实际观测值。
Seismic inversion belongs to nonlinear optimum problem.An Improved Simulated Annealing Genetic Algorithm(ISAGA) for seismic parameters inversion by combining Modified Genetic Algorithm (MGA) with Simulated Annealing Algorithm (SAA) is developed.In the proposed method firstly the initial population is selected by filtering and restoring.Secondly the annealing selection mechanism is proposed which the elder population is permitted to compete.The new algorithm provides not only with strong global search capability of GA,but also with strong local search capability of SAA.Thirdly to improve the convergence speed and the diversity of the population the rates of crossover and mutation are modified self-adaptively.The simulation results indicate that the ISAGA can result in fast convergence rate and high optimization precision;moreover it avoids many shortcomings such as initial model sensitivity.The obtained section can well reflect the geology characteristics.

引文

[1]王宝珍,杨文采.用改进的遗传算法进行地震参数反演研究[J].石油地球物理勘探,1998,33(2):258-264.
    [2]彭真明.地震反演中的非线性优化方法及应用研究[J].成都理工大学,2001.
    [3]张宏兵,尚作萍,谭胜章.地震参数反演的快速模拟退火算法[J].河海大学学报,2005,33(4):434-437.
    [4]路鹏飞,杨长春,郭爱华.改进的模拟退火算法及其在叠前参数反演中的应用研究[J].地球物理学进展,2008,23(1):104-109.
    [5]李守巨,刘迎曦,陈昌林.基于混合遗传算法的混凝土大坝理学参数反演[J].大连理工大学学报,2004,44(2):195-199.
    [6]Wang Chung-Ho,Lu Jiu-Zhang.A hybrid genetic algorithm that op-timizes capacitated vehicle routing problems[J].Expert Systems with Applications,2008(12):415-419.
    [7]Xu Tian-ze,Wei Heng,Wang Zhuan-de.Study on continuous net-work design problem using simulated annealing and genetic algo-rithm[J].Expert Systems with Applications,2008(6):64-69.
    [8]朱建丰,徐世杰.基于自适应模拟退火遗传算法的月球软着陆轨道优化[J].宇航学报,2007,28(4):806-812.
    [9]汪鹏君,陆金刚,曾晓洋.基于整体退火遗传算法的低功耗最佳极性搜索[J].计算机辅助设计与图形学学报,2008,20(1):73-78.