CASIA OpenIR  > 毕业生  > 硕士学位论文
 复杂生态系统叠加建模的搜索计算方法 刘颖 Subtype 工学硕士 Thesis Advisor 何善堉 1994-05-01 Degree Grantor 中国科学院自动化研究所 Place of Conferral 中国科学院自动化研究所 Degree Discipline 控制理论与控制工程 Keyword 生态系统建模 叠加原理 全局最优化方法 Modeling Of Ecological Systems Superposition Procedure Global Optimization Algorithm Abstract 在数学生态学研究中，对真实生态系统的建模是进一步研究的基础，有 着十分重要的意义．叠加原理作为一种实用的复杂生态系统建模方法，具有 较高的应用价值和广泛的适用领域．由于实际生态系统的复杂性，使得在实 现叠加原理时，基于线性模型的传统参数估计方法难以实现或者难以取得 令人满意的结果．本文在分析了生态系统模型的特点后，提出采用全局最优 化方法直接搜索待确定的参数，以此实现叠加原理．由于全局最优化方法不 需要了解目标函数的详细待性，所以具有较高的鲁棒性，很适合于解决生态 模型这类对参数极为敏感的非线性模型的参数估计问题．本文详细讨论了 采用全局最优化方法实现叠加原理的可行性，实现方法以及实现中应注意 的问题，并给出了一系列针对模拟系统和实验观测效据的建模实例．本文也 提出了一种全局最优化算法一随机二分法 Other Abstract In the study of mathematical ecology, the modeling of natural ecosystem, which is the foundation of further research, is of great importance. The Superposition Procedure is of highly value in practice and has extensive applicable area. Owing to the complexity of natural ecological systems, the implementation of Superposition Procedure by using classical parameter estimation methods is difficult and can seldom obtain satisfactory results. In this paper, after analyzing characteristics of ecological models, it is proposed that the global optimization method can be applied to estimate the parameters of ecological models. Because the specific information of objective functions is not needed when applying global optimization methods , this kind of optimization method is rather robust and therefore is suitable to estimate the parameters of nonlinear models, whose responses are very sensitive to their parameters. Besides the possibility of the implementation and techniques, some simulate results are also presented . A new global optimization algorithm - Random Bisection Algorithm - is presented in this paper as well. shelfnum XWLW322 Other Identifier 322 Language 中文 Document Type 学位论文 Identifier http://ir.ia.ac.cn/handle/173211/7103 Collection 毕业生_硕士学位论文 Recommended CitationGB/T 7714 刘颖. 复杂生态系统叠加建模的搜索计算方法[D]. 中国科学院自动化研究所. 中国科学院自动化研究所,1994.
 Files in This Item: There are no files associated with this item.
 Related Services Recommend this item Bookmark Usage statistics Export to Endnote Google Scholar Similar articles in Google Scholar [刘颖]'s Articles Baidu academic Similar articles in Baidu academic [刘颖]'s Articles Bing Scholar Similar articles in Bing Scholar [刘颖]'s Articles Terms of Use No data! Social Bookmark/Share