实数编码自适应遗传算法辅助实现变形监测网平差  

Application of adaptive real-coded genetic algorithms in network adjustment for deformation monitoring

作　　者：陈超[1,2] 张献州[2] 

机构地区：[1]四川中水成勘院测绘工程有限责任公司,成都610072 [2]西南交通大学地球科学与环境工程学院,成都610031

出　　处：《测绘科学》2014年第6期117-120,共4页Science of Surveying and Mapping

摘　　要：变形监测网的平差问题实际上是一种带约束条件的函数优化问题。本文从附加带权基准方程的变形监测网平差统一模型入手,探讨了实数编码遗传算法在变形监测网非线性参数平差应用中的有效性问题:针对平差基准的等式约束条件,通过等式变换对平差参数进行降维处理,达到了约束优化问题向无约束优化问题的转变,实现了不同基准条件下变形监测网的非线性遗传算法参数平差。Adjustment problem of deformation monitoring is a kind of function optimization problem with constraints. This article from the unified adjustment model which had additional benchmark equation with weight of deformation monitoring network, discussed the effectiveness of real-coded genetic algorithms applied in nonlinear parameter adjustment of deformation monitoring network. According to the equality constraint conditions of adjustment datum, this article achieved the transition from constrained optimization problem to unconstrained optimization problem through parameter descending dimension by equation transformation, and realized the nonlinear parameter adjustment by genetic algorithms of deformation monitoring network under the different datum conditions.

关 键 词：实数编码遗传算法 变形监测 平差基准 非线性遗传算法参数平差 参数降维 

分 类 号：P258[天文地球—测绘科学与技术]