欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
建立有效的大坝变形监测模型可以反映大坝运行情况.相关向量机（RVM）具有良好的泛化能力,且适于解决非线性问题,尤其是高维数情况.在RVM理论的基础上,建立大坝变形监测模型,并以此研究其优化改进模型.改进思路如下：利用马尔科夫链适于解决监测数据波动大的优势处理模型残差;同时如何选择核参数会严重影响RVM模型的精度,采用一种改进的粒子群算法寻优核参数.通过实例比较多种优化模型发现,基于RVM理论建立模型的优化方法可大大提高预测的泛化能力及精度.
英文摘要：
Establishing an effective dam deformation monitoring model can reflect the operation of the dam. The relevance vector machine （RVM） has good generalization ability, and can solve high-dimensional non- linear problem effectively. This paper builds a dam deformation monitoring model based on RVM theory, and puts forward optimization models. Markov chain model is suitable for large fluctuations, so we deal with model residual using Markov chain, and how to choose kernel parameter affecting RVM model great- ly. In this paper, the choice of kernel parameter is based on an improved optimization of algorithm of PSO. Comparing with different optimization models through case stady, we could find that the optimization mod- els can greatly improve the generalization ability and prediction accuracy.
同期刊论文项目
同项目期刊论文
Blind source separation model of earth-rock junctions in dike engineering based on distributed optic
Evaluation of various temperature control schemes for crack prevention in RCC arch dams during const
Derivation Method for the Foundation Boundaries of Hydraulic Numerical Simulation Models Based on th
Analysis and back-analysis for temperature field of concrete arch dam during construction period bas
Rate effect on mechanical properties of hydraulic concrete flexural-tensile specimens under low load
A Monitoring Model for High Slope Displacement considering the Effects of Dynamic Structure Mutation
期刊信息
位置：