基于AFSA-RBF模型的混凝土平板坝变形监测  

Monitoring of Concrete Slab Dam Deformation Based on AFSA-RBF Neural Network

作　　者：梁嘉琛 赵鲲鹏[1,2,3] 杨景文[1,2,3] LIANG Jiachen ZHAO Kunpeng YANG Jingwen(State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, China National Engineering Research Center of Water Resources Efficient Utilization and Engineering Safety, Hohai University, Nanjing 210098, China College of Water Conservancy and Hydropower, Hohai University, Nanjing 210098, China)

机构地区：[1]河海大学水文水资源与水利工程科学国家重点实验室,江苏南京210098 [2]河海大学水资源高效利用与工程安全国家工程研究中心,江苏南京210098 [3]河海大学水利水电学院,江苏南京210098

出　　处：《人民黄河》2016年第6期137-140,共4页Yellow River

基　　金：淮安市水利院士工作站资助项目;国家自然科学基金资助项目(51279052;51209077);江苏省杰出青年基金资助项目(BK20140039;BK2012036);江苏省"333高层次人才培养工程"项目(2017-B08037)

摘　　要：针对混凝土平板坝水平位移监测序列呈非线性变化的特点,采用经验模态分解(EMD)方法对混凝土平板坝水平位移监测序列进行分解,并采用计算最大信噪比的方法对信号进行去噪。面板坝的坝顶上下游方向水平位移主要受上下游水位和环境温度的影响,据此建立AFSA-RBF神经网络模型和RBF神经网络模型,对混凝土平板坝上下游方向水平位移进行预测,结果表明:AFSA-RBF模型能够很好地反映混凝土平板坝水平位移变化趋势和规律,预测结果有较高的精度,符合大坝安全监测的要求,可以在混凝土平板坝安全监测和评价中应用。In view of the nonlinear and cyclical change characteristics of the horizontal displacement observation data of concrete slab dam, this paper used the method of empirical mode decomposition （EMD） to decompose the horizontal displacement observation data of concrete slab dam and adopted the method of calculating the maximum SNR to denoise the signal. Concrete slab dam horizontal displacements of dam crest upstream and downstream direction mainly were affected by upstream and downstream water level and environment temperature. Based on this, it established AFSA-RBF and General RBF neural network model, and forecasted the horizontal displacement of concrete slab dam. The results show that AFSA-RBF model can reflect the tendency and rule of the concrete slab dam horizontal displacement change well. The predicted results have higher accuracy, which can meet the requirement of dam safety monitoring. The model can be applied in the safety monitoring and evaluation of concrete slab dam.

关 键 词：混凝土平板坝 水平位移 经验模态分解 AFSA-RBF神经网络 

分 类 号：TV642.51[水利工程—水利水电工程]