DEEP EXCAVATION MULTI-POINT DISPLACEMENT MONITORING MODEL AND DETERMINATION OF RBF CENTER BY FCM
-
摘要: 为建立合理的基坑多测点位移监测模型,采用径向基函数神经网络(RBF)为基本框架,从位移力学机制选择网络输入层,以相关联的多个测点位移为输出层,发挥RBF网络非线性映射功能的同时,根据基坑的开挖进展和位移特征,采用有针对性的预选RBF计算中心与模糊C均值聚类(FCM)算法,共同确定计算中心。实例表明,该计算方法更具合理性,且能获得理想的训练和预测效果。Abstract: To establish reasonable deep excavation multi-point displacement monitoring model,radial basis function artificial neural network(RBF)was taken as frame.Its input layer came from displacement mechanical theory,and output layer was formed by interrelated multi-point displacement.Considering excavation and displacement characteristics,special preselecting RBF centers and Fuzzy C-means Algorithm(FCM)were used together to confirm RBF centers.Instances showed that these methods were more reasonable and possed good training and forecasting results.
-
Key words:
- multi-point /
- deep excavation /
- displacement monitoring model /
- preselecting RBF centers /
- FCM
-
黄铭. 数学模型与工程安全监测[M].上海:上海交通大学出版社,2008. 陈斌,黄铭. 开挖作用下深基坑变形监测数学模型研究[J].建筑技术开发,2006,(01):21-23,39. 王维斌,赵新华. 基于RBF神经网络的活性污泥模型的应用[J].南开大学学报(自然科学版),2008,(02):91-97. Kim Dongwon,Huh Sung-Hoe,Seo Sam-Jun. Self-Organizing Radial Basis Function Network Modeling for Robot Manipulator[A].2005.579-587. 张秀玲,陈丽杰,季颖. 基于径向基函数神经网络的板形模式识别研究[J].工业仪表与自动化装置,2009,(03):7-9. 刘笛,朱学峰,苏彩红. 一种新型的模糊C均值聚类初始化方法[J].计算机仿真,2004,(11):148-151. 邓广慧,唐贤瑛,夏卓群. 基于FCM和RBF网络的入侵检测研究[J].电脑与信息技术,2006,(01):6-8,62. 黄铭,江军,褚伟洪. 考虑堆载预压进程的地基沉降监测分析[J].路基工程,2008,(04):60-61. 黄铭,葛修润,刘俊. 大坝安全监测的多点位移向量模型[J].上海交通大学学报,2001,(04):514-517.
点击查看大图
计量
- 文章访问数: 80
- HTML全文浏览量: 9
- PDF下载量: 46
- 被引次数: 0