王〓泉.基于BP Garson组合模型的堤岸稳定影响因素敏感性分析[J].水电能源科学,2018,36(6):122-124
基于BP Garson组合模型的堤岸稳定影响因素敏感性分析
Sensitivity Analysis of Slope Stability Factor Based on BP Garson Model
  
DOI:
中文关键词:  堤岸稳定  敏感性分析  BP神经网络  Garson算法
英文关键词:slope stability  sensitivity analysis  BP neural network  Garson algorithm
基金项目:
作者单位
王〓泉 上海市水务建设工程安全质量监督中心站 上海 200070 
摘要点击次数: 93
全文下载次数: 
中文摘要:
      鉴于确定各因素(土体强度因素、边界几何因素)影响堤岸稳定的显著程度对研究堤岸失稳的重要意义,采用BP神经网络与Garson算法相结合的方法,对影响堤岸稳定的土体粘聚力、土体内摩擦角、坡高、坡比、墙前水位、墙后水位、墙后荷载7个因素进行了敏感性分析,先利用BP神经网络对非线性映射逼近能力强的优势,建立了影响因素与安全系数间的相关关系,再通过Garson算法,确定了各影响因素对安全系数的贡献量,最终确定了各因素影响堤岸变形的主次关系,为堤岸稳定影响因素敏感性分析提供了一种方法。
英文摘要:
      Considering that finding out the impact of various factors on the stability is of great significance for the instability study of the slope, combination of BP neural network with Garson algorithm was used to analyze the factor sensitivity on slope stability, including slope height, slope ratio, water level in front of the wall, water level behind the wall, soil cohesion, soil friction angle and wall load. First, the correlation between influencing factors and safety factor was established by using the advantages of BP neural network in approximation to nonlinear mapping. Second, the contribution of each factor to the safety factor was determined by Garson algorithm. Finally, the primary and secondary relationship of each factor was determined. 〖JP2〗This study provides a new method for the sensitivity analysis of embankment stability factors.
查看全文  查看/发表评论  下载PDF阅读器