摘要
针对岩质边坡稳定性预测及加固效应具有的非线性和非明确性等特征,为了准确的预测稳定安全系数和加固效应,采用在实际工程中较为常见的二折线滑动面边坡概化模型,以边坡潜在滑块高度、上下滑动面倾角、坡面倾角、滑块面积,以及上下滑动面的黏聚力和摩擦系数作为影响因素,进行基于强度折减法的数值计算获得安全系数,得到足够数量的样本。然后利用遗传算法GA优化BP神经网络的权值和阈值,建立边坡稳定性的安全系数GA-BP预测模型,将其预测结果与标准BP神经网络进行对比分析。在此基础上,建立混凝土置换加固和锚索加固的方案优化方法。最后,将该方法应用于白鹤滩水电站泄洪洞出口边坡典型剖面在天然和开挖状态下的安全系数预测,并分别对混凝土置换加固和锚索加固进行了方案优化。结果表明:GA-BP神经网络模型精度更高,收敛速度更快,比使用标准BP网络模型的效果更优。研究成果对边坡稳定性评价和加固方案优化具有参考意义。
Aiming at the characteristics of nonlinearity and non-definiteness of rock slope stability prediction and reinforcement effect,a generalized model for the slope surface with two-fold line sliding plane normally used in the actual engineering practice is adopted to obtain the safety factor on the basis of carrying out the relevant strength reduction numerical calculation by taking the height of the potential sliding body of slope,slope inclined angle,slope surface inclined angle,area of sliding body as well as the cohesion and friction coefficient of the upper and lower sliding planes as the influencing factors,thus enough number of samples are obtained. Moreover,the weights and thresholds of BP neural network are optimized with genetic algorithm GA,and then a GA-BP model for predicting the safety factor of slope stability is established,from which the predicted result is compared with that from the standard BP neural network. On the basis of this,the method to optimize the schemes of both the replacement for concrete reinforcement and the anchorage reinforcement are proposed. Finally,this method is applied to the prediction of the safety factor of the typical slope section at the outlet of the spillway tunnel of Baihetan Hydropower Station under both the natural and the excavating statuses,and then the scheme optimization are made for both the replacement for concrete reinforcement and the anchorage reinforcement respectively. The result shows that the accuracy of the GA-BP neural network model is higher with fast convergence rate,from which the effect is better than that from the standard BP network model. The study result has certain referential significance for evaluating slope stability and optimizing the reinforcement scheme concerned.
作者
戴妙林
屈佳乐
刘晓青
李强伟
马永志
DAI Miaolin;QU Jiale;LIU Xiaoqing;LI Qiangwei;MA Yongzhi(College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098,Jiangsu, China)
出处
《水利水电技术》
CSCD
北大核心
2018年第5期165-171,共7页
Water Resources and Hydropower Engineering
基金
水利部公益性行业科研专项经费项目"土石坝长效安全运行重大关键技术研究"(201501033)
关键词
岩质边坡
GA-BP算法
安全系数预测
加固方案优化
rock slope
GA-BP algorithm
safety factor prediction
reinforcement scheme optimization
