利用建立多元回归模型的方法对大坝的垂直位移进行预测,往往因为数学模型的局限性和对影响因子分析的不全面导致预测结果不准确.利用BP神经网络良好的非线性问题处理能力和自学习功能,通过训练神经网络,对水电站坝体的垂直位移进行了有效的预测,得到与实测值相对误差小于1％的预测结果,从而实现对大坝更为可靠的安全监测.%The method of using a multiple regression model on prediction the dam' s vertical displacement often result in inaccurate forecasting because of the limitations of mathematical model and the inaccurate analysis on the affecting factors.By training BP Neural Network which has powerful ability in nonlinear problems processing and good self-learning function,an effectively prediction was conduct on the dam' s vertical displacement of hydropower station,makes relative error of the forecast results less than 1％,so as to achieve a more reliable dam safety monito
展开▼
