Prediction Model of Concrete Dam Deformation Based on Adaptive Unscented Kalman Filter and BP Neural Network
2014 ◽
Vol 513-517
◽
pp. 4076-4079
◽
Keyword(s):
By using unscented kalman filter (UKF) theory and introducing adaptive factor into BP neural network, a new prediction model of concrete dam deformation was proposed. Example shows that this model can improve the convergence speed of BP neural network, and the calculation precision of this model meets engineering requirements. Meanwhile, this model can be applied in the safety monitoring of other hydraulic engineering structure.