Abstract: Considering the uncertainty and complexity of dynamic parameters of a computational model used in the flyash dam simulation, a predictive model of the dynamic parameters of ash dam simulation is established. The sampling pool used in the predictive model is obtained from laboratory tested soil parameters. The model also adopts the predictability of the Back Propagation (BP) neural network. The dynamic response of a flyash dam of a power plant was analyzed and the numerical results were compared with the measured results. This analytical approach provides useful recommendations for the response of a flyash dam under a seismic condition.

Key words: Flyash Dam, Back Propagation (BP) Neural Network, Predictive Model, dynamic FEM