It includes three folders, namely 1D S-V, LSTM and RC-ESN. (1) 1D S-V: 1D S-V.m is the matlab code used to solve the one-dimensional dam break; 1D S-V data.mat is the solution result; ic0_H.mat is the data set used for training and testing. (2) LSTM: LSTM.ipynb is the code of LSTM method; lookback4_hd800_nl2_Shift_28.mat is the numerical solution, predicted value, RMSE and ACC of 28 test sets; training set size.xlsx records the predicted horizon and predicted average RMSE under different training set sizes. (3) RC-ESN: RC-ESN.ipynb is the code of RC-ESN method; Res_size1400_Rd_0.1_Shift_number_28.mat is the numerical solution, predicted value, RMSE and ACC of 28 test sets; Res_size_25_Rd_0.1_Shift_number_1.mat is the RMSE under different reservoir sizes; Res_size_1400_Rd_100_Shift_number_1.mat is the RMSE under different spectral radius; RMSE.xlsx records the predicted horizon and predicted average RMSE under different traininf set sizes.
-
Notifications
You must be signed in to change notification settings - Fork 0
License
lcl1527/Data-driven-and-machine-learning-based-prediction-of-wave-propagation-behavior-in-dam-break-flood
About
No description, website, or topics provided.
Resources
License
Stars
Watchers
Forks
Packages 0
No packages published