Academia.eduAcademia.edu

Figure 2 – uploaded by Mohamed Hasnain Isa

See full PDFdownloadDownload figure
TABLE 2: Performance statistics of the ANN model during training and testing (soil depth = 0.5 m).

Table 2 Performance statistics of the ANN model during training and testing (soil depth = 0.5 m).

Related Figures (12)

FIGURE 1: Schematic representation of a multilayer feed forward (MLP) network showing activity levels generated by the neurons in different layers.
FIGURE 1: Schematic representation of a multilayer feed forward (MLP) network showing activity levels generated by the neurons in different layers.
Ficure 3: A trace of time series of rainfall and pore-water pressures at 0.5m soil depth showing pore-water pressure responses to climatic changes. (Data ranges used for training and testing of the ANN model are shown in figure inset.)
Ficure 3: A trace of time series of rainfall and pore-water pressures at 0.5m soil depth showing pore-water pressure responses to climatic changes. (Data ranges used for training and testing of the ANN model are shown in figure inset.)
FiGurE 2: Schematic diagram of typical arrangement of instrumen- tation and location of the study slopes adapted from Mustafa et al., 2012.
FiGurE 2: Schematic diagram of typical arrangement of instrumen- tation and location of the study slopes adapted from Mustafa et al., 2012.
*N = number of observations; Max. = maximum; Min. = minimum; o = standard deviation; 4 = sample mean; of = variance; CV = coefficient of variation. TABLE 1: Normal statistics of data used for training and validating the ANN model (soil depth = 0.5 m).
*N = number of observations; Max. = maximum; Min. = minimum; o = standard deviation; 4 = sample mean; of = variance; CV = coefficient of variation. TABLE 1: Normal statistics of data used for training and validating the ANN model (soil depth = 0.5 m).
FiGure 4: (a) Cross-correlation between pore-water pressure and rainfall. (b) Autocorrelation of pore-water pressures
FiGure 4: (a) Cross-correlation between pore-water pressure and rainfall. (b) Autocorrelation of pore-water pressures
Ficure 5: Comparison between observed and trained time series of pore-water pressures obtained from training the ANN model.
Ficure 5: Comparison between observed and trained time series of pore-water pressures obtained from training the ANN model.
FiGurE 6: Comparison between observed and predicted time series of pore-water pressure obtained using the ANN model.
FiGurE 6: Comparison between observed and predicted time series of pore-water pressure obtained using the ANN model.
FicurE 7: Comparison between observed and trained pore-wate1 pressures obtained during training the ANN model.
FicurE 7: Comparison between observed and trained pore-wate1 pressures obtained during training the ANN model.
FicureE 9: Time and epochs required by the ANN model to reach desired accuracy.
FicureE 9: Time and epochs required by the ANN model to reach desired accuracy.
Ficure 8: Comparison between observed and predicted pore-water pressures obtained using the ANN model.
Ficure 8: Comparison between observed and predicted pore-water pressures obtained using the ANN model.
TABLE 3: Performance statistics of ANN at different soil depths during testing stage.
TABLE 3: Performance statistics of ANN at different soil depths during testing stage.
Related topics:
GeographyHydrogeologySoil SciencePore Water PressureArtificial neural network modelingSpatial and Temporal Variations
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved