Prediction of Slipper Pressure Distribution and Leakage Behaviour in Axial Piston Pumps Using ANN and MGGP

Table 2

Parameters of proposed ANN.

Parameters	Values
Training data set	4799
Validation data set	1028
Testing data set	1028
Number of hidden layers	2
Number of neurons in hidden layers	1-20
Activation functions	Tagsig-Logsig-Pureline
Number of epochs	10000
Learning rate	0.7
Architecture selection	Trial-Error
Target goal mean square error	10-7
Minimum performance gradient	10-7