Table 5 Parametrization of artificial neural network for steam turbine model.

From: Comparative behavior of steam turbine model for dynamical power system analyses by means of multiple fractional and artificial neural network techniques

Description of parameters

Estimation of values

Input neurons

05 (Pressure, flow rate, time, fractional, fractal)

Hidden neurons

08 (Selection is based on minimum MSE)

Output neuron

01 (ratio of output to input as a dynamic response)

Training function

Levenberg-Marquardt optimization

Activation function

Rectified linear unit function

Performance function

Mean square error and correlation

Cycle

Training, validation, and testing

Back to article page