ANN Approach for State Estimation of Hybrid Systems and Its Experimental Validation

Table 1

Values of different ANN parameters.

Parameter	Value
ANN structure	NARX
Number of hidden layers	1
Hidden layer neurons	5
Hidden layer activation function	“tansigmoid”
Output layer activation function	“purelin”
Number of epochs	100
Number of exogenous inputs	4
Number of delayed inputs	0
Number of outputs	3
Number of feedback output delays	2
Training method	Back propagation
Training function	Levenberg–Marquardt
Performance function	Mean square error