Table 8.

Configuration of the four trained models.

Case Study	CWR	C-MAPSS		NGTP *
Dataset		FD001	FD004
Hidden layers	2
Neurons per hidden layer	64-32
Activation functions	ReLU—ReLU
Learning rate	1 × 10−3
FS layer activation function	Tanh	Sigmoid	Linear	Tanh
Regularization rate	1 × 10−6	1 × 10−5	1 × 10−3	20,000
Epochs	2000	800	10,000	10,000

* NGTP: Natural Gas Treatment Plant.