Impact of Noise on a Dynamical System: Prediction and Uncertainties from a Swarm-Optimized Neural Network

Table 1

Parameters used in the hybrid ANN+PSO algorithm.

ANN
NN-type	Feed-forward
Number of hidden layers	1
Transfer function	Tansig
Number of iterations	1500
Normalization range
Weight range
Bias range
Minimum error
PSO
Number of particles in swarm ()	50
Number of iterations ()	1500
Cognitive component ()	1.494
Social component ()	1.494
Maximum velocity ()	12
Minimum inertia weight ()	0.5
Maximum inertia weight ()	0.7
Objective function	RMSE