An Adaptive Genetic Algorithm Optimizes Double-Hidden Layer BPNN for Rapid Detection of Moisture Content of Green Tea in Processing

Table 7

Results of the AGA-BPNN model optimized by combining PCA and CARS.

Model	Calibration model		Predictive model
		RMSECV (%)		RMSEP (%)
AGA-1d-BPNN	0.982	1.58	0.981	1.71
AGA-2d-BPNN	0.992	1.06	0.986	1.51
PCA-AGA-1d-BPNN	0.993	0.95	0.990	1.23
PCA-AGA-2d-BPNN	0.995	0.81	0.994	1.01
CARS-AGA-1d-BPNN	0.977	1.75	0.974	2.02
CARS-AGA-2d-BPNN	0.989	1.25	0.984	1.56