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

Table 6

Prediction results of different models after PCA and CARS.

Model	Calibration model		Predictive model
		RMSECV (%)		RMSEP (%)
PCA-PLSR	0.967	2.13	0.965	2.30
PCA-1d-BPNN	0.988	1.32	0.985	1.53
PCA-2d-BPNN	0.990	1.22	0.986	1.46
CARS-PLSR	0.962	2.30	0.961	2.48
CARS-1d-BPNN	0.971	1.99	0.970	2.16
CARS-2d-BPNN	0.983	1.53	0.982	1.70