Model for estimating the weight-loss ratio of damaged Korla fragrant pears

• DOI: 10.25165/j.ijabe.20241701.8300
• Publication Date: 2024-03-31T14:49:27Z
• Authors: Yang Liu; ; Yurong Tang; Hong Zhang; Hao Niu; Haipeng Lan

To predict the weight-loss ratio of Korla fragrant pears effectively, improve commodity value and study variation laws damaged during storage, this predicted by utilizing generalized regression neural network (GRNN), support vector (SVR), partial least squares (PLSR) error back propagation (BPNN). The prediction performances GRNN, SVR, PLSR BPNN models were compared comprehensively, optimal model was determined. In addition, verified. results show that increases gradually with extension storage time. During is positively related to degree damage. trained can be used pears. most accurate in predicting (R2=0.9929; RMSE=0.2138). It has also been proved have good predictive effect production practice (R2=0.9377, RMSE=0.7138). research findings provide references delivery quality time Keywords: pears, ratio, damages, DOI: 10.25165/j.ijabe.20141701.8300 Citation: Liu Y, Tang Y R, Zhang H, Niu Lan H P. Model for estimating Int J Agric & Biol Eng, 2024; 17(1): 261-266.