Effects of Harvest Maturity and Storage Time on Storage Quality of Korla Fragrant Pear Based on GRNN and ANFIS Models: Part I Firmness Study

• Published in: FOOD SCIENCE AND TECHNOLOGY RESEARCH 2020, Vol.26(3), pp.363-372
• Main Authors: Niu, Hao, Liu, Yang, Wang, Zhentao, Zhang, Hong, Zhang, Yongcheng, Lan, Haipeng
• Format: Article
• Language: English
• Subjects: Adaptive systems; ANFIS; Artificial neural networks; Firmness; Fruits; Fuzzy logic; GRNN; korla fragrant pear; Mathematical models; Maturity; Neural networks; Regression analysis; Storage; storage quality
• ISSN: 1344-6606; 1881-3984
• DOI: 10.3136/fstr.26.363

Generalized-regression-neural-network (GRNN) and adaptive neuro-fuzzy inference system (ANFIS) models were utilized to develop mathematical models for assessing the effects of harvest maturity and storage time on the storage quality of Korla fragrant pear. The results showed that fruit firmness had high correlations with harvest maturity and storage time. At different harvest maturities, the firmness of fragrant pear decreased with increasing storage time; while under the same storage time, a lower harvest maturity resulted in a faster decrease in the fruit firmness of fragrant pear. The GRNN and ANFIS models were validated to be reliable in predicting the variation tendency of fruit firmness of fragrant pear. In addition, the ANFIS model with generalized bell (gbellmf) (RMSE = 0.161; R2 = 0.98) was proved to be the optimum model. The findings of this study can provide reference for the research and practice of the storage of fragrant pear.