草莓果实腐烂指数预测模型的品种和采收期适应性分析

Adaptability Analysis of Strawberry Cultivar and Harvest Time to the Predictive Model for Fruit Decay Index

为验证草莓果实基于乙醇质量分数的果实腐烂指数预测模型的适用性,建立了不同采摘时期的‘丰香’和‘红艳’草莓基于乙醇质量分数的果实腐烂指数预测模型。通过对不同预测模型的参数(斜率和截距)进行差异显著性分析,发现不同品种草莓果实的模型参数差异显著,不同采摘期的同一品种草莓果实的模型参数差异不显著,即模型受果实品种的影响,而与采摘期无关。用与建模相独立的试验数据对模型进行验证。结果表明:预测的相对误差(RE)分别为8.82%、4.72%、14.64%、10.35%,预测精度较高,说明基于乙醇质量分数的草莓果实腐烂指数预测模型对不同品种和采摘期的草莓果实具有较好的适用性。该模型为应用电子鼻实时监测草莓流通过程中果实腐烂状况奠定了基础。

In this study,a universal predictive model for decay index based on ethanol content in strawberry was proposed.Based on this universal model,a separate predictive model for strawberry cultivars Fengxiang and Hongyan harvested at two different time points was developed and validated.The analysis of variance of the model parameters（slopes and intercepts） of different predictive models showed a significant difference between both cultivars but no significant difference between both harvest time points for the same cultivar,suggesting that the predictive models were affected by cultivar rather than harvest time.The relative errors of these four predictive models observed when validated using independent experimental data were 8.82%,4.72%,14.64% and 10.35%,respectively,indicating high prediction accuracy.Thus,different strawberry cultivars and different harvest times had high adaptability to the predictive models.In conclusion,predictive modeling can provide a theoretical basis for non-destructive real-time measurement of fruit quality using gas sensor array.