Research On Apple Chips-processing Suitability Evaluation And Its Correlations With Characteristic Components Of Raw Material

In order to improve the accuracy and applicability of the apple chips-processing suitability evaluation model,34 fresh apple fruit samples of 21 main cultivars from 7 main planting areas were collected and corresponding apple chips were prepared.The BP neural network algorithm was used to construct learning models,which could greatly improve the prediction accuracy of chips-processing suitability for apple raw materials outside the modeling sample.In order to explore the quantitative relationship between the main qualities of apple chips and the characteristic components of raw materials,19 apple raw material samples with large differences in the quality of chips were screened as experimental samples.A variety of linear and nonlinear statistical methods were used to identify the key fruit factors affecting the main qualities of chips,and several quantitative association models were constructed.Finally,the goal of basic substances analysis of chips quality formation from the perspective of data correlation was realized.The main findings of the study were as follows:(1)Chips were prepared by instant controlled pressure drop(DIC,French for détente instantannée controlee,also known as explosion puffing)and 17 indicators were measured.The core indexes of chips were selected by factor analysis and correlation analysis,namely L* value,brittleness,puffing degree,titratable acid,soluble sugar and crude protein.The weights of the core indexes were determined by the analytic hierarchy process,and then the comprehensive quality evaluation scores and grades of chips were calculated.(2)22 indicators of 34 fruit samples of different cultivars and regions were measured.Then the characteristic indicators of apple fruits related to chips qualities were screened out by correlation analysis between apple fruit indicators and chips core indexes,namely shape index,a* value(pulp),pH value,titratable acid content,Vc content,proportion of core,protein content,b* value(pulp),density,soluble solids content,crude fiber content and total sugar content.Learning model with input of fruit characteristic indicators and output of chips comprehensive evaluation scores was established by database of 29 apple samples.5 apple samples were chosen as test samples to verify the prediction accuracy of the learning model.The prediction accuracy of the three learning models obtained by transforming the learning samples was above 90%.(3)19 apple cultivars with large differences in chips qualities were used as research objects.Fresh fruit basic indicators,characteristic indicators(monosaccharides,organic acids,amino acids,monophenols,and pectin)and chips quality(color,texture,and flavor)indicators were determined,and then multiple correlation analysis methods were applied to construct quantitative association models between hips quality indexes and the raw material characteristic indexes.At 0.05 level,the chips quality indexes,namely L* value,a* value,b* value,brittleness,hardness,soluble solids,soluble sugar and titratable acid,and their relevant fruit factors were determined by correlation analysis and stepwise regression.The association analysis was further studied based on big data's regression model and BP neural network model.The verification results showed that the coefficient between the theoretical value and the experimental value exceeds 0.88,which indicated both developed models can forecast effectively.Furthermore,indicated that the selected fruit factors were reasonable and the constructed association models exhibited good fitting effect.It was reasonable and intuitive to use the quantitative correlation analysis method to determine the relationship between main qualities of chips and characteristic components of apple raw material.