基于質(zhì)構(gòu)儀TPA模式構(gòu)建蛋白腸感官評分預測模型及應用

朱美淋 張貞煒 余杰 黃瀟漪 蔣雨含 邢沁澮 賈利蓉

摘要：為建立一套客觀、數(shù)據(jù)化評價動植物雙蛋白腸感官綜合評分的方法，該研究采用質(zhì)構(gòu)剖面分析（TPA）測定24 組不同配方的樣品，結(jié)合感官評價方法，通過逐步回歸分析建立感官綜合評分預測模型，并將其應用于響應面實驗以優(yōu)化雙蛋白腸的配方。結(jié)果表明，動植物雙蛋白腸的感官綜合評分預測模型為Y=－68.110＋0.003×硬度＋79.119×彈性－0.001×咀嚼性－18.393×回復性，該模型相關(guān)系數(shù)R2為0.902，校正決定系數(shù)RAdj2為0.875；響應面回歸模型相關(guān)系數(shù) R2為0.962，校正決定系數(shù) RAdj2為?0.924，預測系數(shù)RPred2為0.861，優(yōu)化后蛋白腸的最佳配方為大豆拉絲蛋白質(zhì)量分數(shù)74%、豬肉質(zhì)量分數(shù)16%、大豆分離蛋白與花生蛋白粉比例7∶3、豬肉肥瘦比例7∶3，該配方下產(chǎn)品的感官評分實際值為8.51，與理論預測值（8.45）誤差較小。該研究得到的動植物雙蛋白腸感官綜合評分預測模型的擬合度較高，實測值與預測值的吻合度較好，模型在響應面實驗中具有可行性，為客觀、快捷地評價新型肉腸的感官品質(zhì)提供了新思路。

關(guān)鍵詞：動植物雙蛋白腸；質(zhì)構(gòu)剖面分析；感官綜合評分；回歸模型；響應面實驗

中圖分類號：TS251.65 ?????文獻標志碼：A ????文章編號：1000-9973（2024）04-0044-06

Construction of Prediction Model for Sensory Score of Protein Sausages

Based on TPA Model of Texture Analyzer and Its Application

ZHU Mei-lin1， ZHANG Zhen-wei1， YU Jie2， HUANG Xiao-yi2，

JIANG Yu-han2，?XING Qin-hui3，?JIA Li-rong1*

（1.College of Biomass Science and Engineering， Sichuan University， Chengdu 610065， China;

2.Yibin Industrial Technology Research Institute of Sichuan University， Yibin 644000，

China; 3.New Hope Group Co.， Ltd.， Chengdu 611400， China）

Abstract： In order to establish an objective and data-based method for evaluating comprehensive sensory scores of animal and plant double protein sausages， in this study， texture profile analysis （TPA） is used to determine 24 groups of samples with different formulas， and a prediction model for comprehensive sensory scores is established by stepwise regression analysis combined with sensory evaluation method， which is applied to response surface experiment to optimize the formula of double protein sausages. The results show that the prediction model for comprehensive sensory scores of animal and plant double protein sausages is Y=－68.110＋0.003×hardness＋79.119×elasticity－0.001×chewiness－18.393×resilience， with the correlation coefficient R2 of 0.902 and the adjusted determination coefficient RAdj2 of 0.875. The correlation coefficient R2 of the response surface regression model is 0.962， the adjusted determination coefficient RAdj2 is 0.924， and the predicted coefficient RPred2 is 0.861. The optimal formula of protein sausages is optimized as follows： drawing soy ?protein mass fraction is 74%， pork mass fraction is 16%， the ratio of soy protein isolate to peanut protein powder is 7∶3 and the ratio of pork fat to lean is 7∶3. The actual value of sensory score of the product prepared under such formula is 8.51， which has a small error with ?the theoretical predicted value （8.45）.The prediction model for comprehensive sensory scores of animal ?and

收稿日期：2023-10-21

基金項目：2020年四川大學-宜賓市校市戰(zhàn)略合作專項資金項目（2020CDYB-1）

作者簡介：朱美淋（1998—），女，碩士，研究方向：健康食品工程。

*通信作者：賈利蓉（1972—），女，教授，博士，研究方向：農(nóng)產(chǎn)品加工及貯藏。

plant double protein sausages has a high fitting degree， and the actual value is in good agreement with the predicted value. The model is feasible in response surface experiment， which has provided a new idea for the objective and rapid evaluation of the sensory quality of new meat sausages.

Key words： animal and plant double protein sausages; texture profile analysis; comprehensive sensory score; regression model; response surface experiment

肉腸類產(chǎn)品包括香腸、薩拉米腸、臘腸等，是消費者喜聞樂見的產(chǎn)品，此類產(chǎn)品膽固醇、飽和脂肪酸等含量較高，隨著消費者對健康的日益重視，用部分植物蛋白替代動物肉的雙蛋白腸產(chǎn)品逐漸進入人們的視野。……