中国食品科学技术学会第十二届年会暨第八居中美食品业高层论坛论文摘要集【610】核磁共振技术结合化学计量学对鲍鱼水分和脂肪含量的预测高兴盛谭明乾*(大连工业大学食品学院辽宁大连116034)摘要自的：建立一种基于核磁共振技术的无损快速预测鲍鱼水分和脂肪含量的方法。方法：采 用CPMG序列测定鲍鱼横向弛豫时间T2，再利用核磁共振数据结合主成分回归分析（PCR)和偏最 小二乘回归法（PLSR)建立两种关于鲍鱼水分和脂肪含量的预测模型，预测结果以模型决定系数 R2和均方根误差（RMSE)对定量预测模型进行评价。结果：发现鲍鱼肉中含有三种不同水组分， 分别是，结合水（T2b)、不易流动水（T21)和自由水（T22)。PCR水分预测模型表明校正集和交 互验证的相关系数R2分别为0. 9946和0. 9941,脂肪预测模型校正集和交互验证的R2分别为0. 9679 和0_9616，交互验证均方根误差（11]\\13£(^)分别为0.14118和0.28998;?1^11水分预测模型校正 集和交互验证的相关系数R2分别为0. 9947和0. 9931，脂肪预测模型校正集和交互验证的R2分别为 0• 9727和0• 9656,交互验证均方根误差（RMSECV)分别为0• 1441g和0. 2719g, R2越大，RMSE 越小，获得的模型效果越好。讨论：低场核磁共振技术结合化学计量学即主成分回归分析（PCR) 和偏最小二乘回归法（PLSR)有望提供一种无损快速预测鲍鱼样品中水分和脂肪含量的方法。关键词低场核磁共振鲍鱼化学计量学预测Predicting moisture content and fat content in abalone by NMR combined with chemometricsGao Xingsheng Tan Mingqian *(School of Food Science and Technology, Dalian Polytechnic University, Dalian 116034, Liaoning)Abstract Purpose ： To establish a fast and non-destructive moisture and fat content prediction method of abalone based on the nuclear magnetic resonance (NMR) technique. Methods： CPMG sequence was employed to determine the transverse relaxation time T2 of abalone. The nuclear magnetic resonance (NMR) data combined with principal component regression ( PCR) and partial least squares regression ( PLSR) was used to establish two kinds of prediction model for abalone moisture and fat content. The model coefficient R2 and root mean square error ( RMSE) were used to evaluate the prediction results. Results ： Three water populations, the combined water flowing water ( T2b ) , not easily flowing ( T21 ) and free water (T22 ) , in abalone meat were observed. The PCR moisture prediction model showed that the calibration set and correlation coefficient R2 interaction validation were 0. 9946 and 0. 9941, respectively. The fat prediction model calibration set and R2 interaction validation were 0. 9679 and 0. 9616, respectively. The interac-• 321 •Abstracts of Food Summit in China 2015 & 12th Annual Meeting of CIFSTtive authentication root mean square errors (RMSECV) were 0. 1411 and 0.2899 g, respectively. The PLSR moisture prediction model calibration set and correlation coefficient R2 interaction validation were 0. 9947 and 0. 9931, respectively. The fat prediction model calibration set and R2 interaction validation were 0. 9727 and 0. 9656 , respectively. The interactive authentication root mean square errors ( RMSECV) were 0. 1441 g and 0. 1441 g, respectively. The greater the R2 is, the smaller the RMSE, and the better the model is. Discussion： The nuclear magnetic resonance (NMR) technology combined with chemomet- rics of principal component regression (PGR) and partial least squares regression (PLSR) may have a potential for the rapid and non-destructive prediction of abalone moisture and fat content.Key words Nuclear magnetic resonance, Abalone, Chemometrics, Prediction[611]光化学比色阵列传感器在鱼鲜度评价中的应用孙文1冯亮2王际辉〃C大连工业大学食品学院辽宁大连116034 2中国科学院大连化学物理研究所辽宁大连116023)摘要鱼肉类食品的鲜度是衡量其品质的重要指标，研究鱼体新鲜度的快速检测技术对于鱼和鱼 类制品的加工储运以及上市销售有着重要的指导意义。针对目前现有的新鲜度快速检测方法存在的 检测灵敏度低等问题，本研究基于光化学比色传感器阵列的方法，以溶肢-凝肢为材料对指示剂进 行固载并通过酸碱调控提高指示剂的灵敏度，以期实现对鱼肉腐败早期的检测。实验选取了八种 pH指示剂分别进行酸或碱调控到各自的pH突变点附近以增加其反应的灵敏度，与未调节的组成一 个4x4的阵列用于鱼鲜度的区分。对新鲜鲅鱼在SOT下放置24h内的不同时间段进行检测，将反 应前后阵列颜色的变化差值进行分层聚类分析可以成功的区分出不同放置时间的鱼。此外可以看出 经过酸碱调控的指示剂颜色变化程度要明显大于未经调节的指示剂。关键词鱼新鲜度检测pH指示剂灵敏度比色传感器Application of Photochemistry Colorimetric Sensor Array in Fish Freshness EvaluationSun Wen1 Feng Liang2 Wang Jihui1 *(l School of Food Science & Technology, Dalian Polytechnic University, Dalian 116034 , Liaoning 2 Dalian Institute of Chemical Physics, Chinese Academy of Science, Dalian 116023 , Liaoning)Abstract Freshness is an important index to measure food quality, rapid detection technology for the pro-• 322 •