Font Size: a A A

Study On Rapid Detection Of Olive Oil Quality Based On Terahertz Spectroscopy

Posted on:2019-10-01Degree:MasterType:Thesis
Country:ChinaCandidate:J J YuFull Text:PDF
GTID:2371330545996964Subject:Food Science
Abstract/Summary:PDF Full Text Request
The quality and safety of food is closely related to people's healthy diet.With the improvement of people's living standard and the frequent occurrence of food safety incidents,food quality and safety detection has drawn more and more attention.Most of the extra virgin olive oil(EVOO)on the Chinese market comes from imports,and its origin,processing and production are uncontrollable,so the origin,quality and safety are deeply concerned by consumers.Our study chose the extra virgin olive oil as the experimental object,and used the terahertz time-domain spectroscopy combined with chemometrics to do the analysis of origin identification,adulteration and oxidative rancidity detection.The specific research contents and results are as follows.(1)Study on the geographic origin of extra virgin olive oil based on terahertz spectroscopy.The spectral characteristics of olive oils in time-domain and frequency range of 0-3.0059 THz band were analyzed.SVM,BPNN,PCA and GA are used respectively to produce the geographic origin identification.The least squares support vector machines(LS-SVM),back propagation neural network(BPNN)and random forests(RF)combined with principal component analysis(PCA)was used to identify the different origin of extra virgin olive oils,and compared with genetic algorithm(GA)to get the optimal discriminated model.According to the high-dimensional spectral characteristic data of terahertz spectrum,the dimension reduction effect of PCA and GA algorithms were compared,and a rapid identification model of olive oils geographic origin was established.The experimental results showed that the GA-LS-SVM model combined with THz absorption spectroscopy provided a new analytical method for the identification of olive oil origin and obtained the best classification results.The accuracies of the calibration and prediction set were as high as 100% and 96.25%,respectively.(2)A qualitative and quantitative analysis of adulterated olive oil was performed using terahertz spectroscopy and chemometric methods.The sunflower seed oil was added to the extra virgin olive oil produced in Italy according to the proportion of 0%,10%,20%,30%,40%,50%,60%,70%,80%,90%,and 100% to obtain samples with different adulteration ratios.The time-domain spectra in 22-30 ps,the frequency domain spectra in 0-3.0059 THz and the absorption spectra of the sample were analyzed.Using LS-SVM,BPNN and partial least squares(PLS)methods,the adulteration ratio was quantitatively analyzed in comparison with the GA algorithm which can select characteristic variables.The results showed that GA-LS-SVM model has the highest prediction accuracy for adulteration ratio with the determination coefficient(Rp2),root mean square error of prediction(RMSEP)and ratio of prediction to deviation(RPD)of 0.887,0.965% and 3.856,respectively.(3)Terahertz spectroscopy was used to identify the extra virgin olive oil with different degrees of oxidative rancidity.By setting the storage conditions to promote the oxidative rancidity to obtain the olive oil samples in different oxidation stages,samples are measured with the terahertz system in the range of 0.3051-2.0065 THz.The identification models of GA combined with SVM,BPNN and PLS were established.Among these models,the identification accuracy of calibration set and prediction set were 98.5% and 99% using GA-SVM model,which was the best prediction model.In addition,the correlation between peroxide values and terahertz bands was analyzed by interval partial least squares(i PLS).The results showed that there was a good linear relationship between terahertz spectrum and oxidation state in the range of 0.6867-0.9613 THz and which were more representative for the oxidation state of olive oil than other spectral ranges.This study showed that terahertz spectroscopy could realize the rapid detection of olive oil quality and meet the real-time monitoring requirement of olive oil quality during processing and storage in the industry.The results provide a theoretical support for the practical application of terahertz spectroscopy in olive oil industry,and also a reference for the application in other agricultural products inspection field.
Keywords/Search Tags:Extra virgin olive oil, Terahertz spectroscopy, Identification of geographic origin, Adulteration, Oxidative Rancidity, Rapid detection
PDF Full Text Request
Related items