Combined LF-NMR and Artificial Intelligence for Continuous Real-Time Monitoring of Carrot in Microwave Vacuum Drying
- 70 Downloads
In this paper, intelligent technology of combined low field NMR (LF-NMR) and back propagation artificial neural network (BP-ANN) was used to monitor moisture content in carrot during microwave vacuum drying. The relationship between different drying powers (200, 300, and 400 W) and NMR signals (A21, A22, A23, and Atotal) was investigated. Results show that as the drying time elapsed, the NMR signals of Atotal and A23 decrease all drying conditions, A21 and A22 tend to increase at high moisture content and then decrease, which is consistent with the state of water while changes during drying. NMR signals can be used as indicators for online monitoring of moisture and control of the drying process. With NMR signals as input variables, a BP-ANN model was built optimized by transfer function, training function, and the number of neurons to model the moisture content (output). Compared with a linear regression model and multiple linear regression model, the BP-ANN model with the topology of 4-25-1, transfer function of tansig and purelin, and training function of trainlm outperformed the fitting performance and accuracy. This shows that the combined approach of utilizing LF-NMR and BP-ANN has great potential in intelligent online monitoring and control applications for carrot drying.
KeywordsArtificial intelligence Artificial neural network LF-NMR Carrot cube Microwave vacuum drying
We acknowledge the financial support from National Key R&D Program of China (Contract No. 2017YFD0400901), Jiangsu Province (China) Agricultural Innovation Project (Contract No. CX(17)2017), Jiangsu Province Key Laboratory Project of Advanced Food Manufacturing Equipment and Technology (No. FMZ201803), and Jiangsu Province (China) “Collaborative Innovation Center for Food Safety and Quality Control” Industry Development Program, National First-class Discipline Program of Food Science and Technology (No. JUFSTR20180205), all of which enabled us to carry out this study.
Compliance with Ethical Standards
Conflict of Interest
The authors declare that they have no conflicts of interest.
- China. (2016). GB5009.3-2016 Determination of moisture in food. Beijing: National Health and Family Planning Commission of the People’s Republic of China.Google Scholar
- Duan, X. M., Feng, X. Q., Song, L., Zhang, B., Cai, X. T., Li, M. M., Yang, F. W., & Fan, L. L. (2013). Advances on development of fruit and vegetable drying by MVD technology. Food and Fermentiin Industries., 39(9), 156–164.Google Scholar
- Hu, X. Y., Lan, W. Q., Zhang, N. N., & Xie, J. (2017). Research progress of low-field nuclear magnetic resonance technology in food. Science and Technology of Food Industry, 38(6), 386–396.Google Scholar
- Li, L. L., Zhang, M., Bhandari, B., & Zhou, L. Q. (2018). LF-NMR online detection of water dynamics in apple cubes during microwave vacuum drying. Drying Technology, 36(16), 2006–2015.Google Scholar
- Momenzadeh, L., Zomorodian, A., & Mowla, D. (2012). Applying artificial neural network for drying time prediction of green pea in a microwave assisted fluidized bed dryer. Journal of Agricultural Science and Technology, 14, 513–522.Google Scholar
- Pariyani, R., Ismail, I. S., Ahmad Azam, A., Abas, F., & Shaari, K. (2017). Identification of the compositional changes in Orthosiphon stamineus leaves triggered by different drying techniques using (1) H NMR metabolomics. Journal of the Science of Food and Agriculture, 97(12), 4169–4179.CrossRefGoogle Scholar
- Sun, Q., Zhang, M., & Mujumdar, A. S. (2018). Recent developments of artificial intelligence in drying of freshfood: a review. Critical Reviews in Food Science and Nutrition. https://doi.org/10.1080/10408398.2018.1446900.
- Ting, X. (2014). Nondestructive detection of fruit quality based on low-field magnetic resonance technology. Hangzhou: D, China Jiliang University.Google Scholar
- Tylewicz, U., Aganovic, K., Vannini, M., Toepfl, S., Bortolotti, V., Dalla Rosa, M., Oey, I., & Heinz, V. (2016). Effect of pulsed electric field treatment on water distribution of freeze-dried apple tissue evaluated with DSC and TD-NMR techniques. Innovative Food Science and Emerging Technologies, 37, 352–358.CrossRefGoogle Scholar
- Yaghoubi, M., Askari, B., Mokhtarian, M., Tavakolipour, H., Elhamirad, A. H., Jafarpour, A., & HeidarianS. (2013). Possibility of using neural networks for moisture ratio prediction in dried potatoes by means of different drying methods and evaluating physicochemical properties. Agricultural Engineering International: CIGR Journal, 15(4), 258–269.Google Scholar
- Yan, K. J., Chu, Y., Huang, J. H., Jiang, M. M., Li, W., Wang, Y. F., Huang, H. Y., Qin, Y. H., Ma, X. H., Zhou, S. P., Sun, H., & Wang, W. (2016). Qualitative and quantitative analyses of Compound Danshen extract based on (1)H NMR method and its application for quality control. Journal of Pharmaceutical and Biomedical Analysis, 131, 183–187.CrossRefGoogle Scholar
- Zhou, K. l., & Kang, Y. H. (2005). Neural network model and MTLAB simulation program design. Beijing: Peking University Press.Google Scholar
- Zou, H. Q., Li, S., Huang, Y. H., Liu, Y., Bauer, R., Peng, L., Tao, O., Yan, S. R., & Yan, Y. H. (2014). Rapid identification of Asteraceae plants with improved RBF-ANN classification models based on MOS sensor E-nose. Evidence-based Complementary and Alternative Medicine, 2014, 425341.PubMedPubMedCentralGoogle Scholar