Hybrid neural network based on software measurement for DM423 biomass during batch cultivation

Authors

ZENG Xiangyan, Department of Biology and Chemical Engineering, Shaoyang University, Shaoyang, Hunan 422000, China; Soybean Processing Techniques of the Application and Basic Research Base in Hunan Province, Shaoyang, Hunan 422000, China
ZHAO Liangzhong, Department of Biology and Chemical Engineering, Shaoyang University, Shaoyang, Hunan 422000, China; Soybean Processing Techniques of the Application and Basic Research Base in Hunan Province, Shaoyang, Hunan 422000, China
LI Bing, Institute of Light Industry & Chemical Engineering, South China University of Technology, Guangzhou, Guangdong 510640, China
LI Lin, Institute of Light Industry & Chemical Engineering, South China University of Technology, Guangzhou, Guangdong 510640, China

Abstract

The biomass concentration estimated by the static feedforward multiplayer neural network of 5-5-1 topology and hybrid neural network - microbe growth model with five inputs of culture time, temperature, pH and dissolved oxygen and glucose concentration. The result showed that the static feedforward multiplayer neural network mean squared error (MSE) of testing samples 1.73×10^-3. And hybrid neural network - microbe growth model offered a much better generalization accuracy than that of single neural network model, with MSE of testing samples of 0.25×10^-3, it was found that there was some deviation between estimated biomass and actual values while microbe were growing in the stationary phase.

Publication Date

5-28-2016

First Page

30

Last Page

33

DOI

10.13652/j.issn.1003-5788.2016.05.007

Recommended Citation

Xiangyan, ZENG; Liangzhong, ZHAO; Bing, LI; and Lin, LI (2016) "Hybrid neural network based on software measurement for DM423 biomass during batch cultivation," Food and Machinery: Vol. 32: Iss. 5, Article 7.
DOI: 10.13652/j.issn.1003-5788.2016.05.007
Available at: https://www.ifoodmm.cn/journal/vol32/iss5/7

References

[1] 曾祥燕, 赵良忠, 蒋盛岩. BP 神经网络优化槐花中芦丁的提取工艺[J]. 天然产物研究与开发, 2013, 3(25): 312-316.
[2] 许哲军, 何云, 郭美锦, 等. 重组毕赤酵母表达期菌体浓度的软测量模型[J]. 食品与生物技术学报, 2006, 25(5): 28-34.
[3] 曾祥燕, 赵良忠, 孙文兵, 等. 基于PCA和BP神经网络的葡萄酒品质预测模型[J]. 食品与机械, 2014, 30(1): 40-44.
[4] 李冰, 郭祀远, 李琳, 等. PRNN对Bacillus cereusDM423分批培养过程中生物量的软测量[J]. 华南理工大学学报: 自然科学版, 2009, 37(4): 111-115.
[5] 黎移新. 基于前馈神经网络的脐橙糖度及有效酸度检测[J]. 食品与机械, 2009, 25(1): 78-81.
[6] Wilson J A, Zorzetto L F M. A generalized approach to process state estimation using hybrid artificial neural network/mechanistic models[J]. Computers Chem. Engng., 1997, 21(9): 951-963.
[7] Psichogios C D, Ungar L H. A hybrid neural network-first principles approach to process modeling[J]. AIChE Journal, 1992, 38(10): 1 499-1 511.
[8] 朱海燕, 朱晓莲, 黄頔. 基于动态BP神经网络的预测方法及其应用[J]. 计算机与信息技术, 2007(Z1): 3-6.
[9] 黎移新. 多层前馈神经网络几种算法的样本顺序敏感性[J]. 食品与机械, 2010, 26(6): 100-103.
[10] 隋青美, 于正欧. 发酵过程混合神经网络建模方法比较[J]. 山东工业大学学报, 2001, 26(4): 95-97.
[11] Pan Li-jun, Zhang Chan-chan, Jiang Shao-tong, et al. Medium optimization of L-lactic acid fermentation by lactic acid bacteria with rape cake hydrolysate as nitrogen source[J]. Food Sci., 2007, 10(6): 313-316.
[12] 曹乐平, 谢培甫. Vfb基于补偿模糊神经网络的立式转鼓水果分级系统设计[J]. 食品与机械, 2007, 23(2): 70-73.
[13] 袁丽红. 微生物学实验[M]. 北京: 化学工业出版社, 2010: 67-69.
[14] 宁正祥. 食品分析[M]. 北京: 中国轻工业出版社, 1994: 90-100.
[15] 王树青. 生化反应过程模型化及计算机控制[M]. 杭州: 浙江大学出版社, 1998.
[16] Leal R R, Butler P, Lane P, et al. Data fusion and artificial neural networks for biomass estimation[J]. IEE Proc-Sci. Meas. Technol., 1997, 144(2): 69-72.
[17] 戚以政, 汪叔雄. 生化反应动力学与反应器[M]. 北京: 化学工业出版社, 1998: 57-63.
[18] Ratkov A L, Georgiev T Z, Ivanova V, et al. Investigation of L-lysine fed-batch fermentation process with droppings by mathematical modeling[J]. IEEE 26th Int. Conf. Information Technology Interfaces, 2004(6): 543-547.