Journal of South China University of Technology(Natural Science Edition) ›› 2004, Vol. 32 ›› Issue (12): 11-15.

Previous Articles     Next Articles

BP Neural Network-based Software Measurement for DM423 Biomass During Flask Cultivation

Li Bing Guo S-i yuan Li Lin Li X-i liu   

  1. College of Food and Biological Engineering‚South China Univ.of Tech.‚Guangzhou 510640‚Guangdong‚China
  • Received:2004-08-25 Online:2004-12-20 Published:2015-09-08
  • Contact: 李冰(1972-)‚女‚讲师‚博士‚主要从事食品与生物化工研究。 E-mail:lcbingli@scut.edu.cn
  • About author:李冰(1972-)‚女‚讲师‚博士‚主要从事食品与生物化工研究。

Abstract: The relationship between the initial conditions (initial glucose content and initial biomass) and the biomass during the flask cultivation of Baeillus cereus DM423was described by BP neural network.A2-6-3topo-logy of the BP neural network was then constructed.The estimation results of biomass by the BP neural network show that there is little difference between the calculated values and the experimental ones‚which means that the constructed neural network has a good generalization.The proposed network also promises rather good prediction results when the initial glucose contents are4.25and27.30g/L‚and the average relative error of the estimation is1.95%.

Key words: BP neural network, flask cultivation, biomass, estimation, generalization

CLC Number: