CIESC Journal

• 化工学报 • 上一篇    下一篇

神经网络建模方法在维生素C发酵过程中的应用

花强,王树青   

  1. 浙江大学工业控制技术研究所!杭州310027,浙江大学工业控制技术研究所!杭州310027
  • 出版日期:1996-08-25 发布日期:1996-08-25

APPLICATION OF NEURAL NETWORK MODEL IN VITAMIN C FERMEMTATION PROCESS

Hua Qiang;Wang Shuqing(Institute of Industrial Process Control,Zhejiang Universty,Hangzhou 310027)   

  • Online:1996-08-25 Published:1996-08-25

摘要: 传统的标准“黑箱”型人工神经网络已较为广泛地应用于生化过程中的状态预估等多个方面,然而结合过程先验知识或部分机理模型的混合神经网络建模方法能给出更令人满意的结果.本文将其应用于2-酮基-l-古龙酸(2-KLG)发酵过程的状态估计,并将其结果与传统神经网络模型进行了比较,混合模型明显优于单一神经网络方法.

Abstract: Conventional standard artificial neural network has been applied widely in state estimation of bioprocesses,but hybrid neural network,which combines the prior knowledge or partially known principles of the process,can give more satisfactory results.The hybrid model is used in state estimation of 2-keto-l-gulonic acid fermentation in this paper,and the results are compared with that of standard neural network.

中图分类号: