Process Modeling and Optimization of Biotechnology Systems Using Parameter Function Neural Networks

使用参数函数神经网络的生物技术系统的过程建模和优化

• 批准号: 0227748
• 负责人: W. Fred Ramirez
• 金额: $ 10万
• 依托单位: University of Colorado at Boulder
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-09-01 至 2005-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0227748&HistoricalAwards=false
• 关键词: Process; Modeling; Optimization; Biotechnology; Systems

This project aims to develop a new framework for using neural networks and fundamental process knowledge to model, optimize and control dynamic biotechnology processes. This research group has recently proposed parameter function neural networks as an effective means of using both fundamental process knowledge and neural networks to model dynamic processes. They use the parameter function concept to model the industrially important process of the production of hemoglobin though induced recombinant bacteria using experimental data available from Baxter Hemoglobin Therapeutics. In general, this project investigates the feasibility of using the parameter function neural network modeling technique to successfully model a complex industrial biotechnology fermentation and find new improved operational strategies for that process. The anticipated results of this project may have widespread impact on modeling and optimization of industrial processes, especially the biotechnology processes of fed-batch bioreactors.

该项目旨在开发一个新框架，利用神经网络和基本过程知识来建模、优化和控制动态生物技术过程。该研究小组最近提出了参数函数神经网络作为使用基本过程知识和神经网络来建模动态过程的有效手段。他们利用百特血红蛋白治疗公司提供的实验数据，利用参数函数概念对通过诱导重组细菌生产血红蛋白的工业重要过程进行建模。总的来说，该项目研究了使用参数函数神经网络建模技术成功模拟复杂的工业生物技术发酵的可行性，并为该过程找到新的改进操作策略。该项目的预期结果可能会对工业过程的建模和优化产生广泛的影响，特别是补料分批生物反应器的生物技术过程。