Model Based Control of Dissolved Oxygen at Low Concentrations and in Shear Sensitive Fermentations

基于模型的低浓度和剪切敏感发酵中溶解氧的控制

• 批准号: 9622536
• 负责人: Muhammad Karim
• 金额: $ 29.41万
• 依托单位: Colorado State University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-08-15 至 2000-10-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9622536&HistoricalAwards=false
• 关键词: Model; Based; Control; Dissolved; Oxygen

9622536 Karim This proposal is on research to develop effective means of on-line monitoring, supervision and control of dissolved oxygen (DO) for a variety of fermentation conditions. Most fermentation systems are complex, time-varying, and highly nonlinear in nature, especially in a batch culture system, and therefore control of the important variables can be difficult. It is proposed to use a nonlinear model predictive control approach by using a recursive nonlinear autoregressive model with an exogenous input (NARX) to approach this problem. Specific objectives of this research are to: (1) develop a reliable experimental setup to control DO, (2) develop on-line identification with reliable models, i.e. NARX and adaptive NARX (ANARX), (3) use the identified model to implement model predictive control, and (4) test the NARX and ANARX based DO controllers in different fermentations, i.e. both low and high density bacterial cell cultures, including fermentations in the presence of two viscous aqueous phases, and plant cell cultivations which require low shear stress environments. ***

9622536 Karim该提案是关于研究开发用于各种发酵条件的溶解氧（DO）的在线监测、监督和控制的有效手段。 大多数发酵系统本质上是复杂的、时变的和高度非线性的，特别是在分批培养系统中，因此重要变量的控制可能是困难的。 提出了一种非线性模型预测控制方法，该方法利用带有外生输入的递归非线性自回归模型（NARX）来解决这个问题。 这项研究的具体目标是：（1）开发可靠的实验装置来控制DO，（2）开发具有可靠模型的在线识别，即NARX和自适应NARX（ANARX），（3）使用识别的模型来实施模型预测控制，以及（4）在不同发酵中测试基于NARX和ANARX的DO控制器，即低密度和高密度细菌细胞培养，包括在两种粘性水相存在下的发酵，以及需要低剪切应力环境的植物细胞培养。 ***