Research Initiation Award: A Novel Approach to the Control of Bioreactors

叙述：生物反应器的控制是复杂的， 事实上，微生物或组织细胞对 他们的成长环境发生了很小的变化。 此外还有 对生物系统过程动力学知之甚少 从而使设计和实施有效控制 战略是一项艰巨的任务，如果不是不可能的话。 另一个令人沮丧 问题是缺乏适当的传感器来监测重要的 过程变量，例如pH、温度、营养物浓度， 和副产物浓度。 为了帮助克服这些固有的问题，这项研究提出， 使用混合控制方案;将一个传统的 控制和第二个叠加的“专家系统”与事实， 以及直观的，不确定的，判断性的知识， 控制动作。 它建议，无论是基于模型和模型- 免费的专家系统进行调查。 以丁醇/丙酮发酵为模型体系， 由于文献中的强大数据库，该系统可以 以不同的产量和选择性生产各种产品 在不同的环境条件下，研究人员已经 以前有过这个系统的经验。 意图是 这项研究的结果将加强对生物过程的控制， 从而提高其可靠性和生产率。

Narrative: The control of biological reactors is complicated by the fact that microbial or tissue cells are extremely sensitive to small changes in their growth environment. In addition, there is little detailed knowledge about biological system process dynamics thus making the design and implementation of effective control strategies a difficult if not impossible task. Another frustrating problem is the lack of appropriate sensors to monitor important process variables, such as pH, temperature, nutrient concentration, and by-product concentration. To help overcome these inherent problems this research proposes to use a hybrid control scheme; incorporating one level of conventional control and a second superimposed "expert system" with factual, as well as intuitive, uncertain, and judgemental knowledge for extended control action. It is proposed that both a model-based and a model- free expert system be investigated. The butanol/acetone fermentation has been chosen as the model system because of a strong data base in the literature; the system can produce various products with variable yields and selectivities under different environmental conditions, and the investigators have had previous experience with this system. It is intended that results from this study will enhance the control of bioprocesses and therefore, increase their reliability and productivity.