SGER: Model Uncertainty and Robustness in Nonlinear Model Predictive Control for Biomedical Applications

SGER：生物医学应用非线性模型预测控制中的模型不确定性和鲁棒性

• 批准号: 0328247
• 负责人: Victor Vasquez
• 金额: $ 9.66万
• 依托单位: Board of Regents, NSHE, obo University of Nevada, Reno
• 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: 2003-07-01 至 2005-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0328247&HistoricalAwards=false
• 关键词: SGER; Model; Uncertainty; Robustness; Nonlinear

Intellectual Merit:An issue in the development of any process control strategy is the sensitivity and robustness of the process models and the application of viable computational procedures where the input is uncertain information or about processes or specific events. This SGER project is exploratory research aimed at exploring sensitivity and robustness issues in process control with biomedical applications. In addition, a better understanding of the effects of uncertainties and the development of more robust design and analysis procedures could also benefit industry through enhanced product effectiveness and better-trained engineers. New uncertainty-analysis methodologies might also be used by regulatory agencies in the approval of new procedures and technologies in medicine. The main goal of the project is to explore the robustness due to uncertainties in predictive control algorithms, ultimately for biomedical applications - specifically exploring the robustness of nonlinear model predictive control (MPC) in the presence of model uncertainty. Some case studies, such as insulin delivery systems, will ultimately be analyzed.Broad Impact:The research has the potential to influence the way medication is administered to patients. Drugs might be administered to match the patient's needs, rather than according to a prescribed schedule. Applications run the gamut from controlling insulin dosages to the administering of various drugs for cancer treatment.

智力优势：任何过程控制策略开发中的一个问题是过程模型的灵敏度和鲁棒性，以及输入是不确定信息或关于过程或特定事件的可行计算程序的应用。 该SGER项目是探索性研究，旨在探索生物医学应用过程控制中的灵敏度和鲁棒性问题。 此外，更好地了解不确定性的影响和制定更健全的设计和分析程序，也可以通过提高产品效率和培训更好的工程师使工业受益。 新的不确定性分析方法也可能被监管机构用于批准新的医疗程序和技术。 该项目的主要目标是探索由于预测控制算法中的不确定性而导致的鲁棒性，最终用于生物医学应用-特别是探索存在模型不确定性的非线性模型预测控制（MPC）的鲁棒性。 一些案例研究，如胰岛素输送系统，将最终进行分析。广泛的影响：该研究有可能影响药物给病人的方式。 药物可以根据病人的需要给药，而不是按照规定的时间表给药。 应用范围从控制胰岛素剂量到各种癌症治疗药物的给药。