Statistics in polymer reaction engineering

高分子反应工程统计

• 批准号: 105672-2006
• 负责人: Duever, Thomas
• 金额: $ 2.59万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2007
• 资助国家: 加拿大
• 起止时间: 2007-01-01 至 2008-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=362635
• 关键词: Statistics; polymer; reaction; engineering

In this research program, statistical methods are used to analyse data and develop mathematical models of chemical processes in particular polymerization processes used to manufacture various plastics and engineering polymers.  These models can be used to design reactors and ancilliary equipment for a process, optimize new and exisiting processes and control processes to produce material to specification and prevent unsafe operating conditions.  The statistical aspects of the model development address the fundamental and important question of how to bring process data into the mathematical model in an optimal fashion.  In this work, relatively recent developments in the field of applied statistics involving Markov Chain Monte Carlo methods will be used to solve the problems mentioned above.  This has the potential to lead to a general, flexible regression framework and software package.  This will significantly improve the ability of practicing engineers and scientists to correctly conduct regression analysis thereby efficiently extracting information from process models.

在该研究项目中，统计方法用于分析数据并开发化学过程的数学模型，特别是用于制造各种塑料和工程聚合物的聚合过程。  这些模型可用于设计工艺的反应器和辅助设备，优化新的和现有的工艺以及控制工艺，以生产符合规格的材料并防止不安全的操作条件。  模型开发的统计方面解决了如何以最佳方式将过程数据带入数学模型的基本且重要的问题。  在这项工作中，涉及马尔可夫链蒙特卡罗方法的应用统计领域的相对最新发展将用于解决上述问题。  这有可能产生通用、灵活的回归框架和软件包。  这将显着提高执业工程师和科学家正确进行回归分析的能力，从而有效地从过程模型中提取信息。