Globally Optimal Process Network Synthesis

CBET-0829211人工智能优点本项目的目标是利用PI开发的无限维状态空间(IDEAS)方法，提高过程网络全局优化综合的最新水平。他计划进行以下研究：1.IDEAS的概念和计算进展2.间歇反应堆网络综合-可达区域的识别3.全局最优燃料电池网络综合4.全局最优低温液化网络的自动生成在开展这项研究时，他将：a.为IDEAS无限线性规划(ILP)寻找新的求解方法，该方法产生的下界保证收敛到全局最小值。寻求间歇过程网络的新概念公式。这是第一次将思想应用于动态过程网络，如间歇反应堆。还将讨论与网络因果关系相关的问题。C.要么开发新的燃料电池模型，要么使用现有的燃料电池模型，以确定这些模型是否满足纳入IDEAS框架的属性，包括必须对模型属性做出的任何假设。将思想应用于低温液化网络将产生新的低温循环。将使用包括用户友好的图形用户界面的软件包进行这种周期的合成。该软件包将被纳入学生的课堂作业，以及解释该方法的威力及其局限性的文档。更广泛的影响已经证明，IDEAS方法促进了过程网络综合中的技术水平，并为设计者提供了一种不预先限制网络结构的综合方法。该方法不仅适用于电气工程(燃料电池、混合电力系统等)、运筹学(无限维线性规划的数学发展、大规模线性规划的并行求解等)和低温工程，而且同样适用于化学工程过程设计。研究活动将为K-12、本科生和研究生提供培训，让他们接触高级数学和集群计算技术。这项工作还将通过促进更多K-12学生的研究项目来促进多样性，如加州大学洛杉矶分校工程与多样性卓越中心(CEED)提供的项目。在过去的这个夏天(2007)，PI让一名少数族裔高中生通过加州大学洛杉矶分校CEED运营的学生科学、数学成就和研究技术(SMARTS)项目，使用IDEAS方法做一个暑期研究项目。PI计划让更多的各级学生参与进来，以培养他们对工程和高等教育的兴趣。思想的概念进步可能会影响几个研究领域(数学编程、低温、反应工程、燃料电池系统工程)和几个行业(电力生产、氢气生产、制冷等)。

CBET-0829211ManousiouthakisIntellectual Merit The goal of this project is to advance the state of the art in globally optimal synthesis of process networks using the Infinite DimEnsionAl State-space (IDEAS) method developed by the PI. He plans the following research: 1. Conceptual and Computational Advancement of IDEAS 2. Batch Reactor Network Synthesis - Identification of the Attainable Region 3. Globally Optimal Fuel Cell Network Synthesis 4. Automatic Generation of Globally Optimal Cryogenic Liquefaction Networks In carrying out this research, he will: A. Seek new solution methods for the IDEAS Infinite Linear Program (ILP) that yield lower bounds that are guaranteed to converge to the global minimum. B. Seek a novel IDEAS formulation for batch process networks. This is the first time that application of IDEAS will be pursued for a network of dynamic processes such as batch reactors. Issues related to network causality will also be addressed. C. Either develop a new or use existing fuel cell models to establish whether these models satisfy the properties for incorporation into the IDEAS framework, including any assumptions that must be made on the model properties to make this the case. The application of IDEAS to cryogenic liquefaction networks will yield novel cryogenic cycles. Synthesis of such cycles will be pursued using software packages which include a user-friendly GUI. This software package will be incorporated into classroom work for students, along with documentation explaining the power of the method and its limitations. Broader Impact The IDEAS methodology has been shown to advance the state of the art in process network synthesis and to provide a designer with a synthesis methodology that does not restrict a priori the network structure. The method applies just as well to chemical engineering process design as it does to electrical engineering (fuel cells, hybrid electrical systems, etc.), operations research (mathematical developments in infinite dimensional linear programming, parallelized solution of large scale linear programs, etc.), and cryogenic engineering. The research activity will provide training for K-12, undergraduate, and graduate students by exposing them to high level mathematics and cluster computing technology. This work will also contribute to diversity by facilitating more research projects for K-12 students like those provided by the UCLA Center for Excellence in Engineering and Diversity (CEED). This past summer (2007) the PI had a minority high school student do a summer research project using the IDEAS method through the Science Mathematics Achievement and Research Technology for Students (SMARTS) program run by UCLA CEED. The PI plans on involving more students of all levels in order to foster their interest in engineering and higher education. The conceptual advancement of IDEAS could impact several research areas (mathematical programming, cryogenics, reaction engineering, fuel cell systems engineering) and several industries (power production, hydrogen production, refrigeration, and others).