CAREER: Research and Educational Advances in Optimal Groundwater Remediation Design

Minsker 9734076本研究的目标是：1)开发解决现场规模地下水修复设计问题的改进方法；2)使从业者和学生更容易获得最优地下水修复设计工具；3)使用主动学习技术来改进各级学生的教育；以及4)教研究生如何开发和应用耦合的优化和模拟模型。为了提高前期工作中建立的好氧原位生物修复设计优化控制模型的计算效率，将开发多尺度优化方法。这将使对异质、实地规模的站点进行准确建模成为可能。将开发一种混合遗传算法，以解决局部最优的困难，并允许以有效的方式考虑固定成本。一旦确定了计算效率最高的多尺度方法和混合遗传算法，就将建立一个将多尺度方法整合到混合遗传算法中的野外尺度原位生物修复设计的集成模型。本科生研究助理将编程图形用户界面，以方便来自行业的学生和从业者使用在本研究中开发的设计模型。然后，这些模型将被用作教育游戏的核心，教会学生如何使用优化设计工具来设计具有成本效益的补救解决方案。这个游戏将在本科生和研究生课程中进行，并在推广项目中使用小型学生团队之间的竞赛形式。还将开设一门关于耦合优化和仿真建模的新研究生课程，向学生传授与开发和应用此类模型相关的复杂性。本课程将作为展示研究成果和开发改进方法的论坛。***

Minsker 9734076 The objectives of this research are to: 1) develop improved methods for solving field-scale groundwater remediation design problems; 2) improve access to optimal groundwater remediation design tools by practitioners and students; 3) use active learning techniques to improve education of students at all levels; and 4) teach graduate students how to develop and apply coupled optimization and simulation models. Multiscale optimization methods will be developed to improve the computational efficiency of an optimal control model for aerobic in situ bioremediation design developed in previous work. This will enable accurate modeling of heterogeneous, field-scale sites. A hybrid genetic algorithm will be developed to address difficulties with local optima and to allow fixed costs to be considered in an efficient manner. Once the multiscale method and hybrid genetic algorithm with the best computational efficiency have been identified, an integrated model for field-scale in situ bioremediation design will be developed that will incorporate multiscale methods into a hybrid genetic algorithm. Undergraduate research assistants will program graphical user interfaces to facilitate use of the design models developed in this research by students and practitioners from industry. The models will then be used as the core for an educational game to teach students the utility of optimal design tools for designing cost-effective remediation solutions. The game will be played in undergraduate and graduate courses and in outreach programs using a contest format between small student teams. A new graduate course on coupled optimization and simulation modeling will also be created to teach students the complexities associated with developing and applying such models. The course will serve as a forum for presenting research results and developing improved methods. ***