Coordination-based optimization framework for engineering systems design considering both individual and cooperative performance objectives

当今的工程问题非常复杂，无法用单片系统的方法来解决。不同的系统必须合作应对社会技术挑战。例如，不同的能源生产、转换和储存系统必须合作以满足需求，同时促进可持续性和尽量减少对环境的不利影响。到目前为止，工程设计文献中还没有定量的方法来对这种异构系统的配置进行系统优化设计。因此，本研究计划的目标是开发必要的理论和计算工具，用于基于仿真的分析、设计和优化工程系统，这些系统的构建和配置不仅可以实现个人目标，还可以实现集体目标，以便它们可以提供两个系统都无法单独提供的功能。建议研究的第一个5年阶段将侧重于解决以下问题：(1)捕获、理解和解释互联系统之间的意外或不希望的相互作用；(2)处理合作系统配置的时间依赖性质并评估其性能和效率；建模和量化与互联系统网络的不同组成和集体设计目标相关的不确定性；(4)集成涉及多学科和不同空间/时间尺度的异构系统。我们的方法将基于数学规划、协调方法、基于图的网络理论和贝叶斯统计。这项研究的成功实施将为未来发展大规模系统网络的控制和生命周期管理方法奠定必要的基础，以确保强大和有效的互操作性。从教育的角度来看，研究方法将把跨学科的协作系统思维注入工程学生和实践者，他们将学习以创造性和创新的方式制定和解决复杂的工程问题。

Today's engineering problems are so complex that they cannot be solved by means of monolithic system approaches. Different systems must cooperate to meet socio-technical challenges. For example, different systems of energy generation, conversion and storage must cooperate to satisfy demands while fostering sustainability and minimizing adverse environmental impact. To date, no quantitative methodology exists in the engineering design literature for systematic optimal design of such configurations of heterogeneous systems. Therefore, the goal of this research program is to develop the necessary theoretical and computational tools for simulation-based analysis, design and optimization of engineering systems that are built and configured to accomplish not only individual but also collective objectives so that they can provide functionalities that neither of the systems is capable of providing by itself. The first 5-year stage of the proposed research will focus on addressing the issues of (1) capturing, understanding and accounting for unexpected or undesired interactions among the interconnected systems, (2) dealing with the time-dependent nature of configurations of cooperating systems and assessing their performance and efficiency, (3) identifying, modeling and quantifying uncertainties related to the varying composition and collective design objectives of interconnected system networks and (4) integrating heterogeneous systems involving multiple disciplines and different space/time scales. Our approach will be based on mathematical programming, coordination methods, graph-based network theory and Bayesian statistics. The successful implementation of this research will result in a foundation necessary for future development of methods for control and life-cycle management of very large-scale system networks to ensure robust and efficient interoperability. From an educational perspective, the research methodology will infuse interdisciplinary collaborative systems thinking into engineering students and practitioners who will learn to formulate and solve complex engineering problems in creative and innovative ways.