Distributed optimization of large-scale chemical / petrochemical plant operations.

化工/石化行业经营的加工厂规模非常大，高度集成和复杂的系统。这类工厂的性能目标是最大限度地提高运营效率，同时确保对环境的影响最小。当前的过程控制和优化方法不能满足日益高效的操作的要求。我们正接近一个技术转折点，当前的过程自动化范例将不能满足社会强加的性能要求。因此，过程系统工程的一个关键前沿是创造技术和工具，这些技术和工具将为建立下一代过程自动化系统提供基础。更准确地说，对于化工/石化行业中遇到的超大规模系统(如时变、紧密集成、非线性等)，需要新的方法来进行过程控制和优化。我们研究计划的长期目标一直是，并将继续是创造优化工艺操作的方法和工具。在拟议的研究计划中，我们将重点研究大规模过程系统中的分布式性能优化方法。现有工作的基础假设是部件控制/优化子系统是完全同步的，这限制了现有研究成果的适用性。异步性是“真实世界”过程控制和优化应用程序的规则。拟议的研究计划的重点将是为基于协调的分布式优化系统创建一个理论框架和过程系统工程技术，以应对化工/石化厂通常面临的异步性。拟议方案产生的预期贡献代表着实现分布式、基于协调的过程控制和优化系统的潜力的重要一步。

The chemical / petrochemical industries operate processing plants that are very large-scale, highly integrated and complex systems. The performance objectives for such plants are to maximize operating efficiency, while ensuring a minimal environmental impact. Current process control and optimization approaches cannot meet the demands for increasingly efficient operations. We are nearing a technological breakpoint, where the current process automation paradigm will not be able to meet the performance demands imposed by society. Thus a key frontier in process systems engineering is the creation of techniques and tools that will provide the basis for the creation of the next generation of process automation systems. More precisely, new approaches are required for process control and optimization for the very large-scale systems that are encountered in the chemical / petrochemical industries (e.g., time-varying, tightly integrated, nonlinear and so forth). The long-term objective of our research program has been, and continues to be, the creation of methods and tools for optimizing process operations. In the proposed research program, we will focus on distributed approaches to performance optimization in large-scale process systems. The underpinning assumption in existing work is that the component control / optimization subsystems are perfectly synchronized, which limits the applicability of available research results. Asynchronicity is the rule for"real-world" process control and optimization applications. The focus of the proposed research program will be in the creation of a theoretical framework and process systems engineering techniques for coordination-based, distributed optimization systems in the face of the asynchronicity that is typically faced in chemical / petrochemical plants. The expected contributions arising from the proposed program represent an essential step in realizing the potential for distributed, coordination-based process control and optimization systems.