Advanced Multiscale Decision-making for Process Operations and Energy Systems, and Incorporating Big Data Analytics

在能源和流程行业的所有部门，都在大力开发和运营敏捷制造技术，以响应客户需求和动荡的市场，同时控制成本、提高效率和减少污染。这就产生了对解决复杂的综合规划和调度问题的系统的需求，这些问题弥合了不同职能和战略决策层之间的差距。这些具有挑战性的多尺度决策问题需要更高效的计算工具和求解策略。这项研究计划的目标是开发新技术，旨在及时获得更具竞争力的集成解决方案。尽管在处理上述问题的一些组合方面取得了进展(包括首席研究员过去的研究工作)，但仍需要作出重大研究努力来克服这些多尺度的挑战。本研究计划旨在设计整合了基于逻辑的方法、分解技术、随机优化和重构策略的改进的求解算法。由于现有的大多数方法都是特定于问题或仅适用于短时间视界的，因此聚类将被作为一种潜在的工具来研究，以有效地缩小模型规模并实现计算可处理性，同时保持解的准确性。该研究计划将考虑的其他方法是使用连续时间尺度、滚动范围战略和使用大数据分析的代理模型。*将考虑两个广泛的应用领域：一个专注于流程运营和企业范围的集成设施，另一个涉及先进的能源生产系统。将通过几种电网稳定办法解决可再生能源间歇性的挑战，包括能源储存、流程工业电气化以及通过Power-to-X技术将过剩电力转换为有用的产品。开发的计算工具的应用包括做出设计或运营决策、选择系统组件、单元互连，甚至供应链或生态系统中的相互作用。该计划将为几名研究生和本科生提供培训，让他们接触到大型多尺度决策问题的实际解决方案中的先进技术，并有可能对未来的流程和能源行业产生重大影响，为加拿大带来巨大的经济利益。参与研究计划的学生将非常适合目前工业界和学术界过程系统工程领域的许多职位。

Across all sectors within the energy and process industry, tremendous efforts have been devoted towards the development and operation of agile manufacturing techniques to respond to customer needs and volatile markets while at the same time control costs, improve efficiency, and reduce pollution. This has created a demand for systems to solve complex integrated planning and scheduling problems that bridge the gap between the different functional and strategic decision-making levels. These challenging multiscale decision problems require more efficient computational tools and solution strategies. The goal of this research program is to develop new techniques aimed to obtain more competitive integrated solutions in a timely manner. Although progress (including past research efforts by the principal investigator) has been made in dealing with some of the combinatorial aspects of the problems cited above, significant research efforts are still required to overcome these multiscale challenges. This research program aims at devising improved solution algorithms that integrate logic-based methods, decomposition techniques, stochastic optimization, and reformulation strategies. Since most of the existing methodologies are problem-specific or applicable only to short time horizons, clustering will be investigated as a potential tool to effectively shrink model size and achieve computational tractability while at the same time maintaining solution accuracy. Other approaches that this research program will consider is the use of continuous time scales, rolling horizon strategies, and surrogate models that employ big data analytics.******Two broad application domains will be considered: one focuses on process operations and enterprise-wide integrated facilities and the other deals with advanced energy production systems. The challenges of intermittence of renewable energy will be addressed through several grid stabilization approaches, including energy storage, electrification of the process industry, and conversion of surplus electricity to useful products through Power-to-X technologies. Applications of the developed computational tools include making design or operating decisions, selection of system components, interconnections of units, or even interactions along supply chains or the ecosystem. The program will provide training for several graduate and undergraduate students, exposing them to advanced techniques in the practical solution of large multiscale decision problems, and has the potential to significantly impact tomorrow's process and energy industry with tremendous economic benefits to Canada. Students involved in the research program will be well-suited for the many positions that are currently available in the area of process systems engineering in both industry and academia.*****