Nonlinear Dynamical Systems Theory for Real-Time DistributedScheduling of Heterarchical Manufacturing Systems

[9612003] Duffie本研究涉及的是一种异构制造系统，在这种系统中，控制高度分布在松散耦合的自治实体之间，这些实体只保留了最少的全局信息。本研究的目标是建立一个非线性动力系统理论，用于零件驱动、批量低至统一、准时生产的异构制造系统的实时分布式控制建模与分析。需要解决的主要问题是：(1)通用多机配置的创新连续反馈控制拓扑，(2)描述通用多机配置行为的动态模型，以及(3)对系统动力学的严格分析，包括状态轨迹、稳定性、稳态值、收敛特性和鲁棒性。具有挑战性的研究问题包括建模非线性和分析系统动力学时，有一个以上的机器为一个给定的加工步骤，当零件必须访问多个这样的加工步骤。实时调度性能和对干扰的鲁棒性将在实验室中进行评估。本研究新颖，与行业需求高度相关。有许多重要的问题需要克服，使这个相对较高的风险，但如果成功，将代表着重大的进步。

9612003 Duffie This research involves heterarchical manufacturing systems in which control is highly distributed among loosely-coupled autonomous entities that retain a minimal amount of global information. The goal of this research is to establish a nonlinear dynamical systems theory for modeling and analysis of real-time distributed control in part-driven heterarchical manufacturing systems with batch sizes as low as unity and Just-In-Time production. Major issues to be addressed are (1) innovative continuous feedback control topologies for general multiple machine configurations, (2) dynamical models that characterize the behavior of general multiple machine configurations, and (3) rigorous analysis of system dynamics including state trajectories, stability, steady-state values, convergence properties, and robustness. Challenging research issues include modeling nonlinearities and analyzing system dynamics when there is more than one machine for a given processing step and when parts have to visit more than one such processing step. Real-time scheduling performance and robustness to disturbances will be evaluated in the laboratory. This research is novel and highly relevant to industry needs. There are many non-trivial issues to surmount making this relatively high risk, but if successful, will represent major advances.