(105d) Resource-Aware Control of Spatially Distributed Processes: Integrating Control, Computation and Communication

The integration of wireless sensor networks in process control systems is an appealing goal that promises to expand the capabilities of existing control technology beyond what is possible with wired devices alone, and is a key mechanism for enabling the emerging paradigm of smart plant operations. An important challenge with this evolution is the development of resource-aware control methods that can systematically balance the desired stability and performance requirements against the intrinsic constraints on the sensing, computation and communication resources of the networked devices as well as the occasional unreliability of the communication medium due to interference in the field or environmental impact. Despite the significant and growing body of research work on networked control, the majority of existing methods have targeted lumped parameter systems described by ordinary differential or difference equations.  Many important engineering applications, however, are characterized by spatial variations and are naturally modeled by partial differential equations (PDEs). Typical examples include transport-reaction processes and fluid flow systems where the control problem often involves the regulation of spatially distributed variables using spatially-distributed control actuators and measurement sensors. Compared with the extensive literature on control of distributed parameter systems in process control, results on networked control of spatially distributed processes remain limited at present. Major bottlenecks include the infinite-dimensional nature of such processes, as well as the complex dynamics and uncertainties that characterize their models. Over the past few years, research work in our group has begun to address the development of networked control and scheduling methods for spatially distributed processes that arise commonly in the modeling of transport-reaction processes and fluid flows. This presentation will provide an overview of our recent results in this area and outline some of the open challenges that still need to be addressed.