Plantwide process control with asynchronous sampling and communications

Abstract The need for economic efficiency has driven the development of large and complex chemical plants. These plants typically contain a large number of process units, often with different time scales, interacting with each other due to the use of recycle streams and heat integration. This causes considerable difficulties in plantwide control. This paper aims to address this issue by developing a flexible networked-based plantwide control approach. In this approach, a plantwide process is modeled as a network of process units which is controlled by a network of autonomous controllers. The controllers within the network operate with different sampling rates and communicate with each other asynchronously, to allow cost-effective control designs and efficient utilization of communication bandwidth. This networked control approach ensures the plantwide stability even when communications between controllers breakdown. Using the concept of dissipative systems, the effects of different local controller sampling rates and information exchange rates on the stability and performance of the entire plant are represented by a set of dissipativity conditions. The conditions that individual local controller has to satisfy to achieve the plantwide stability and user specified performance are derived. Autonomous controllers can then be designed individually to form a controller network. This effectiveness of the proposed approach is demonstrated by an illustrative example of a process network that consists of a reactor and a multi-stages distillation column.