Chemical plants are large-scale networked systems with tightly interconnected component subsystems coupled through material, energy and information flows and recycle. The ability of these systems to operate profitably and reliably requires the design and implementation of advanced control systems that can manage the complex dynamics and interactions between the component subsystems. This realization has motivated significant research work on the control and supervision of large-scale process networks, leading to a host of control design methodologies ranging from centralized to decentralized and distributed control architectures (e.g., see , , ,  for some recent results and references).
In addition to handling complex network dynamics, another important consideration in the design of control systems for large-scale process networks is the integration of computing, communication and control at different levels of plant operations. With the emergence of the Smart Plant paradigm in recent years  and the increased calls for the deployment of wireless sensor/actuator networks and other advanced communication technologies in process monitoring and control systems, there is a need to address the fundamental challenges posed by the integration of communication and control. The intrinsic limitations on the computational and communication resources of wireless devices, for example, favor keeping the information exchange over the network to a minimum so as to conserve network resources, while maximizing process performance favors increased communication levels between the plant subsystems.
An approach to resolve this conflict was proposed in  using a quasi-decentralized networked control structure designed to balance plant performance requirements with communication costs. A key idea was to embed, within each local control system, a set of predictive models that generate estimates of the states of the neighboring units when communication is suspended, and to update the models when communication is restored. Using a fixed communication rate policy, an explicit characterization of the minimum allowable communication rate was obtained. In a subsequent study , a quasi-decentralized control structure using a state-dependent communication logic was developed. The key idea there was to have each unit monitor the evolution of the local state and prompt the other units to broadcast their measurements to update their models only when the unit is on the verge of instability. Compared with the fixed communication rate approach, this approach allows the plant to respond to changes in operating conditions by varying the communication rate; however, it can also lead to increased network utilization and the possibility of delays and data losses when all plant units attempt to access the network at the same time.
An alternative approach for dealing with communication constraints is the use of event triggered control strategies (e.g., -). In these strategies, communication over the network is typically suspended and restored in response to certain events tied to the desired closed-loop stability and performance properties. These strategies are appealing in the context of sensor/actuator networks where reducing network utilization can also reduce the energy expenditures of battery-powered wireless devices. In these studies, however, the problems of detecting and handling faults in the networked control system design were not addressed. These are important problems as fault-handling is essential to maintaining the desired closed-loop stability and performance properties. Furthermore, the presence of faults, if not appropriately dealt with, can lead to unnecessary increases in communication levels.
In this work, we present an event-triggered fault-tolerant control framework for networked process systems controlled over a resource-constrained communication medium. The objective is to maintain closed-loop stability in the presence of faults while minimizing unnecessary network utilization. To meet this objective, a set of stabilizing distributed controllers are initially designed on the basis of an approximate plant model. The local controllers utilize the available model to generate the local control action during periods of communication suspension. Updates of the model states using real-time measurements are triggered whenever the local model estimation error breaches a certain stability threshold. This threshold is designed using Lyapunov techniques and is explicitly parameterized in terms of the fault size, the plant-model mismatch and the various controller design parameters. This parameterization provides the basis for selecting a suitable fault accommodation strategy that enforces closed-loop stability, while simultaneously optimizing network resource utilization. To aid fault accommodation, fault detection and estimation are carried out locally within each subsystem using a moving-horizon optimization scheme. Based on the fault identification result, several fault accommodation measures are assessed, including adjusting the local model or controller parameters, tightening the local model estimation error threshold to increase communication frequency, and system reconfiguration at the supervisory level. The choice of a suitable strategy is made on the basis of the desired balance between control and communication requirements. Finally, the developed methodology is illustrated using a reactor-separator process network.
 P. D. Christofides, J. Davis, N. H. El-Farra, D. Clark, K. R. D. Harris and J. N. Gipson, ``Smart Plant Operations: Vision, Progress and Challenges,'' AIChE J., 53, 2734-2741, 2007.
 K. R. Jillson and B. E. Ydstie, "Process networks with decentralized inventory and flow control," J. Process Control, 17, 399-413, 2007.
 S. Jogwar, M. Baldea and P. Daoutidis, "Dynamics and control of process networks with large energy recycle," Ind. Eng. Chem. Res., 48, 6087-6097, 2009.
 Y. Sun and N. H. El-Farra, "Quasi-decentralized model-based networked control of process systems,'' Comp. Chem. Eng., 32, 2016-2029, 2008.
 Y. Sun and N. H. El-Farra, “Quasi-decentralized networked process control using an adaptive communication policy,” Proceedings of American Control Conference, pp. 2841–2846, 2010.
 B. T. Stewart, S. Wright and J. B. Rawlings, "Cooperative distributed model predictive control for nonlinear systems," J. Process Control, 21, 698-704, 2011.
 P. D. Christofides, J. Liu and D. M. de la Pena. Networked and Distributed Predictive Control: Methods and Nonlinear Process Network Applications. Springer-Verlag, London, 2011.
 M. Mazo and P. Tabuada, “Decentralized event-triggered control over wireless sensor/actuator networks,” IEEE Transactions on Automatic Control, 56, 2456–2461, 2011.
 X. Wang and M. D. Lemmon, “Event-triggering in distributed networked control systems,” IEEE Transactions on Automatic Control, 56, 586–601, 2011.
 Y. Hu and N. H. El-Farra, ``Quasi-decentralized output feedback model predictive control of networked process systems with forecast-triggered communication," Proceedings of American Control Conference, pp. 2612-2617, Washington, DC, 2013.