Predictive Control of Nonlinear Chemical Processes under Asynchronous Measurements and Controls

AbstractIn many process control problems measurement and control instances might not be available in a periodically-equally-distributed way. Moreover, due to the sensor processing time, actuators/sensors calibration, or computation, inevitable delays can often arise. Also information losses caused, for example, by temporary components failure, or the presence of unreliable communication media, might represent a non-trivial problem. This leads to asynchronous availability of measurement and control inputs, i.e. the controller, sensor, and actuator work in an event-driven, rather than a continuous way. In order to avoid instability and performance loss all these issues must be considered during the control design. In this paper, it is shown that predictive control methods based on continuous time models can be used to stabilize event-based nonlinear systems under variable delays, and limited information losses. It is demonstrated that by using the suggested approach asymptotic convergence is ensured.