MPC-based reference governor control of a continuous stirred-tank reactor

Model predictive control for steady-state performance in integrated continuous bioprocesses

Perfusion bioreactors are commonly used for the continuous production of monoclonal antibodies (mAb). One potential benefit of continuous bioprocessing is the ability to operate under steady-state conditions for an extended process time. However, the process performance is often limited by the feedback control of feed, harvest, and bleed flow rates. If the future behavior of a bioprocess can be adequately described, predictive control can reduce set point deviations and thereby maximize process stability. In this study, we investigated the predictive control of biomass in a perfusion bioreactor integrated to a non-chromatographic capture step, in a series of Monte-Carlo simulations. A simple algorithm was developed to estimate the current and predict the future viable cell concentrations (VCC) of the bioprocess. This feature enabled the single prediction controller (SPC) to compensate for process variations that would normally be transported to adjacent units in integrated continuous bioprocesses (ICB). Use of this SPC strategy significantly reduced biomass, product concentration, and harvest flow variability and stabilized the operation over long periods of time compared to simulations using feedback control strategies. Additionally, we demonstrated the possibility of maximizing product yields simply by adjusting perfusion control strategies. This method could be used to prevent savings in total product losses of 4.5–10% over 30 days of protein production.

Reinforcement Learning-Based Control and Networking Co-design for Industrial Internet of Things

Industrial Internet-of-Things (IIoT), also known as Industry 4.0, is the integration of Internet of Things (IoT) technology into the industrial manufacturing system so that the connectivity, efficiency, and intelligence of factories and plants can be improved. From a cyber physical system (CPS) perspective, multiple systems (e.g., control, networking and computing systems) are synthesized into IIoT systems interactively to achieve the operator's design goals. The interactions among different systems is a non-negligible factor that affects the IIoT design and requirements, such as automation, especially under dynamic industrial operations. In this paper, we leverage reinforcement learning techniques to automatically configure the control and networking systems under a dynamic industrial environment. We design three new policies based on the characteristics of industrial systems so that the reinforcement learning can converge rapidly. We implement and integrate the reinforcement learning-based co-design approach on a realistic wireless cyber-physical simulator to conduct extensive experiments. Our experimental results demonstrate that our approach can effectively and quickly reconfigure the control and networking systems automatically in a dynamic industrial environment.