Real-Time Monitoring of Powder Mass Flowrates for MPC/PID Control of a Continuous Direct Compaction Tablet Manufacturing Process
thesisposted on 30.07.2020 by Yan-Shu Huang
In order to distinguish essays and pre-prints from academic theses, we have a separate category. These are often much longer text based documents than a paper.
To continue the shift from batch operations to continuous operations for a wider range of products, advances in real-time process management (RTPM) are necessary. The key requirements for effective RTPM are to have reliable real-time data of the critical process parameters (CPP) and critical quality attributes (CQA) of the materials being processed, and to have robust control strategies for the rejection of disturbances and setpoint tracking.
Real-time measurements are necessary for capturing process dynamics and implement feedback control approaches. The mass flow rate is an additional important CPP in continuous manufacturing compared to batch processing. The mass flow rate can be used to control the composition and content uniformity of drug products as well as an indicator of whether the process is in a state of control. This is the rationale for investigating real-time measurement of mass flow of particulate streams. Process analytical technology (PAT) tools are required to measure particulate flows of downstream unit operations, while loss-in-weight (LIW) feeders only provide initial upstream flow rates. A novel capacitance-based sensor, the ECVT sensor, has been investigated in this study and demonstrates the ability to effectively measure powder mass flow rates in the downstream equipment.
Robust control strategies can be utilized to respond to variations and disturbances in input material properties and process parameters, so CQAs of materials/products can be maintained and the amount of off-spec production can be reduced. The hierarchical control system (Level 0 equipment built-in control, Level 1 PAT based PID control and Level 2 optimization-based model predictive control) was applied in the pilot plant at Purdue University and it was demonstrated that the use of active process control allows more robust continuous process operation under different risk scenarios compared to a more rigid open-loop process operation within predefined design space. With the aid of mass flow sensing, the control framework becomes more robust in mitigating the effects of upstream disturbances on product qualities. For example, excursions in the mass flow from an upstream unit operation, which could force a shutdown of the tablet press and/or produce off-spec tablets, can be prevented by proper control and monitoring of the powder flow rate entering the tablet press hopper.
In this study, the impact of mass flow sensing on the control performance of a direct compaction line is investigated by using flowsheet modeling implemented in MATLAB/Simulink to examine the control performance under different risk scenarios and effects of data sampling (sampling time, measurement precision). Followed by the simulation work, pilot plant studies are reported in which the mass flow sensor is integrated into the tableting line at the exit of the feeding-and-blending system and system performance data is collected to verify the effects of mass flow sensing on the performance of the overall plant-wide supervisory control.