CHIU, Min Sen

Associate Professor

PhD (Chem. Eng.) Georgia Tech., 1991
BSc (Chem. Eng.) Natn’l Taiwan, 1983

Contact information
Blk E5, 4 Engineering Drive 4, #03-08, Singapore 117576
Tel: (65) 6516 2223   Fax: (65) 6779 1936
Email: checms@nus.edu.sg

RESEARCH

Model-Free Controller Design Methods

Model-based techniques have been the predominant controller design methods used in the chemical and biochemical industries, for example, PID tuning rules based on the low-order transfer function models. Although such models can be easily obtained from identification experiments and practical for industrial applications, these empirical models suffer several shortcomings. First, they may not provide adequate information for the underlying process dynamics and thus performance of the resulting model-based controller design may become poor if the modeling error is too large. Second, even when these models have acceptable modeling accuracy, the model with better accuracy does not necessarily guarantee the better performance of the resulting model-based controller. As a result, model-based design methods need to resort to the trial and error procedure to cope with such interplay between model identification and controller design, which places additional burden on the users of these model-based design methods. To alleviate the aforementioned problems, model-free design methods without resorting to the identification of a process model are investigated in this research to address various controller design problems of importance to industrial process control.

Data-Based Methods for Nonlinear Process Control

In the chemical process industries, an increasing number of process variables are measured and stored in the distributed control systems. To draw benefits from this rich information, techniques are needed to turn the available process data into working knowledge. The aim of this research is to develop control strategies for nonlinear systems directly based on the process data available in the control systems.

Control of Pharmaceutical Crystallization

One of the most important problems that can arise in the development of a pharmaceutical crystallization process is the control of polymorphism, in which there exist different crystal forms for the same chemical compound. Different polymorphs can have very different properties such as bioavailability, which motivates the design of controlled processes to ensure consistent production of the desired polymorph to produce reliable therapeutic benefits upon delivery. In this study, we are developing batch control strategies with the goal of optimizing batch productivity while providing robustness to variations that can occur in practice, for exmaple variations in the feed-stocks.

SELECTED PUBLICATIONS

Nuella, I., Cheng, C., and M. S. Chiu. "Adaptive PID Controller Design for Nonlinear Systems", Industrial Engineering Chemistry Research, 48, 4877-4883 (2009).

Kansha, Y. and M. S. Chiu. "Just-in-Time Learning-Based Generalized Predictive Control", Journal of Process Control, 19, 1067-1072 (2009).

Hermanto, M. W., Braatz, R. D., and M. S. Chiu. "High-order Simulation of Polymorphic Crystallization Using Weighted Essentially Non-oscillatory Methods", AIChE J., 55, 122-131 (2009).

Cheng, C. and M. S. Chiu. "Robust PID Controller Design for Nonlinear Processes Using JITL Technique", Chemical Engineering Science, 63, 5141-5148 (2008).

Kansha, Y., Jia, L. and M. S. Chiu. "Self-Tuning PID Controllers Based on the Lyapunov Approach", Chemical Engineering Science, 63, 2732-2740 (2008).

Kalmukale, A. K., Chiu, M. S. and Q. G. Wang. "Partitioned Model-Based IMC Design Using JITL Modeling Technique", Journal of Process Control, 17, 757-769 (2007).

Hermanto, M. W., Chiu, M. S., Woo, X. Y. and R. D. Braatz. "Robust Optimal Control of Polymorphic Transformation in Batch Crystallization", AIChE J., 53, 2643-2650 (2007).

Jia, L., Chiu, M. S. and S. S. Ge. "A Noniterative Neuro-Fuzzy Based Identification Method for Hammerstein Processes", Journal of Process Control, 15, 749-761 (2005).

Jia, L., Chiu, M. S. and S. S. Ge. "Iterative Identification of Neuro-Fuzzy Based Hammerstein Model With Global Convergence", Industrial Engineering Chemistry Research, 44, 1823-1831 (2005).