(6cp) Data-Driven Modeling and Control of Batch and Batch-like Processes

Research Interests: Process Modeling and Control, Performance Monitoring, System Identification, Time-frequency Analysis and Forecasting, Machine learning and Causality Analysis

The key motivation that drives my research is the development of methods and strategies for the design of advanced process control systems that (i) account for the complex process dynamics including the non-availability of measurements throughout the process dynamics, (ii) ensure an efficient operation of processes by having relevant model-based control algorithms, (iii) ensure optimal operation of the process by having data-driven process monitoring tools, and (iv) could detect faults in the process as they occur and determine its root cause. My research interests are in the general area of process systems engineering, with emphasis on the development of modeling, control, monitoring and fault detection tools for chemical engineering processes using data-driven approaches. The central objective of my research is the development and implementation of data-driven advanced process control solutions for different chemical engineering applications. By bringing together tools from control and systems theory, chemical engineering and data analytics, we have been able to develop and implement practical solutions to important process control problems including (a) modeling and control of batch chemical engineering processes, (b) analysis and design of controllers, and (c) process network reconstruction using causality analysis.

Postdoctoral Project: “Development of data-driven modeling and control solutions for chemical engineering processes”

Under supervision of Prof. Prashant Mhaskar, McMaster Advanced Control Consortium (MACC), Department of Chemical Engineering, McMaster University

PhD Dissertation: “Data-Driven Modeling and Control of Batch and Batch-Like Processes”

Under supervision of Prof. Prashant Mhaskar, McMaster Advanced Control Consortium (MACC), Department of Chemical Engineering, McMaster University

Master’s Dissertation: “Causality Analysis for Topology Reconstruction and Interaction Assessment”

Under supervision of Prof. Arun K. Tangirala, Indian Institute of Technology Madras

Research Experience: I am currently a postdoctoral fellow working on data-driven modeling and control strategies for chemical engineering systems. My current research projects are in collaborations with McMaster Manufacturing Research Institute (MMRI) and Linde, where I have developed novel data-based modeling and control approaches and is an extension of my doctoral research (e.g., [1-3,5-7]). My research contributions include the development of a framework to utilize big data aspects for modeling and control of the batch particulate process using a dynamic subspace identification based approach coupled with a static product quality model; optimal synthesis of a hydrogen plant startup process; data-driven closed-loop control of a uni-axial rotational molding process. In my masters’ research, I focused on causality analysis and made some significant contributions to the problem of causal network reconstruction and interaction assessment from process data (e.g., [4]). The central objective of my research is the development and implementation of data-centric advanced process control algorithms.

Teaching Interests: Process control, Modern control theory, System identification, Time-series analysis

I view teaching as an integral and very important part of my overall academic activity. I plan to actively participate in the enhancement of the entire undergraduate and graduate curriculum. Given my interdisciplinary background with B.Tech in Electronics and Instrumentation Engineering and Masters’ and PhD in Chemical Engineering, and my experience as a teaching assistant in several undergraduate and graduate courses at Indian Institute of Technology Madras and McMaster University, I feel confident that I can effectively teach any courses in the undergraduate and graduate Chemical Engineering curriculum.

The focus in my teaching would be on developing a curiosity for the subject and enabling students to apply the concepts learned in the classroom to practical problems through careful design of course projects. As a teacher, I believe in the transfer of fundamental subject knowledge to students in such a way that it cultivates their critical thinking skills. I will strive to instill a sense of curiosity in my students that will challenge them to go beyond the typical course requirement. Further, I will value the individual backgrounds and experiences of my students and create a learning environment that’s fair for everyone. I will strive to nurture an environment that will encourage them to seek areas that excite them, as true learning occurs best when it is most meaningful. Further, I will adapt my classroom practices to continuously improve the effectiveness of my teaching based on feedback.

Future Directions:Building upon the training and research experiences that I had during my studies and postdoctoral fellowship; I would like to continue contributing to the systems engineering field. Some of the main directions that I plan to pursue in my future research activities include (a) development of machine learning approaches suitable for process control, (b) development of performance monitoring/decision support tools for large-scale plants, (c) development of hybrid modeling techniques for process control applications, and (d) study the problem of causal inference and its application to process control and monitoring.

Selected Publications (1 book, 6 published peer-reviewed journal articles, 5 conference papers):

[1] Gomes F. P. C., Garg, A., Mhaskar, P. and Thompson M. R. (2019). Data-Driven Advanced in Manufacturing for Batch Polymer Processing Using Multivariate Nondestructive Monitoring, Ind. Eng. Chem. Res. 58 (23), 9940-9951.

[2] Garg, A., Gomes, F. P. C., Mhaskar, P. and Thompson M. R. (2018). Model Predictive Control of Uni-Axial Rotational Molding Process, Comp. & Chem. Eng. 121, 306-316.

[3] Garg, A.and Mhaskar, P. (2018). Utilizing Big Data for Batch Process Modeling and Control, Comp. & Chem. Eng. 119, 228-236.

[4] Garg, A.and Tangirala, A. K. (2018). Metrics for Interaction Assessment using Causality Analysis, Ind. Eng. Chem. Res., 57(3), 967-979.

[5] Garg, A.and Mhaskar, P. (2017). Subspace Identification Based Modeling and Control of Batch Particulate Processes, Ind. Eng. Chem. Res., 56(26) 7491-7502.

[6] Garg, A., Corbett, B., Mhaskar, P., Hu, G., & Flores-Cerrillo, J. (2017). Subspace-Based Model Identification of a Hydrogen Plant Startup Dynamics. Comp. & Chem. Eng., 106, 183-190.

Books:

[7] Mhaskar, P., Garg, A. and Corbett, B. (2019). Modeling and Control of Batch Processes. Springer International Publishing. http://doi.org/10.1007/978-3-030-04140-3