(3cn) At the Interface of Ab-Initio Modeling and Data Analytics: A Hybrid Approach to Process Design, Control, and Optimization

The effort to mimic a chemical plant’s operations or to design and operate a completely new technology in silico is a highly studied research field under Process Systems Engineering. As the rising computation power allows us to simulate and model systems in greater detail through careful consideration of the underlying phenomena, identifying the optimal design/operating criteria of such systems becomes a formidable challenge due to the noise, complexity, lack of analytical forms, and the excessive computational effort and time required to find a viable solution. My primary research interest is developing computational algorithms for solving such large-scale complex optimization problems without the full knowledge of the underlying mathematical model (black-box) through amalgamating data analysis, data-driven modeling, and global optimization theory.

The novel research strategy that I intend to bring will create computational tools that hybridize first-principles modeling and state-of-the-art data analytics for the design, control, and optimization of engineering systems. The ultimate goal of such an integrated approach is to pipeline the accelerated adoption of new technologies, starting from the initial design to its implementation in real-life. Specific application areas will include:

1. Control and maintenance optimization with applications to intensified processes,
2. Clean energy design for wastewater treatment applications in chemicals production,
3. Multi-scale modeling for solid-state batteries in the era of “Materials Genomics.”

Teaching Interests:

My experience in being a teaching assistant to five undergraduate chemical engineering courses (Heat Transfer Operations, Fluid Operations, Numerical Methods, Introduction to Chemical Engineering, and Water Technology Innovation), a guest lecturer in 3 graduate-level process optimization courses, and a co-instructor in Data Sciences and Multi-Scale Systems Engineering course has equipped me with the skills to design and teach any course from the undergraduate curriculum. Given my strength and expertise in computational modeling, I could be valuable in teaching Introduction to Chemical Engineering, Process Design, Optimization of Chemical and Biological Processes, and Numerical Analysis. A key component of my educational strategy will be incorporating the emerging field of data analytics to the chemical engineering undergraduate curriculum, where I will also be teaching the systems approach to tackle challenging tasks and problems. Additionally, I am looking forward to designing and delivering an advanced course on Applications of Data Analytics in Chemical Engineering with a focus on process modeling design and optimization.

My main goal as a prospective instructor is to create a solid foundation of Chemical Engineering knowledge for each student while challenging them to think outside the box. Inside the classroom, I plan to follow an unconventional teaching strategy where I will be implementing a classical direct instruction technique such as lecturing, together with interactive, experiential, and indirect instructional techniques. I will also be adopting new technologies in my teaching strategy to capture different learning styles of diverse classrooms. I will be providing the class material in various formats including, lecturing on the board, using visual aids (both in presentation form and the lecture material can be supported by relevant videos), solving computer-based exercises, and providing organized lecture notes and reading materials prior to the class. This range of approaches towards the students’ differences in learning will also be applied in their assessment. Outside the classroom, I will be preparing help sessions and providing additional office hours to personally make myself available for one-on-one guidance on the course material.

Selected Publications:

• Beykal, B.; Onel, M.; Onel, O.; Pistikopoulos, E.N. A Data-Driven Optimization Algorithm for Differential Algebraic Equations with Numerical Infeasibilities. AIChE Journal (Under Review).
• Beykal, B.; Avraamidou, S.; Pistikopoulos, I.P.E.; Onel, M.; Pistikopoulos, E.N. DOMINO: Data-driven Optimization of bi-level Mixed-Integer NOnlinear Problems. Journal of Global Optimization, 2020, DOI: 10.1007/s10898-020-00890-3
• Beykal, B.; Boukouvala, F.; Floudas, C.A.; Pistikopoulos, E.N. Optimal Design of Energy Systems Using Constrained Grey-Box Multi-Objective Optimization. Computers & Chemical Engineering, 2018, 116, 488-502. (Invited for special issue: Multi-scale Systems Engineering – in memory & honor of Professor C.A. Floudas)
• Beykal, B.; Boukouvala, F.; Floudas, C.A.; Sorek, N.; Zalavadia, H.; Gildin, E. Global Optimization of Grey-Box Computational Systems Using Surrogate Functions and Application to Highly Constrained Oil-Field Operations. Computers & Chemical Engineering, 2018, 114, 99-110. (Invited for special issue: FOCAPO/CPC 2017)