- BASc, Technion Israel Institute of Technology
- MSc, Technion Israel Institute of Technology
- PhD, Technion Israel Institute of Technology
Chemical process control:
- Robust control
- Nonlinear control
- Model Predictive Control (MPC)
Models of systems are always inaccurate. Control systems which are based on process models have to be designed to deal with this model mismatch. The applicability of advanced model model-based control algorithms in the chemical industry depends on their ability to tolerate model error.
We conduct extensive research in the areas of robust control algorithms which are tolerant to model mismatch. For example, we investigate robustness of Model Predictive Controllers, which is currently the most widely used multivariable control algorithm in the chemical industry. We are also studying Robust Economic Predictive Control where the objective is to maximize economic profits of chemical factories in the presence of constraints and using models with uncertainty.
- modeling, control and optimization of biological processes
- metabolic flux analysis
- dynamic metabolic flux models
- run-to-run (batch to batch) optimization of pharmaceutical process
We are investigating systematic mathematical modeling approaches to predict the physiology, growth behavior and productivity of CHO and hybridoma cells in both batch and fed-batch systems. The models proposed in these studies are significantly more comprehensive that the ones presented in the literature since it explicitly link biomass and monoclonal antibody production to all aminoacids present in the culture and to intracellular compounds. We perform metabolic flux analysis to understand the metabolism of mammalian and bacterial cells used for manufacturing of monoclonal antibodies or vaccine antigens.
The models are used for product optimization and are verified with extensive experimental data. On this subject we maintain active industrial collaborations with Sanofi Pasteur and Sigma-Aldrich.
For example we are currently investigating the optimization of growth media using successive optimization while simultaneously identifying the model from batch data.
Monitoring/Fault Detection of chemical and biological/biotechnological processes:
- Multivariate statistical methods for monitoring and fault detection.
- Inferential (soft) sensors
Monitoring is a key activity in process systems engineering which is essential for ensuring safe and profitable operation of chemical systems. Our group investigates the application of statistical tools to monitor and detect faults in different chemical systems including pharmaceutical operations, water purification and fermentation of animal cells. We are also studying the use of stochastic models to detect faults in chemical processes and for diagnosing the sources of these faults.
- A novel multivariate statistical model was developed to predict fouling in ultra-filtration operation for drinking water applications. The predictions are based on the fluorescence spectra of retentate and permeate streams. This project was done in collaboration with GE Water and Power.
- A spectro-fluorometer was used to monitor manufacturing of an acellular vaccine in collaboration with Sanofi Pasteur.
- We are currently actively investigating the use of flow cytometry to monitor a vaccine manufacturing process by tracking the evolution of the cell population within a large industrial bioreactor.
- We are investigating the use of stochastic models (using PCEs-polynomial chaos expansions) to perform model based fault detection of chemical processes with tolerance to errors in the model.
- Zavatti, V., Legge, R. and H. Budman, Monitoring of an antigen manufacturing process, Bioprocesses and Biosystems Engineering, (39), 855-869, 2016.
- Du Y., Duever T. and H. Budman, Fault Detection and Diagnosis with Parametric Uncertainty using Generalized Polynomial Chaos Computers & Chemical Engineering, 76,63-75, 2015.
- Mandur, J., Budman, H., Simultaneous model identification and optimization in the presence of model mismatch, Chemical Engineering Science, 129, 106-115, 2015.
- Ohadi, K, Agamohseni, H, Legge, R, Budman, H., Fluorescence-based soft sensor for at situ monitoring of chinese hamster ovary cell cultures, Biotechnology and Bioengineering, 111(8), 1577-1586, 2014.
- Kumar D,Budman H, "Robust Nonlinear MPC based on Volterra series and Polynomial Chaos Expansions", Journal of Process Control, (24(1), 304-317, 2014
- Mandur J, Budman H, "Robust Optimization of Chemical Processes using Bayesian description of Parametric Uncertainty", Journal of Process Control, invited paper to the special issue of the IFAC Dycops 2013, 24(2), 422-430, 2014
- Meshram, M., Naderi, S., McConkey, B., Ingalls, B.,Scharer, J., Budman, Modeling the coupled extracellular and intracellular environments in mammalian cell culture, Metabolic Engineering, 19, 57-68, 2013.
- Shams, M, Duever, T. And H. Budman, Fault detection, identification and diagnosis using CUSUM based PCA, Chemical Engineering Science, 66(20),4488-4498, 2011