Chemical Engineering Professor Nikolaos Kazantzis' Theory Becoming An Accepted Practice
Last month, a pioneering process systems engineering theory developed by WPI professor Nikolaos Kazantzis was featured at the 2022 European Control Conference (ECC) in London, a high-profile gathering of academic and industrial professionals in the systems and control fields. The theory, known as the “Kazantzis-Kravaris/Luenberger (KKL) or non-linear Luenberger observer methods,” is a nonlinear control and state estimation theory that reliably assesses and improves the overall performance profile of complex process systems.
Emerging chemical, material, and clean energy technologies are helping confront climate change and enhance the prospects for a sustainable future, and Kazantzis’ theory could help give regulators and researchers a more accurate view of how the attendant process systems behave in the real world. If regulators and researchers continue to rely on conventional, linear analysis, society may not be fully aware of the inherent risks, potential upside opportunities and inevitable process system performance trade-offs that come along with those advancements, Kazantzis said.
“It’s important to rely on sound science rather than arbitrary decision-making,” he said.
Kazantzis’ work conceptually builds on linear control and estimation theory but, as he points out, “the way inherently complex processes/systems behave in the real world does not always follow the idealized behavior prescribed by a linear path.” In light of this realization, Kazantzis rigorously developed nonlinear analogues as well as the requisite analytical and computational tools, leading to a more accurate characterization of how process systems behave in an environment endowed with layers of complexity that cannot be captured by conventional methods.
Kazantzis helped develop a new approach that allows for a reliable dynamic reconstruction of a (bio)chemical substance’s concentration profile that is not always attainable under stringent accuracy/reliability requirements, due to inherently nonlinear behavior and various uncertainties.
Developed 20 years ago, the work has recently generated interest in the fields of artificial intelligence, machine learning and data science; potential real-life applications have piqued the interest of governments and industry in the areas of clean energy production systems, environmental systems, chemical risk assessment, and environmental health exposure studies. For example, international efforts by the United Nations Environment Program, the Organization for Economic Cooperation, and Development and the European Environment Agency aim to standardize the classification and regulation of chemicals in an effort to enhance protection of human health and the environment.
At a presented tutorial at the ECC, academics and other experts in the field recognized Kazantzis’ contributions and had the opportunity to also assess their impact in emerging and promising areas of scientific inquiry, such as machine learning, data science and forecasting.
Kazantzis, who was recently re-elected as a senior member and associate at Hughes Hall College, University of Cambridge, said he is grateful to have his work recognized at the ECC, but noted he has not worked alone and science advances through collective efforts. He has further developed his thinking and broadened his theory’s real-world impact over the years through collaborations with partners across the globe and his work with undergraduate and graduate students.
“I feel a deep sense of gratitude to be recognized by members of our scientific community whom I have personally admired for a long time, inspired by their own contributions as well as benefited from fruitful collaborations and interactions over the years,” Kazantzis said.