Faculty - John J. Hwalek
B.S. Clarkson College of Technology, 1977
M.S. University of Illinois, 1980
Ph.D. University of Illinois, 1982
Process information systems • heat transfer
- Dynamic Modeling Using Neural Networks
- Engineering Education
Dynamic models of chemical engineering operations are powerful tools which can be used for control, optimization and simulation. Models of complex operations are difficult to formulate particularly when the constitutive relationships (e.g., heat transfer coefficients, kinetic rate equations, etc.) used in the model are not well known. Neural networks have been used to model the dynamics of chemical engineering operations. Typically, the neural network model replaces the first principles model. The neural network in then trained using data for a specific system. The neural network model can not be extended to other similar systems unless it is retrained. My research interest is in exploring hybrid models in which neural networks replace only those parts of the first principles model that are unknown (constitutive equations, physical property variation, etc.) while retaining known relationships (material and energy balances, etc.). Of primary interest is how to train the embedded neural network models and how to extended them to similar systems.
My other research interest in engineering pedagogy. In particular, I am interested in three areas: integration of modern computing tools, integration of design across the curriculum and the use of cooperative learning and teaming to enhance learning.
Modern computing tools (mathematical analysis programs, spreadsheets, dynamic simulators, etc.) give students a powerful way to explore concepts and carry out design calculations. I am investigating ways to integrate these tools into the curriculum that allow students to focus on developing higher order learning skills instead of lower level computations.
Engineering education is often criticized for waiting until late in the curriculum to introduce design. The result of delaying the introduction of design is that students have difficulties connecting the science and mathematics they learn early in the curriculum to engineering problem solving. I have been developing and testing design experiences that require students to apply the knowledge they have gained to solve engineering problems. The major challenge is formulating problems that require analysis and synthesis but are at a level appropriate to the students’ current knowledge and skill base.
Many research studies have shown the benefits of cooperative learning. Cooperative learning can be implemented formally through team projects and informally in the classroom or in study groups. My interest is to promote cooperative learning by educating students about its advantages and instituting it through formal and informal methods.
Cutlip, Michael B., Hwalek, John J., Nuttall, H. Eric, Shacham, Mordechai, Brule, Joseph, Widmann, John, Han, Tae, Finlayson, Bruce, Rosen, Edward M., and Taylor, Ross, “A Collection of Ten Numerical Problems in Chemical Engineering Solved by Various Mathematical Software Packages”, Computer Applications in Engineering Education, Vol. 6, No. 3, 1998.
J. Hwalek, “The Use of Mathematical Software in Chemical Engineering”, Chemical Engineering Summer School, Snowbird, Utah, August, 1997.