Research interests
For my detailed publication history please check my publication list or visit my Google Scholar and ResearchGate profiles.
In modern chemical production plants the occurring discrete events (alarms, warnings, operator actions, system messages, etc.) are recorded in alarm management databases. These historical datasets give us the opportunity to build data models for advanced fault diagnosis solutions and in general to help and reduce the operator work. This concept is illustrated in the figure below .
Recently, we published two novel solutions in advanced alarm management. We demonstrated how we can determine the important and characteristic process alarms with a deep learning based approach and developed a Bayes' theroem based alarm suppression approarch.
The search for compounds exhibiting desired physical and chemical properties is an essential, yet complex problem in the chemical, petrochemical, and pharmaceutical industries. During the formulation of this optimization-based design problem two tasks must be taken into consideration: the automated generation of feasible molecular structures and the estimation of macroscopic properties based on the resultant structures. This concept is illustrated in the following figure.
For the structural characteristic-based property prediction task numerous methods are available. However, the inverse problem, the design of a chemical compound exhibiting a set of desired properties from a given set of fragments is not so well studied. Since in general design problems molecular structures exhibiting several and sometimes conflicting properties should be optimized, we proposed a methodology based on the modification of the multi-objective Non-dominated Sorting Genetic Algorithm-II (NSGA-II).
We have published a research article in the topic and an additional one is under revision at the moment.
The key idea of our research is that strategic plans, sustainability reports and scientific studies reflect these variables, therefore, with the tools of text mining, the most important focus points and interactions can be determined. These key aspects and their connections can be represented by a network structure and compared to the subsystems of the dynamic models of sustainability to explore the deficiencies of the models or the lack of focus of the related policies and documentations.
In our study, the proposed methodology through the analysis of five strategical documents is demonstrated and the determined aspects with the structure of the famous World3 model compared.
The energy demand of the corn-based bioethanol production could be reduced using the agricultural byproducts as bioenergy feedstock for biogas digesters. The release of lignocellulosic material and therefore the acceleration of degradation processes can be achieved using thermal and mechanical pretreatments, which assist to hydrolyze the cell walls and speed the solubilization of biopolymers in biogas feedstock.
Our study was focused on liquid hot water, steam explosion and ultrasonic pretreatments of corn stover.
