Selected projects completed in 2023
You can find other projects on the website of the FAV ZČU departments.
CZ.01.1.02/0.0/0.0/21_374/0026768
DMS of the new generation
The output of the project is a modern web-based Document Management System utilizing elements of self-learning artificial intelligence and machine learning, focusing on neural network technologies. The product is unique on the market thanks to the use of the latest technologies and the most recent advances in semantic search and automatic natural language processing. This has resulted in a significant increase in search accuracy and the automation of many tasks that previously had to be performed manually. By focusing on transferable learning and language-independent methods, trained models can be transferred from one market to another, enabling deployment in multilingual environments.
The competitive advantage of the solution is further enhanced by the integration of Blockchain technology, which, when needed, allows for the creation of transparent and immutable records in a selected Blockchain network. The Document Management System has been designed to achieve a high degree of configurability and administrative control by customer administrators. The system also includes a data structure designer and a graphical workflow designer, allowing for verification of created processes.
The final product incorporates the latest knowledge and research outcomes from specific areas of artificial intelligence, acquired in cooperation with a research and development organization and other partners. A partial goal of the project was to innovate the existing system while maintaining its full range of functionality. The project includes both experimental development and industrial research activities. The solution builds on the applicant’s many years of experience and on their previously successful, but now technologically outdated, product. For the development of new functionalities and to ensure the final product is the best available solution, the applicant partnered with the University of West Bohemia in Pilsen, a leading technology-focused scientific and research institution.
Provider: Ministry of Industry and Trade - OP PIK
Project leader: Ing. Miloslav Konopík, Ph.D., VP2
Solution period: 2021–2023
21-16406S
Nonlinear Acoustics and Transport Processes in Porous Periodic Structures
The project is concerned with multi-time scale modelling of nonlinear phenomena related to acoustic wave propagation in periodic porous structures. Besides mechanical fluid-structure interaction, the mass-heat transfer will be examined in the context of acoustic streaming influenced by temperature gradients and deforming solid scaffolds containing piezoelectric components and resonating tunable parts. Microstructure geometry, interface roughness, porosity and cavitation will be respected to model mass and/or heat fluxes in response to acoustic waves. Using analytical and numerical methods based on the perturbation theory and time-space homogenization, multi-scale models will be derived and implemented to provide computational basis for simulations of the above mentioned effects and to optimize the composition and geometry of periodic scaffolds. Algorithms for reduced order modelling will be developed to allow for efficient treatment of the slowly evolving dynamic systems driven by acoustic waves. The research outputs will contribute to improved designs of thermo-acoustic engines.
Provider: Czech Science Foundation
Project leader: prof. Dr. Ing. Eduard Rohan, DSc., KME a VP3
Solution period: 2021–2023
21-31457S
Fast flow-field prediction using deep neural networks for solving fluid-structure interaction problems
Fluid-structure interaction (FSI) problems that occur in various industries including aeronautics, turbomachinery and nuclear energy tend to be very complex and their simulation extremely demanding. The innovative idea in the proposed project is to overcome high computational costs by applying deep learning to solve FSI problems. To be more specific, the idea is to substitute a traditional CFD solver with a deep neural network. Deep neural networks are just starting to be popular for fast flow-field predictions and have been successfully used for a narrow range of problems, however, never have they been applied to predict fluid flows with a moving boundary, not to mention complex FSI problems. The target of the proposed project is to develop and implement a deep-neural-network architecture capable of predicting flow fields with a moving boundary and couple it with a structure solver to obtain a powerful FSI framework. The developed neural-network architecture and FSI solver will be benchmarked on problems of external aerodynamics in subsonic and transonic regimes.
Provider: Czech Science Foundation
Project leader: prof. Dr. Ing. Jan Vimmr, Ph.D., KME a VP3
Solution period: 2021–2023
FW01010153
Lightweight axial fan blade for harsh operating conditions
The aim of the project is to design a hybrid lightweight blade of axial fan for difficult operating conditions, which consist in an elevated operating temperature above 200 °C and in the unfavorable composition of the transported air.
Provider: TAČR
Project leader: Ing. Jan Krystek, Ph.D., KME a VP3
Solution period: 2020–2023
CZ.02.1.01/0.0/0.0/17_048/0007267
Research and Development of Intelligent Components of Advanced Technologies for the Pilsen Metropolitan Area (InteCom)
The research goal of the project is to develop and test in the laboratory the IT technologies (mostly as software or industrial digital systems) in scalable price and performance parameters to convert existing theories, knowledge, principles, methods and algorithms of automation, robotics, artificial intelligence, monitoring, diagnostics and signal processing in a form that will significantly accelerate and make visible the possibility of their application in practice and thus contribute to the solution of a big social theme in the context of Industry 4.0.. The project outputs will have potential for applications mainly in enterprises of the Pilsen metropolitan area and should to strengthen the cooperation in the future between the NTIS Research Center and the application sphere.
Provider: Ministry of Education, Youth and Sports
Project leader: Doc. Ing. Eduard Janeček, CSc., NTIS
Solution period: 2018–2023
CZ.02.1.01/0.0/0.0/16_026/0008389
Research Cooperation for Higher Efficiency and Reliability of Blade Machines
The research activities are aimed at building a strong interdisciplinary team, which will develop new methods aimed at streamlining and refining development and projection work on flow parts of bladed machines. The methods include new procedures for the identification and validation of the impact of the flow-through sealing elements on the stability of bladed machine rotors and enable a more comprehensive solution of rotor dynamics problems in order to predict and prevent their unstable behavior. Relevant is the research of the representation of the static and dynamic deviation model of flow geometry from nominally designed, which is obtained in combination with 3D measuring procedures, and its consequences on the current parameters and efficiency of the proposed machine. The intention is to increase the reliability and durability of the rotor systems while ensuring maximum efficiency.
Provider: Ministry of Education, Youth and Sports
Project leader: Doc. Ing. Eduard Janeček, CSc., NTIS
Solution period: 2018–2023