http://hdl.handle.net/20.500.14044/33868 2026-07-21T07:04:47Z 2026-07-21T07:04:47Z Faltin, Zsolt Beneda, Károly http://hdl.handle.net/20.500.14044/32929 2025-09-18T08:35:56Z 2024-01-01T00:00:00Z

Combined Numerical and Experimental Investigation of a Centrifugal Compressor with Surge Supression Holes at the Impeller Hub Faltin, Zsolt; Beneda, Károly Centrifugal compressor surge is a very dangerous phenomenon to the working environment of a compression system. Unfortunately, the most efficient working range is close to the boundary of stable operation and surge situation. One of the main aims of this research project is to find a solution which can help to operate centrifugal compressor close to the instabilities with a much higher level of safety. A numerical investigation was made to examine the flow field and the centrifugal compressor behaviour at the onset of surge conditions both when compressor discharge bleed air was set and when the bleed air valve was closed. With the help of the numerical investigation, the calculation of the flow field was calculated in every compressor impeller blade passage. In this way the development and the behavior of the injected bleed air to the main flow field in the impeller blade passages could be analyzed. The results of the numerical investigation was validated later with a test bench with the exact geometry of the numerical investigation.

2024-01-01T00:00:00Z Serban, Andrada A. Micskei, Zoltán http://hdl.handle.net/20.500.14044/32926 2025-09-18T08:38:51Z 2024-01-01T00:00:00Z

Application of Mutation testing in Safety- Critical Embedded Systems: A Case Study Serban, Andrada A.; Micskei, Zoltán Mutation testing is a technique used for evaluating test efficiencies, by analyzing whether existing tests could detect minor modifications inserted in the source code. Despite its proven benefits and added value to the verification and validation process, mutation testing is yet to become a widespread practice in safety-critical software development, mostly due to issues around its scalability in industrial environments. In this case study, conducted at Knorr-Bremse Rail Systems, we created a lightweight mutation testing tool, tailored to the specific test environment of the company, showing how such tools can be created with a notably smaller workload, than estimated in previously published case studies. Mutation testing was used to analyze automatically generated and manually complemented coverage-based tests of an entire braking system component. Mutation testing was able to reveal deficiencies not uncovered by standardized, coverage-based testing. The experience added to the body of knowledge on the application of mutation testing, in safety-critical embedded systems, strengthening the fault-finding capability of mutation testing reported by earlier related studies, but pointing out how the one-mutant- per-line optimization was less useful in the given setting. The findings resulted in the definitive, strategic implementation of the created tool within the company’s component testing workflow and could help replicate the results in other case studies, aiding companies in introducing mutation testing in their work environment.

2024-01-01T00:00:00Z Somlyai, László Vámossy, Zoltán http://hdl.handle.net/20.500.14044/32925 2025-09-18T08:38:39Z 2024-01-01T00:00:00Z

Improved RGB-D Camera-based SLAM System for Mobil Robots Somlyai, László; Vámossy, Zoltán The paper presents an improved Simultaneous Localization and Mapping (iSLAM) and 3D reconstruction system for a mobile robot carrying an RGB-D camera. The developed system generates a three-dimensional point cloud from the color and depth camera data of the RGB-D camera. The matching is performed on successive point clouds that partially overlap. After feature detection on the color camera images, the method selects the 3D points in the successive point clouds that belong together. During the iterative multi-step matching algorithm based on Singular Value Decomposition (ISVD), it minimizes the matching error between the selected point clouds and deletes non-real point pairs in the process. Our previous SLAM method has been improved in several ways. On the one hand, a conditional averaging filter (Distance Image Filter: DIF) was created for the depth camera data to reduce the noise. The matching algorithm iteratively determines the matching transformation and the estimated displacement from the feature points of several recent point cloud segments. It defines a parameter for the accuracy and quality of each matching and includes the sub-results of pairs in the final displacement estimate by weighting these accuracy parameters. In this way, the algorithm yields significantly improved accuracy values, which in all cases are of comparable magnitude to those of methods in the literature, and for some published test sets exceed their characteristics. Since parallel programming methods are used to run the fits to previous states, the operation runtime remains fast. If the robot returns to its previous location, the improved loop closure detection method detects this fact, refines the estimates, and improves the accuracy. Finally, the proposed system also produces a 3D point cloud of the environment.

2024-01-01T00:00:00Z Kenzhetayev, Yernar Nagy, István http://hdl.handle.net/20.500.14044/32924 2025-09-18T08:38:27Z 2024-01-01T00:00:00Z

A Real-time, Remotely Operated Vehicle, using a Raspberry Pi and a Virtual Reality Headset Kenzhetayev, Yernar; Nagy, István The usage of modern information and communication technologies, such as virtual and mixed reality, offers new options for controlling and monitoring IoT devices. For example, Head Mounted Displays (HMDs) are gaining popularity as a tool to enhance user productivity and enjoyment. This development is also related to the recent advancements in computer technology and the decline in the price of that technology: HMDs are now more functional while also being more widely available on the market. This paper presents a two- wheel robot car that can be controlled remotely in real-time using HMD. The remote control is done in Virtual Reality with the help of Unity 3D. The open-source game engine decreases cost and development time. There are separate objects for the steering wheel, transmission, screen, and stop button. Both controllers and the user’s hands can be used as input manipulators. The Oculus headset's external cameras use hand recognition to implement this feature. The Raspberry Pi 4 has three main functions: first is to control DC motors with GPIO pins, second is to send video stream from the camera to HMD and third is to accept control signals from HMD and perform them. The data transfer of the Virtual Reality headset and Remotely Operated Vehicle (ROV) is done through server-client communication. Raspberry plays the role of a server, which is written on the Flask framework of the python programming language. This server works using asynchronous principles and the OpenCV library for working with images. GPIO pins are controlled by the server, and it receives requests as well. VR headset is a client, which is written in C# on the Unity game engine. The device interacts with the server when the user does any action and transfers the video stream to the screen in real-time. The configuration of input systems is done with the help of the official Oculus Software Development Kit. The convenience of new input systems and their inherent advantages and disadvantages are discussed. Full-scale tests and findings on whether the suggested approach is practical for actual offshore activities are described as well. The results reveal that it is easier and less expensive to modify the input layout while maintaining the same message-sending technology without the limitations of a physical control panel for the ROV operator.

2024-01-01T00:00:00Z