Tadeusz Kościuszko University of Technology (Polish: Politechnika Krakowska im. Tadeusza Kościuszki) is a public university located in central Kraków, Poland, established in 1946 and, as an institution of higher learning granted full autonomy in 1954.Over 37,000 students graduated from the Polytechnic to this day with degrees. Doctorate degrees were granted to 1200 persons and Habilitated degrees – to additional 300. The number of students admitted each year reaches 4500.
This paper examines the geometric principles which governs the construction of late Gothic vaults, by using the example of the vault in the chancel of St. Zygmunt’s Church in Szydłowiec (Poland). Scientific studies concerning the Szydłowiec temple and similar European construction were analyzed. On this basis, the geometrical rules for the construction of the vault, which is unique in Poland, were established. The initial figure was a regular octagon, the side of which determined the length of the spans. The radius of the two-bend prinzipalbogen, inscribed in the octagon, defines the curvature of all ribs. The research used measurements of the vault and the original plan from the time of construction. As a result of the analysis, the use of a two-bend prinzipalbogen was confirmed. Axial symmetry, parallel lines, and collinearity of nodal points are other geometric principles that builders followed.
This study addresses inverse non-stationary heat conduction problems involving an initially unknown heat source function, identified using a limited set of temperature measurements within the domain. The proposed method combines Monte Carlo simulations with the random walk technique to stochastically estimate temperatures at arbitrary spatial points, yielding a meshless temperature approximation and collocation framework that incorporates measurement data and known parameters via nodal indication numbers. These results are embedded in an objective function measuring the discrepancy between computed and measured temperatures. Analytical differentiation with respect to source parameters yields explicit expressions, transforming the ill-posed inverse problem into a well-posed direct one in the form of a small system of algebraic equations. The approach is introduced for the one-dimensional case and extended to 2D and 3D geometries using the Moving Least Squares (MLS) approximation, which supports irregular node distributions and complex domains.
Context: Due to limited contextual understanding, traditional Static Application Security Testing (SAST) tools often struggle to detect complex software vulnerabilities, especially in dynamically typed languages. Large Language Models (LLMs), with their advanced contextual comprehension and language-agnostic nature, offer a compelling alternative to conventional security analysis methods. Objectives: This study aims to evaluate the effectiveness of contemporary LLMs as tools for SAST, focusing on their ability to detect eleven common vulnerability classes, including SQL Injection (SQLi), Cross-Site Scripting (XSS), and Path Traversal. Methods: Thirty state-of-the-art LLMs-including Gemini 3 Pro, Sonnet 4.5, and DeepSeek R1 — were compared with a traditional SAST tools like Semgrep. The models were evaluated using the OWASP Benchmark version 1.2 and a custom external validation dataset. The models were assessed using widely accepted metrics such as precision, recall, and F1-score. Results: Several LLMs outperformed Semgrep in terms of the F1-score. While Semgrep achieved an F1-score of 0.66, the highest score, 0.88, was obtained by the Gemini 3 Pro model. Evaluation on the custom external dataset confirmed these findings, validating the models’ superior performance over traditional methods and their ability to generalize to previously unseen code. Conclusion: The findings demonstrate the potential of LLMs in static code analysis and suggest that, with further development, LLM-based approaches surpass traditional tools and potentially even manual review in vulnerability detection effectiveness.
The pollution of water, including salt and fresh water, has become an emergency problem. Pollutants come from different sources and have various characteristics, starting from industry and fertilizers used in agriculture, sewage related to human living, and other sources. Diverse sources of pollution require a comprehensive approach to water purification. One possible approach may be the use of appropriate sorbents. Currently, one of the most promising materials used is zeolites. This is because they can come from various sources, including waste raw materials such as fly ash, and, therefore, allow for the use of a circular economy approach. Moreover, these materials can be modified, which enables their selective use for selected types of pollutants. Eventually, these materials become economically viable options. The main aim of this article is to present and analyze possible solutions to water pollution based on zeolite materials. For this purpose, a critical literature review was prepared. The review reveals that zeolites perform particularly well in ion-exchange-driven removal of inorganic contaminants, while their effectiveness for organic micropollutants under realistic conditions is often limited. The identified trade-offs between removal efficiency, regeneration stability, and scalability indicate that zeolites are best applied as function-specific rather than universal sorbents. From a sustainability perspective, this targeted applicability is supported by advantages, such as low material cost, long service life, and the possibility of using naturally occurring or waste-derived precursors, which, together, enable resource-efficient water treatment processes, reduced reliance on energy-intensive technologies, and the valorization of industrial byproducts within circular economy frameworks.
Magnetorheological (MR) dampers are semi-active devices that take advantage of the controllable properties of MR fluids. The technology has been most utilized in vibration suppression applications: real-time semi-active controlled automotive chassis systems. Upon exposing the fluid in the control valve to magnetic fields of sufficient magnitude, the dampers generate an additional force. The force generated by such dampers is a complex function of the input velocity or displacement and the current (or magnetic flux), not to mention the fluid’s rheological properties and the solenoid’s performance; the device by design is multidisciplinary. Therefore, developing an accurate and capable MR damper model has been proved to be a challenge. In the paper, the authors present the results of a parameter identification study based on a modified Bouc-Wen model of damper hysteresis applied to three distinct MR damper configurations with realistic force envelopes and tuning range. The obtained results show that the updated model copies the complex relationships of the damper force against the displacement, velocity, and current very well regardless of the input displacement (velocity) frequency or the current magnitude. The comparison is revealed for all three tested damper units in the form of force-displacement, force-velocity loops, respectively, or static force-velocity plots. The study is complemented using the developed HiL (Hardware-in-the-Loop) control system in order to demonstrate the proposed model application for control system studies in a vehicle suspension.
