W niniejszej pracy zbadano wpływ gazu Browna (HHO), gazowej mieszaniny wodoru i tlenu wytwarzanej w procesie elektrolizy wody, na zużycie oleju napędowego w małym silniku o zapłonie samoczynnym. Badania eksperymentalne przeprowadzono na stacjonarnym agregacie prądotwórczym pracującym w różnych warunkach obciążenia elektrycznego, z różnymi prędkościami przepływu gazu HHO dostarczanego do układu dolotowego silnika. Zużycie paliwa mierzono grawimetrycznie i analizowano w odniesieniu do obciążenia silnika oraz stosunku HHO do oleju napędowego. Wyniki pokazują, że dodatek HHO może zmniejszyć zużycie oleju napędowego, szczególnie przy wyższych obciążeniach silnika, z zaobserwowaną redukcją do około 4% w optymalnych warunkach. Jednakże zapotrzebowanie na energię elektrolizera potrzebnego do wytwarzania HHO znacznie przekracza uzyskane oszczędności paliwa. Można stwierdzić, że pomimo mierzalnej poprawy sprawności spalania, stosowanie systemu HHO zasilanego z instalacji elektrycznej pojazdu jest nieuzasadnione energetycznie i ekonomicznie.
Despite a long-term decline in the total number of road accidents in Poland, pedestrian safety remains a serious public health concern. Pedestrians continue to account for a substantial share of fatalities and serious injuries, with careless entry onto the roadway and illegal crossing being the dominant behavioral causes. This study investigates regional differences in pedestrian-caused road accidents across all 16 Polish provinces between 2018 and 2024. The analysis is based on official police accident statistics and applies descriptive statistics, normalized safety indicators (accidents per population and fatalities per accident), and correlation analysis to assess relationships between accident causes and their consequences. The results show strong regional differentiation. Highly urbanized provinces such as Mazowieckie and Śląskie record the highest absolute number of pedestrian accidents, while eastern provinces, including Lubelskie and Podkarpackie, exhibit significantly higher fatality rates per accident despite lower incident counts. Behaviors such as lying or sitting on the roadway are associated with the highest accident-to-incident ratios (up to 75%) and the highest mortality levels, reaching 50% of accidents. The findings demonstrate that accident frequency alone does not adequately reflect pedestrian safety and highlight the need for region-specific preventive measures. The results may support evidence-based regional road safety policies aimed at reducing pedestrian fatalities.
Wastewater is contaminated liquids discharged from homes and industrial plants. A wastewater treatment plant is a complex of basic technological facilities directly used for wastewater treatment, along with auxiliary facilities located on a shared site, necessary for supplying electricity and water, creating appropriate operating conditions, and maintaining the treatment plant. Brewing industries produce wastewater with high organic loads, characterized by high values of parameters such as BZT5 and ChZT. For this reason, these plants typically have their own wastewater treatment systems or cooperate with local municipal wastewater treatment plants. There is growing interest in energy recovery in this industry, for example, through the use of biogas generated during treatment processes, which contributes to cost reduction and a reduction in pollutant emissions. However, the variable composition and load of wastewater remain a challenge, as does the need to properly combine mechanical, biological, and chemical methods to reduce waste volume and its environmental impact. Preparations containing surfactant compounds are used to clean these types of industrial installations. The work was supplemented with an analysis of the content of anionic and nonionic surfactants. Their levels in wastewater—both raw and treated—can be determined using dedicated cuvette tests and the MBAS (for anionic surfactants) and BiAS-thio (for nonionic) methods. The assessment of treatment plant efficiency was based on the pollution reduction index, comparing the content of specific parameters before and after treatment, and the permissible pollutant values in treated wastewater discharged to surface waters.
The shipbuilding industry in Poland, Europe and the United States has been struggling in the 21st century with asymmetric competition from Asia. For several renowned shipyards – including those in Gdansk – identifying pathways for development and creating innovative solutions has become essential for survival. This article presents the dilemmas and research outcome that form part of the preparation for a project aimed at developing an innovative solution grounded in nature-based technologies supported by artificial intelligence. The project on efficient water management for planned production processes in the shipyard constitutes one component of the broader EcoTech initiative, which addresses a wider set of developmental needs of the shipbuilding sector, particularly in Gdansk. This article tackles the challenge of holistically identifying the wide spectrum of developmental conditions as well as the methodological responsibility guiding research and development activities intended to generate breakthrough solutions with business significance.
We analyze the formation of the boundary layer and point to its affinity for diffusion. Based on Prandl'swork, we discuss the properties of the boundary layer and its transition to the vortex state. We also describe theways to reduce the unfavorable effects of the boundary layer encountered in nature.
The article presents an analysis of the influence of geometric parameters of a porous medium represented in a binary form – grid size, obstacle width, and porosity – on the tortuosity of transport paths and the efficiency of pathfinding algorithms. Numerical simulations were carried out for grids of 100-200 nodes, obstacle widths of 1-13 nodes, and porosity values ranging from 0.9 to 0.5, using the Dijkstra, A*, BFS, and Greedy BFS algorithms. The results confirmed the existence of a percolation threshold at ϕ ≈ 0.6 and showed that decreasing porosity increases path tortuosity. For high porosity (ϕ = 0.9), paths were nearly straight (τ ≈ 1.03), while for low porosity (ϕ = 0.5-0.6) they became highly tortuous (τ > 1.3). Among the tested methods, the A* and Greedy BFS algorithms proved to be the most computationally efficient, confirming the effectiveness of heuristic approaches in modeling transport phenomena in porous structures.
The ongoing development of hardware solutions enabling more efficient signal processing is influencing the development of capabilities for reasoning about the state of technical objects. At the same time, solutions for processing data in close proximity to the technical object – edge computing – are becoming more widespread. In an era of new possibilities and expectations regarding signal processing, it is necessary to select a computing unit and a computational algorithm implementation tailored to the process being carried out. This article examines the performance of various implementations of the commonly used FFT algorithm in terms of execution time and energy consumption on a selected computing unit – the ADSP-SC589 DSP controller. The study examined implementations using the SIMD architecture of the computing unit, as well as implementations using a hardware calculation accelerator built into the device structure. The results obtained indicate significant differences in performance between the various implementations.
Automatic speech recognition systems rely on statistical or neural models capable of modelling temporal dependencies present in acoustic signals. Among classical approaches, Hidden Markov Models (HMM) remain an important component of many speech recognition systems, particularly in tasks involving limited datasets or domain-specific vocabularies. One of the key design decisions in HMM-based systems concerns the representation of phonetic context and the number of states used to model acoustic sequences. This study investigates the impact of different phonetic representations and state allocation strategies in HMM models for the task of isolated Polish word recognition. The analysis considers three types of phonetic decomposition: phonemes, diphones and triphones. Additionally, three strategies of assigning the number of hidden states are evaluated: a constant number of states for all models, a dynamically adjusted number of states depending on the number of phonetic units in a word, and the classical speech recognition topology assuming three states per phonetic unit. Experiments were conducted on a custom dataset consisting of 3,600 recordings of 20 Polish command words spoken by nine speakers. Acoustic features were represented using MFCC coefficients and modelled with Gaussian Mixture Hidden Markov Models trained using the Baum–Welch algorithm. The obtained results indicate that dynamically assigning the number of states proportional to the number of phonemes (three states per phoneme) achieves the highest recognition accuracy. At the same time, increasing the phonetic context from phonemes to diphones and triphones did not improve performance on the analysed dataset, likely due to the increased model complexity and the limited size of the training corpus. The analysis of confusion matrices further reveals that HMM models capture phonetic similarities between words, which can lead to systematic recognition errors in phonetically similar commands.
In this article, damage to 2D/3D truss structures is identified using the Coati Optimization Algorithm (COA), a metaheuristic technique. The fundamental idea of COA was to imitate two of the coatis' natural behaviors: (i) hunting and attacking iguanas, and (ii) escaping predators. Moreover, by using natural frequencies from finite element analysis for truss structure to set up the objective function, the final results prove the effectiveness of this algorithm in solving real-world structures.
Nowadays, observations from the Global Navigation Satellite System (GNSS) are one of the most valuable datasets used for ionosphere remote sensing. One of the fields of such studies is to detect and describe the different-scale ionospheric effects, including the main ionospheric trough. This phenomenon is recognised as a large-scale depletion in plasma density with sharp gradients at the edges, observed at the boundary between the high- and middle-latitude ionosphere. Due to the connection between the trough and auroral oval, it exhibits a high dependence on the geomagnetic activity. This work presents a cross-evaluation of ionospheric trough detection using various GNSS-based methods. The assessment utilises multi-station geometry-free (GF) linear combination (LC) GNSS data, converted to vertical directions, and global ionospheric maps. In the former case, data from several dozen stations located in the Northern Hemisphere are used to provide a spatial view of the analysed ionospheric phenomenon. The study focuses on the patterns of trough during high geomagnetic activity that occurred in March 2012. The results confirm that the network-derived GF LC GNSS time series, scaled to a vertical ionospheric path, can be successfully used for ionospheric trough monitoring. Such datasets provide complex signatures of trough during two selected cases corresponding to different phases of the storm. In contrast, the results derived from the global ionospheric maps suffer from generalisation, the level of which depends on the generating algorithm. The comparison of such products provided by UPC and ESA reveals the outperformance of the former, characterised by RMS values at the level 1.7-1.8 TECu. In contrast, the patterns of ionospheric trough in the ESA product are significantly deteriorated, even to 4.4 TECu RMS.
Decision–action systems deployed in high-stakes domains require auditability and bounded (profiled) verifiabil-ity of the reasoning core, which cannot be ensured by empirical safeguards alone. We introduce AGL (Actionable Granular Logic) as a formal framework that combines an auditable knowledge-state layer (MT-FOGL) with work-flows described by a regular-program syntax in the style of FO-PDL. Vague and probabilistic assessments are encapsulated as Information Granules and exposed to the core solely via threshold atoms, thereby keeping the rule/procedure interface within classical two-valued logic. To control verifiability, we restrict quantification and program tests to guarded profiles GF/RGF, preserving decidability with known worst-case complexity bounds. Complex estimation mechanisms are deliberately kept outside the verifiable core (the Decidability Split). The approach is illustrated using non-normative decision patterns in a medical context, intentionally independent of any particular clinical guideline version.
Large-scale deployment of AI in oncology is constrained less by standalone algorithmic performance than by system-level safety, accountability, interoperability, and regulation-aware governance. Grounded in approximately one year of practical pre-deployment work within the OnkoBot project, this paper specifies a deployment- and governance-first reference model for integrated oncology AI platforms under the EU AI Act and the Medical Device Regulation (MDR). The paper introduces Architecture for Medical AI Collaboration (AMAC), an implementation-neutral, system-level envelope that enforces strict online/offline separation between clinical operation and model/knowledge learning and evolution, gate-controlled releases via a Clinical Governance Gateway (CGG) with explicit human-in-the-loop (HITL) escalation, and tamper-evident auditability across clinical, technical, and interoperability boundaries. AMAC is anchored by the Community of Collaborative Evolving Medical Assistants (CEMA), a supervised multi-agent computational core that performs coordinated clinical reasoning under bounded autonomy. Concrete deliverables include: (i) a reference architecture outline with explicit responsibilities and auditable control points; (ii) a phase-gated deployment pathway (Preparation → Prototype → Pilot → Integration → AMAC operation) with required evidence packs, decision gates, and rollback/suspension mechanisms; and (iii) enforceable socio-technical gate criteria, including Socio-Technical Readiness Levels (STRL), readiness metrics, and accountability mapping (RACI). The model is intentionally non-normative and does not encode clinical guidelines; it provides a minimal, auditable governance architecture designed to make large-scale clinical AI integration feasible, controllable, and regulation-compatible in complex oncology environments.
The objective of this study was to determine the statistical relationship among density, volume, mass, and selected geometric parameters of rice grains. A gas pycnometer was employed for grain volume measurement, while a flatbed scanner and specialized software were utilized for the determination of geometric features. Over 70 geometric parameters were identified. Among these, for whole grains, the most effective shape-describing coefficients were Rb, W5, Nv, and LminE, whereas for broken grains, Lsz, LminE, Maver, and Uw proved to be superior. Correlation coefficients between density and geometric features ranged from 0.895 to 0.995 at a significance level of p<0.007. Based on these findings, it will be feasible to develop a grain quality assessment system utilizing 2D images and to infer the mass fraction of grains belonging to different quality grades
The research presented in this article represents a further stage in studies on the strength of components printed using 3D printing technology, specifically FDM (Fused Deposition Modelling). The article presents the results of tensile strength tests on samples printed from PA12 and PA12+CF15 materials, while previous studies by the author focused on PLA material. Basic material data provided by manufacturers and distributors of materials used in the FDM method, such as tensile strength and Young’s modulus, refer to the most favourable model orientation during printing. However, in additive technologies, particularly FDM, the constructed object shows significant layering differences (in the Z direction). The direction of material deposition (in the XY plane) is also crucial. Additionally, the strength is influenced by the degree and type of infill within the model and the temperature during printing. For these reasons, it is essential to understand the relationship between technological parameters and the resulting strength for specific materials. This study aimed to determine the tensile strength of samples printed with varying infill percentages. In the context of the new material, PA12+CF15, it is essential to understand how the addition of carbon fibers affects the mechanical properties of prints compared to traditional materials, such as PA12 and PLA. Carbon fibers can significantly increase the strength and stiffness of the composite, potentially leading to applications in producing parts with high strength requirements. Therefore, studying the strength of materials concerning various printing parameters is crucial for developing the potential of FDM technology and its industrial applications. PA12+CF15 is composed of polyamide 12 (PA12), a thermoplastic material with good chemical resistance, abrasion resistance, and flexibility. The addition of 15% carbon fibers (CF15) reinforces the composite structure, leading to increased stiffness, mechanical strength, and deformation resistance. The study shows that this addition enhances PA12’s strength by approximately 13%, also facilitating printing by reducing shrinkage.
Every year a very large number of people die on the roads. From year to year the value decreases, but it is still a very large number. The purpose of this article is to forecast the number of road accidents in Poland. The study was divided into two parts. The first was the analysis of annual data from police statistics on the number of road accidents in Poland in 2000-2021, and on this basis the forecast of the number of road accidents for 2022-2031 was determined. The second part of the study, dealt with monthly data from 2000-2021. Again, the analyzed forecast for the period January 2022 – December 2023 was determined.
The paper concerns the most significant and characteristic stages of manufacturing and testing the properties of aircraft thin-wall composite structures. Structures of this kind, due to the high requirements in terms of safety, durability and economy of aircraft operation, force the emergence of a number of often mutually contradictory design assumptions. The basic problem is to ensure the necessary strength and stiffness of the structure at the lowest possible weight. In a typical thin-wall structure, due to the small thickness of the skin, it is the covering that buckles, while the frame elements do not lose stability. Therefore, when testing thin-wall structures, it is extremely important to properly prepare the model so that the loss of stability is only local. The choice of stiffness of individual elements and the adopted technological process, which directly determines the properties of the system, are of fundamental importance. The text presents an exemplary solution of this problem, concerning the concept of research model with the given technological process. The model underwent the loading condition, which corresponds to the real conditions occurring during the flight. In the next stage, the characteristic properties of the composite thin-wall structure, which is a representative part of the aircraft, were recorded. The obtained results make it possible to determine the influence of the adopted solution on the character of the skin deformation and provide a basis for modifications and comparative analyses.
Despite a general decline in recent years, road traffic accidents remain a significant public safety concern in both Poland and Montenegro. Although accident rates were affected by the COVID-19 pandemic, the persistent frequency of such incidents underscores the urgent need for further preventive measures to enhance road safety. The aim of this study is to forecast the number of road traffic accidents in Poland and Montenegro for the period 2024-2030. To achieve this, historical data on annual accident counts were obtained from Monstat (Montenegro) and the Polish Police. These datasets were then analyzed using selected neural network models to generate projections for the specified timeframe. The results suggest a potential stabilization in the number of traffic accidents in the near future. This forecast is influenced by several factors, including the steady increase in car ownership and ongoing investments in road infrastructure, such as the construction of new motorways and local roads. It should be noted, however, that the inherent uncertainty in data sampling – used for training, testing, and validating the models – places natural limitations on the precision of the forecasts.
Bioethanol is one of the most important liquid biofuels and is capable of significantly reducing fossil fuel consumption and greenhouse gas emissions. A wide range of raw materials are used for its production. First- and fourth-generation bioethanol is distinguished. The ethanol production process can be carried out using biological or synthetic technologies. Fermentation allows the production of ethanol from renewable raw materials, while synthetic production allows for a high-purity product, but requires the use of petrochemical raw materials. Process optimization includes, among other things, modernizing process water recovery systems, using biological methods involving algae, and integrating bioethanol production with other energy processes. Life-Cycle Assessment (LCA) indicates that greenhouse gas emissions from field fertilization and the high water consumption of the entire process remain a significant environmental issue. The use of bioethanol as a transport fuel additive is supported by European Union policy, while the first-generation bioethanol market is successfully developing in Brazil and its production is currently the cheapest. Bioethanol, especially second generation, is an important element of energy transformation, but its economic competitiveness requires further technological innovation and regulatory support.
The aim of the study was to investigate the effect of vertical displacement of the sample on the results of X-ray diffraction (XRD) in Bragg-Brentano geometry. Measurements were performed on an S275JR steel sample using a Phaser D2 diffractometer (Cu Kα, λ = 1.541874 [Å]) with a step size of 2θ = 0.01°. The shifts in the positions of the 2θ peaks and half-widths (FWHM) were analyzed. A comparison between the theoretical model and experimental peak shifts has been calculated. The lattice constant was determined using the Nelson-Riley method, the crystallite size using the Scherrer method, and the parameters using the Williamson-Hall method. A vertical displacement of 1 mm produced an approximately 0.8° shift of the (110) peak. Based on the diffraction data, the lattice parameter was determined using the Nelson-Riley extrapolation method (2.8643-2.8678 [Å]), the crystallite size was evaluated using the Scherrer method (110-260 [Å], with the largest value for the (110) peak), and lattice distortions were assessed using the Williamson–Hall approach (approx. 0.26-0.30[%]). The results highlight the significance of precise sample positioning, as even a small displacement can lead to noticeable errors in peak locations and consequently in the derived structural parameters.
The paper presents the design of a mobile air quality monitoring station. Its mobility is defined by a lightweight and compact structure, allowing easy transportation and use without additional equipment. To enhance user convenience, a system for storing and analysing the collected data was implemented on an internet server, connected to a web application. The device includes a PM1.0/PM2.5/PM10 sensor, a barometer and hygrometer both with built-in temperature sensors, and two gas sensors. A GPS module provides precise location data, and a GSM module enables wireless transmission of results. The system is powered by Li-Ion batteries, with extended operation time thanks to a photovoltaic panel. The system was tested by comparing its measurements with air quality data from airly.eu, ensuring time and location consistency. The comparison confirmed the device’s accuracy while further tests in mobile mode, i.e., walking, cycling, and driving showed that the system’s mobility increases the number of measurement points, enhancing local air pollution monitoring. Mobile stations could be beneficial, especially in areas with a low density of monitoring stations, e.g. by using vehicles on fixed routes, such as public transport. This would allow effective air quality control and localization of point-source pollution.
