Publications Repository - Gdańsk University of Technology

Page settings

polski
Publications Repository
Gdańsk University of Technology

Show publications from the year

Publications from the year 2023

Show all
  • Crack Resistance of Asphalt Concrete Subjected to Environmental Factors
    • Cezary Szydłowski
    • Marcin Stienss
    2023 Full text NanoWorld Journal

    The paper presents an analysis of the influence of environmental factors on the cracking susceptibility of asphalt concrete resulting in a change in the durability of asphalt pavement. In order to assess the phenomenon, laboratory tests were carried out taking into account the destructive effects of moisture, freeze-thaw cycle and long-term ageing. The influence of both factors occurring simultaneously was also verified. Due to the different methods of assessing resistance to water and frost, the tests were carried out according to four variants of conditioning the samples: Polish Technical Requirements WT:2-2010 and WT:2-2014, AASTHO (American Association of State Highway and Transportation Officials) T 273, AASHTO TP 140 (MIST). Long-term aging was conducted according to AASHTO R 30 for 120 h and also in extended time to 216 h. Finally, ten variants of conditioning samples were tested. Asphalt concrete for wearing course was investigated. Parameters of mixtures with neat (50/70) and SBS-modified (45/80-55) binders were compared. The evaluation of changes in fracture toughness was carried out based on the results of testing SCB (Semicircular bending) specimens with a 10 mm notch depth at a temperature of +10 °C, at a loading rate of 1 mm/min. Fracture toughness, fracture energy and Flexibility Index were assessed. A group of 40 results for each of the mixtures was obtained, taking into account both the dispersion due to material heterogeneity (four samples for each of the conditioning variants) and variation in material properties caused by environmental factors (10 conditioning variants). The influence of water and frost action causes a decrease in fracture toughness and a reduction in fracture energy. Long-term ageing increases the stiffness of the mixture, thus increasing the fracture toughness. A significant increase in the brittleness of the material is observed, which reduces the deformation at fracture and thus the fracture energy. On the basis of the obtained results, it can be concluded that the destructive effect of environmental factors is clearly more visible in the case of a mixture with a neat binder. The comprehensive laboratory test results allowed for probabilistic analysis and reliability estimation related to the assessment of the durability of asphalt mixtures. It could be concluded that asphalt pavements made with the use of modified bitumen are characterized by significantly higher durability.

  • Credit default swaps and banks
    • Muhammad Mushafiq
    2023

    This chapter aims to explore the evolving role of credit default swaps (CDS) in managing and transferring default risk from the perspective of banks from a holistic perspective. This chapter examines credit default swaps (CDSs) as derivative financial instruments that transfer credit risk on debt securities. While CDSs offer benefits such as risk management and risk trading, they also introduce potential systemic risks, as evidenced by their role in the 2008 financial crisis. The collapse of American International Group (AIG) serves as a case study, highlighting the complex interactions between banks, CDSs, and the broader financial system. AIG’s exposure to mortgage-backed securities through CDSs led to significant losses and threatened the stability of the financial system.

  • Cross-Lingual Knowledge Distillation via Flow-Based Voice Conversion for Robust Polyglot Text-to-Speech
    • Dariusz Piotrowski
    • Renard Korzeniowski
    • Alessio Falai
    • Sebastian Cygert
    • Kamil Pokora
    • Georgi Tinchev
    • Ziyao Zhang
    • Kayoko Yanagisawa
    2023

    In this work, we introduce a framework for cross-lingual speech synthesis, which involves an upstream Voice Conversion (VC) model and a downstream Text-To-Speech (TTS) model. The proposed framework consists of 4 stages. In the first two stages, we use a VC model to convert utterances in the target locale to the voice of the target speaker. In the third stage, the converted data is combined with the linguistic features and durations from recordings in the target language, which are then used to train a single-speaker acoustic model. Finally, the last stage entails the training of a locale-independent vocoder. Our evaluations show that the proposed paradigm outperforms state-of-the-art approaches which are based on training a large multilingual TTS model. In addition, our experiments demonstrate the robustness of our approach with different model architectures, languages, speakers and amounts of data. Moreover, our solution is especially beneficial in low-resource settings.

  • Cross-Linking, Morphology, and Physico-Mechanical Properties of GTR/SBS Blends: Dicumyl Peroxide vs. Sulfur System
    • Agata Rodak
    • Agnieszka Susik
    • Daria Kowalkowska-Zedler
    • Łukasz Zedler
    • Krzysztof Formela
    2023 Full text Materials

    In this work, ground tire rubber and styrene–butadiene block copolymer (GTR/SBS) blends at the ratio of 50/50 wt%, with the application of four different SBS copolymer grades (linear and radial) and two types of cross-linking agent (a sulfur-based system and dicumyl peroxide), were prepared by melt compounding. The rheological and cross-linking behavior, physico-mechanical parameters (i.e., tensile properties, abrasion resistance, hardness, swelling degree, and density), thermal stability, and morphology of the prepared materials were characterized. The results showed that the selected SBS copolymers improved the processability of the GTR/SBS blends without any noticeable effects on their cross-linking behavior—which, in turn, was influenced by the type of cross-linking agent used. On the other hand, it was observed that the tensile strength, elongation at break, and abrasion resistance of the GTR/SBS blends cured with the sulfur system (6.1–8.4 MPa, 184–283%, and 235–303 mm3, respectively) were better than those cross-linked by dicumyl peroxide (4.0–7.8 MPa, 80–165%, and 351–414 mm3, respectively). Furthermore, it was found that the SBS copolymers improved the thermal stability of GTR, while the increasing viscosity of the used SBS copolymer also enhanced the interfacial adhesion between the GTR and SBS copolymers, as confirmed by microstructure evaluation.

  • Crystal Facet Engineering of TiO2 from Theory to Application
    • Szymon Dudziak
    • Marta Kowalkińska
    • Anna Zielińska-Jurek
    2023 Full text

    Recently, the surface structure effect on photocatalytic activity has gathered increasing attention due to its reported influence on the charge carrier trapping and separation. Detailed control over the surface structure can be achieved by exposing the specific crystal facets. As a result, the photogenerated electrons and holes can be effectively separated between the different facets of semiconductor crystals. TiO2 is the most studied photocatalyst, with the particles exposing {0 0 1}, {1 0 0}, {1 0 1}, {1 1 0}, {1 1 1}, and {1 0 5} crystal facets. The performed studies have shown that the efficiency of the photocatalytic process strongly depends on the nature of the crystal facet exposed at the photocatalyst surface. In this regard, this chapter focuses on the comparison of possible surface-related parameters and photocatalytic activity of anatase, rutile, and brookite polymorphs with exposed different crystal facets. Particularly, computational data on their different possible surface structures are summarized, focusing on the geometry, energy, and possible reconstructions. This is followed by the general description of the hypothetical Wulff constructions and existing stabilization/synthesis strategies. Such an approach could help to further design, simulate, and optimize photocatalyst surface for efficient photoreduction and photooxidation processes.

  • Crystalline Silicon (c-Si)-Based Tunnel Oxide Passivated Contact (TOPCon) Solar Cells: A Review
    • Hayat Ullah
    • Stanisław Czapp
    • Seweryn Szultka
    • Hanan Tariq
    • Usama Bin Qasim
    • Hassan Imran
    2023 Full text ENERGIES

    Contact selectivity is a key parameter for enhancing and improving the power conversion efficiency (PCE) of crystalline silicon (c-Si)-based solar cells. Carrier selective contacts (CSC) are the key technology which has the potential to achieve a higher PCE for c-Si-based solar cells closer to their theoretical efficiency limit. A recent and state-of-the-art approach in this domain is the tunnel oxide passivated contact (TOPCon) approach, which is completely different from the existing classical heterojunction solar cells. The main and core element of this contact is the tunnel oxide, and its main role is to cut back the minority carrier recombination at the interface. A state-of-the-art n-type c-Si-based TOPCon solar cell featuring a passivated rear contact was experimentally analyzed, and the highest PCE record of ~25.7% was achieved. It has a high fill factor (FF) of ~83.3%. These reported results prove that the highest efficiency potential is that of the passivated full area rear contact structures and it is more efficient than that of the partial rear contact (PRC) structures. In this paper, a review is presented which considers the key characteristics of TOPCon solar cells, i.e., minority carrier recombination, contact resistance, and surface passivation. Additionally, practical challenges and key issues related to TOPCon solar cells are also highlighted. Finally, the focus turns to the characteristics of TOPCon solar cells, which offer an improved and better understanding of doping layers and tunnel oxide along with their mutual and combined effect on the overall performance of TOPCon solar cells.

  • Cu( II ) ions removal from wastewater using starch nanoparticles ( SNPs ): An Eco‐sustainable approach
    • Vikas S. Hakke
    • Vividha K. Landge
    • Shirish H. Sonawane
    • G. Uday B. Babu
    • Sivakumar Manickam
    • Grzegorz Boczkaj
    2023 Full text CANADIAN JOURNAL OF CHEMICAL ENGINEERING

    The complex structured starch particles were reduced to the nanoscale size range through hydrolysis utilizing low concentration acid assisted with ultrasound irradiation. The synthesized starch nanoparticles (SNPs) were characterized by TEM, FTIR, and XRD techniques. The synthesized SNPs possessed surface activated entities, as many cationic functional groups were confirmed through the FTIR spectrum. Also, these SNPs were effectively utilized to separate heavy Cu metal ions from the synthetic ion solution. The SNPs were characterized using FESEM, XPS and BET for the surface modification after the adsorption process. The weak electrostatic interaction between the SNP surface and Cu ion was confirmed from the XPS spectrum and energy dispersive X-ray spectroscopy. The maximum efficiency of Cu ions removal was about 93% at an optimal pH 5 and 25 mg/mL dosage. The adsorption equilibrium was obtained in 60 min. The nitrogen isotherm BET analysis of SNPs after adsorption shows a higher specific surface area of 18.552 m2/g, attributed to the interaction and presence of Cu ions on the SNP surface. The process feasibility was validated by the Langmuir isotherm model. The process exhibits pseudo-second-order adsorption kinetics and follows the Langmuir isotherm. The RL predicted by the Langmuir isotherm mechanism is 0.017, implying favorable adsorption. The process is reproducible and allows to separate heavy metal ions from the wastewater through bio-sorption effectively.

  • Cu-HKUST-1 and Hydroxyapatite–The Interface of Two Worlds toward the Design of Functional Materials Dedicated to Bone Tissue Regeneration
    • Marzena Fandzloch
    • Weronika Bodylska
    • Joanna Trzcińska-Wencel
    • Patrycja Golińska
    • Katarzyna Roszek
    • Joanna Wiśniewska
    • Michał Bartmański
    • Agnieszka Lewińska
    • Anna Jaromin
    2023 Full text ACS Biomaterials Science & Engineering

    A novel composite based on biocompatible hydroxyapatite (HA) nanoparticles and Cu-HKUST-1 (Cu-HKUST-1@HA) has been prepared following a layer-by-layer strategy. Cu-HKUST-1 was carefully selected from a group of four Cu-based metal−organic frameworks as the material with the most promising antimicrobial activity. The formation of a colloidal Cu- HKUST-1 layer on HA nanoparticles was confirmed by various techniques, e.g., infrared spectroscopy, powder X-ray diffraction, N2 sorption, transmission electron microscopy imaging, electron paramagnetic resonance, and X-ray absorption spectroscopy. Importantly, such a Cu-HKUST-1 layer significantly improved the nanomechanical properties of the composite, with Young’s modulus equal to that of human cortical bone (13.76 GPa). At the same time, Cu-HKUST-1@HA has maintained the negative zeta potential (−16.3 mV in pH 7.4) and revealed biocompatibility toward human dermal fibroblasts up to a concentration of 1000 μg/mL, without inducing ex vivo hemolysis. Chemical stability studies of the composite over 21 days in a buffer-simulated physiological fluid allowed a detailed understanding of the transformations that the Cu-HKUST-1@HA undergoes over time. Finally, it has been confirmed that the Cu-HKUST-1 layer provides antibacterial properties to HA, and the synergism reached in this way makes it promising for bone tissue regeneration.

  • Cyclic behavior of FeCoCrNiMn high entropy alloy coatings produced through cold spray
    • Pasquale Cavaliere
    • Angelo Perrone
    • Alessio Silvello
    • Aleksandra Laska
    • Gianni Blasi
    • Irene G. Cano
    • Behzad Sadeghi
    • S. Nagy
    2023 JOURNAL OF ALLOYS AND COMPOUNDS

    eCoCrNiMn high entropy alloy powders were employed to produce coatings on carbon steel through high pressure cold spray (at 1100 °C and 7 MPa in temperature and pressure respectively). X-ray diffraction of the sprayed material revealed a dense coating with the retention of the original crystallographic structure. Once splatted, particles revealed high flattening ratio with consequent excellent adhesion of the coating to the substrate. Residual stresses and hardness increased through the thickness of the coating. The cyclic behavior of the coatings was evaluated through bending tests. The cyclic bending tests were performed by linearly increase the maximum deformation at a fixed number of loading cycles. Other bending tests were conducted at fixed maximum stroke at the same deformation levels indicated for the previous test up to the coating fracture. For ach 100 cycles the test was stopped in order to observe the coating surface to gain the possibility of evaluating the aspect of the superficial fracture. The fracture surface of the coatings was observed through scanning electron microscopy in order to describe the cracking features related to the different loading conditions imposed to the coatings.

  • Cyfrowy HR. Organizacja w warunkach transformacji technologicznej
    • Małgorzata Sidor-Rządkowska
    • Łukasz Sienkiewicz
    2023

    Publikacja jest pierwszą na rynku pozycją opisującą kompleksowo zagadnienia związane z cyfryzacją HR. Autorzy szczegółowo omawiają poszczególne aspekty procesu wdrażania rozwiązań cyfrowych, a także podkreślają kluczową rolę, jaką dział HR odgrywa w kształtowaniu nowej cyfrowej kultury organizacyjnej w firmie. W monografii omówiono zagadnienia takie jak m.in.: * nowe role HR w procesie digitalizacji firmy, * cyfrowe zarządzanie danymi personalnymi, * analityka predykcyjna i algorytmy w HR, * wpływ kluczowych technologii na zmiany w obszarze pracy. Autorzy poruszają także etyczny aspekt transformacji cyfrowej, wskazując przykładowe dylematy związane z ochroną i wykorzystaniem danych osobowych, przejrzystością procesów czy dobrostanem pracowników.

  • Cykl obrazów: CZŁOWIEK WOBEC| uwspólnienie udział w ogólnopolskiej wystawie zbiorowej EMOTIONAL ECOLOGY / EKOLOGIA EMOCJONALNA. Ekologia, humanizm, etyka. Świadomość historii wzornictwa i nowe ścieżki rozwojowe jako element procesu. | cykl: ‘Design w procesie’
    • Agnieszka Kurkowska
    2023

    „Zależności pomiędzy organizmami i miejscem ich przebywania, są obszarem zainteresowania kreatorek i kreatorów. Czerpią oni impuls płynący z emocji własnych i innych organizmów. W efekcie synergii emocji, powstaje perspektywa przestrzeni naturalnej. Ekologia opiera się na systemie wymiany pomiędzy miejscem przebywania a kreatorami. Relacja ta zachodzi dokładnie pomiędzy jednostką a środowiskiem społecznym i naturalnym. Człowiek współstanowi o niesamowitej relacji emocjonalnej, powstającej w pewnym splocie istnienia organizmu ludzkiego i jego otoczenia. W ten sposób kreuje się wielo-organizm. Jego funkcje, sposoby zachowania, uwarunkowania kulturowe wynikają bezpośrednio z natury, z szacunku do otaczającego świata. Nowy sposób wymiany pomiędzy człowiekiem a środowiskiem proponują twórcy kierujący się tropem emocji. Stąd nie jesteśmy ludźmi sami przez się, ale poprzez istnienie (proces życiowy) w środowisku zewnętrznym, organizowanym przez emocje.

  • Cykl rzeźb pt. 'struktury biomorficzne'
    • Agnieszka Kurkowska
    2023

    Dotykam świata formą. Tworzę ją z marzeń i obrazów przyrody. Jestem architektem, co pozwala mi dotykać wzrokiem. Rysuję, rzeźbię, sytuuję gliniane obiekty i sprawdzam, czy miejsce je akceptuje. Poszukuję odczucia zgodności. Zgodność kierunkuje podejmowane próby. Wyobraźnia pyta o mieszkańców. Formy mogą mieścić domy. Schronienia bezpiecznie osadzone w kaszubskim krajobrazie. Brzmiące lasem, pachnące tradycją. Nie znam ich dalszych losów. Zawieszam myśli pomiędzy tym co wiem, a tym czego się spodziewam. Moje doświadczenia tworzą kształty. Spokój miejsca daje siłę do pracy. Utkane z marzeń, ułożone z doświadczeń, dotykające z nadzieją, formą zapożyczoną, przemyślaną i podarowaną miejscu. Najważniejsze jest dotknięcie, rozciągłe po nieregularności obiektu. Dotknięcie podłoża i powietrza. Tła. Uwolniony zapach i przeczucie. Ulotność, która nadaje sens. „Dotykanie jest zjawiskiem dwustronnym. To świat nas dotyka, gdy my go dotykamy” * Zarys i płaszczyzna styku. Definiują przebieg relacji. Oczekuję od nich potwierdzenia celu poszukiwań. Ten etap procesu niesie obawy niepowodzeń, ale i nadzieję adaptacji. Kto powiedział, że architekt nie może w ten sposób poszukiwać formuły dla domu spasowanego z miejscem? Mniej dosłownie. Bardziej delikatnie. Przyglądając się nieuchwytności i ciesząc słońcem dotykającym gliny z odciskiem mojej skóry. „Dzięki dotykowi nasza obecność objawia się nie w przestrzeni, tylko wobec innej, bliskiej rzeczywistości” * *cytaty: Brach-Czaina Jolanta 2022. Błony umysłu. Dowody na istnienie, Warszawa

  • Cylindrical orifice testing in laminar flow with the orifice diameter ratio β = 0.5
    • Anna Golijanek-Jędrzejczyk
    • Andrzej Mrowiec
    2023 Full text Scientific Reports

    The paper presents the results of an experimental study of a cylindrical orifice with the orifice diameter ratio β = 0.5 and the flow opening length‑to‑diameter ratio L/d = 1, with hydraulic oil flowing in the DN50 measuring channel. The measurements of the values characterising the oil flow were made in the laminar flow regime, for the Reynolds numbers ranging between Re = 100 to 950. Based on the experimental tests, standard flow characteristics were created for four kinematic viscosity values in the range of 13.4–33.3 cSt, for which the average value of the discharge coefficient C in the tested flow rate range of q v < 0.5 dm 3 /s was determined.

  • Cytokine IL6, but not IL-1β, TNF-α and NF-κB is increased in paediatric cancer patients
    • Anna Synakiewicz
    • Anna Stanisławska-Sachadyn
    • Anna Owczarzak
    • Małgorzata Skuza
    • Teresa Stachowicz-Stencel
    2023 Full text Acta Biochimica Polonica

    Cytokines are responsible for maintaining homeostasis as cell growth, differentiation, migration and apoptosis mediators. They play a pivotal role in immune responses to inflammatory reactions. In oncological diseases, the cross-talk between cells of the immunological system and cells of the tumour microenvironment is led by cytokines. Also, the overproduction of cytokines may change the tumour microenvironment and stimulate tumour development and growth. To test whether pro-inflammatory cytokines or associated with them transcription factor levels are changed in a group of 53 paediatric cancer patients, serum levels of IL-1β, IL-6, TNF-α and NF-κB were assessed and compared to measures in 25 healthy controls. Increased levels of IL-6 were found among patients in active oncological treatment (P=0.002) but not among patients whose treatment was completed. Our data suggest that IL6, but not IL-1β, TNF-α and NF-κB, is elevated as a result of the immune response in the microenvironment around the tumour and in blood cancers, among patients who were not infected at the time of blood collection. Thus, IL6 levels might serve as a potential biomarker of oncohematological diseases.

  • Czy chemia wszechświata różni się od chemii na planecie Ziemia?
    • Bartłomiej Cieślik
    2023

    Astronomowie, fizycy i chemicy od lat zadają sobie pytanie na ile nasza planeta jest wyjątkowa. Żyjemy na jednym z nielicznych ciał niebieskich, na którym występuje woda w stanie ciekłym. Na naszej planecie pojawiła się niezwykle szeroka gama prostych i wyjątkowo skomplikowanych związków organicznych. Warto jednak zadać pytanie czy faktycznie ziemia jest tak wyjątkowa pod kątem chemicznym, a jeśli tak to co na to wpływa. Współcześnie naukowcy wiedzą już o wszechświecie na tyle dużo, że można spróbować odpowiedź na to pytanie.

  • Czynniki kształtujące skuteczność outsourcingu w podmiotach leczniczych
    • Piotr Grudowski
    • Mateusz Muchlado
    2023

    Problem badawczy podjęty przez autorów tej książki można więc sformułować następująco: Jakie procesy podlegają outsourcingowi i jakie czynniki decydują o ich skuteczności w szpitalach w Polsce? W konsekwencji celem badawczym przyjętym w niniejszej pracy jest identyfikacja procesów zlecanych na zewnątrz w polskich szpitalach wraz z czynnikami warunkującymi ich skuteczność. Celem utylitarnym jest zaś opracowanie metody umożliwiającej kadrze zarządzającej polskimi szpitalami dobór dostawców oraz ocenę procesów poddanych outsourcingowi. Niniejsza monografia składa się z trzech głównych rozdziałów. W rozdziale 1 omówiono pojęcie outsourcingu oraz wykorzystanie go w zarządzaniu procesami. Opisano też czynniki sprawiające, że organizacje decydują się na zlecanie procesów na zewnątrz, jak również szanse i zagrożenia z tym związane. Przedstawiono charakterystykę podejścia procesowego, a także metody klasyfikacji procesów, których zastosowanie jest niezbędne przy wydzielaniu procesu na zewnątrz. Scharakteryzowano też najczęściej wykorzystywane w praktyce oraz sugerowane w literaturze przedmiotu metody oceny skuteczności procesów. W rozdziale 2 przedstawiono organizacje stanowiące podmiot badań oraz ich charakterystykę w świetle literatury. Zawarto tu również wnioski płynące z przeglądu literatury dotyczącej outsourcingu w podmiotach leczniczych w Polsce i za granicą, które odnoszą się do rodzaju procesów wydzielanych oraz motywów prowadzących do outsourcingu. Opisano uwarunkowania prawne i normatywne wpływające na wybór dostawcy outsourcingowego i ocenę skuteczności procesów poddanych outsourcingowi. W rozdziale 3 przedstawiono zastosowaną przez autorów metodykę badawczą. Zaprezentowano wybrane wyniki badań, które posłużyły do weryfikacji postawionych hipotez, a także do osiągnięcia celu utylitarnego, czyli opracowania metody oceny skuteczności procesów poddanych outsourcingowi. Metoda ta została omówiona w końcowej. części rozdziału. W podsumowaniu przedstawiono ujęte syntetycznie wyniki przeprowadzonych badań, wnioski i rekomendacje, w tym te związane z funkcjonowaniem placówek ochrony zdrowia w warunkach kryzysu i niepewności. Książka adresowana jest przede wszystkim do osób zajmujących się zarządzaniem w sektorze ochrony zdrowia, zarówno praktyków, jak i badaczy tej problematyki. Powinna się też spotkać z zainteresowaniem studentów kierunków związanych z zarządzaniem w usługach publicznych.

  • Data, Information, Knowledge, Wisdom Pyramid Concept Revisited in the Context of Deep Learning
    • Bożena Kostek
    2023

    In this paper, the data, information, knowledge, and wisdom (DIKW) pyramid is revisited in the context of deep learning applied to machine learningbased audio signal processing. A discussion on the DIKW schema is carried out, resulting in a proposal that may supplement the original concept. Parallels between DIWK pertaining to audio processing are presented based on examples of the case studies performed by the author and her collaborators. The studies shown refer to the challenge concerning the notion that classification performed by machine learning (ML) is/or should be better than human-based expertise. Conclusions are also delivered.

  • Data-driven, probabilistic model for attainable speed for ships approaching Gdańsk harbour
    • Jakub Montewka
    • Marcin Życzkowski
    • Filip Zarzycki
    2023

    The growing demand for maritime transportation leads to increased traffic in ports. From this arises the need to observe the consequences of the specific speed ships reach when approaching seaports. However, usually the analyzed cases refer only to the statistical evaluation of the studied phenomenon or to the empirical modelling, ignoring the mutual influence of variables such as ship type, length or weather conditions. In this paper, a different approach to the issue is proposed, which summarizes the most important factors and leads to a probabilistic speed model for manoeuvring ships in the port of Gdańsk. For this purpose, data from the Automatic Identification System were used. This resulted in a dataset with almost 2.5k traffic scenarios. To obtain results from the dataset, three different machine learning algorithms based on Bayesian networks were then applied. The developed models can be used to predict the speed as a function of the given parameters as well as to determine the values of individual parameters for a given speed. In addition, the use of the constructed models allowed the analysis of the strength of mutual influences for two connected nodes or the sensitivity of changes for individual variables. The discussion also raised questions about validation of the algorithms and measures to improve accuracy. The average predictive accuracy of the models of about 75% (depending on the learning algorithm used) achieved at this stage is promising, but further work is expected that can increase the predictive power of the models.

  • Data-Driven Surrogate-Assisted Optimization of Metamaterial-Based Filtenna Using Deep Learning
    • Peyman Mahouti
    • Aysu Belen
    • Ozlem Tari
    • Mehmet Belen
    • Serdal Karahan
    • Sławomir Kozieł
    2023 Full text Electronics

    In this work, a computationally efficient method based on data driven surrogate models is pro-posed for the design optimization procedure of a Frequency Selective Surface (FSS)-based filtering antenna (Filtenna). A Filtenna acts as a as module that simultaneously pre-filters unwanted sig-nals, and enhances the desired signals at the operating frequency. However, due to a typically large number of design variables of FSS unit elements, and their complex interrelations affecting the scattering response, FSS optimization is a challenging task. Herein, a deep-learning-based al-gorithm, Modified-Multi-Layer-Perceptron (M2LP), is developed to render an accurate behavioral model of the unit cell. Subsequently, the M2LP model is applied to optimize FSS elements being parts of the Filtenna under design. The exemplary device operates at 5 GHz to 7 GHz band. The numerical results demonstrate that the presented approach allows for almost 90% reduction of the computational cost of the optimization process as compared to direct EM-driven design. At the same time, physical measurements of the fabricated Filtenna prototype corroborate the relevance of the proposed methodology. One of the important advantages of our technique is that the unit cell model can be re-used to design FSS and Filtenna operating a various operating bands without incurring any extra computational expenses.

  • Dataset Related Experimental Investigation of Chess Position Evaluation Using a Deep Neural Network
    • Dawid Wieczerzak
    • Paweł Czarnul
    2023 Full text

    The idea of training Articial Neural Networks to evaluate chess positions has been widely explored in the last ten years. In this paper we investigated dataset impact on chess position evaluation. We created two datasets with over 1.6 million unique chess positions each. In one of those we also included randomly generated positions resulting from consideration of potentially unpredictable chess moves. Each position was evaluated by the Stockfish engine. Afterwards, we created a multi class evaluation model using Multilayer Perceptron. Solution to the evaluation problem was tested with three different data labeling methods and three different board representations. We show that the accuracy for the model trained for the dataset without randomly generated positions is higher than for the model with such positions, for all data representations and 3, 5 and 11 evaluation classes.