1. Transverse compressive behavior of plain and polypropylene fiber-reinforced concrete infilled steel tubes as hybrid steel-concrete nodes for discrete reticulated shell structuresŽiga Unuk, 2025, izvirni znanstveni članek Opis: This study investigates using concrete infilled steel tubes as hybrid nodes for discrete reticulated shell structures under compressive loading transverse to the tube axis. Three configurations were tested: empty, plain concrete infilled and polypropylene fiber-reinforced concrete infilled steel tubes. Displacement-controlled loading was applied through welded steel plates. Simplified beam theory models and nonlinear finite element analysis were used to evaluate the structural behavior. The concrete infilled steel tube specimens had higher stiffness and load-bearing capacity than the empty ones. The structural behavior of the concrete infilled specimens could be split into pre- and post-split-tensile cracking phases. No significant difference was found between the plain and fiber-reinforced concrete specimens under loads up to 1500 kN, which was the load limit of the testing machine. An additional fiber-reinforced concrete specimen, tested on a higher capacity machine, had a load-bearing capacity of 2400 kN and exhibited ductile failure.
Ključne besede: concrete infilled steel tube, fiber-reinforced concrete, compressive loading, discrete reticulated shell structure node, simplified calculation, nonlinear finite element analysis
Objavljeno v DKUM: 19.01.2026; Ogledov: 0; Prenosov: 0
 Celotno besedilo (13,83 MB)
Gradivo ima več datotek! Več... |
2. Controlling the mechanical response of stochastic lattice structures utilizing a design model based on predefined topologic and geometric routinesInga Krešić, Jasmin Kaljun, Nebojša Rašović, 2024, izvirni znanstveni članek Opis: The paper explores enhancing the mechanical behavior of stochastic lattice structures through a semi-controlled design approach. By leveraging the Gibson-Ashby model and predefined geometric routines, the study aims to optimize the mechanical response of lattice structures under compressive stress. Transitioning from stochastic to semi-controlled tessellation using Rhinoceros 7 software enables more predictable deformation behavior. Design parameters such as node formation, strut thickness, and lattice generation patterns are correlated with relative density to regulate stiffness and strength. Experimental validation using Acrylic Styrene Acrylonitrile (ASA) filament demonstrates the effectiveness of the proposed design model. The research emphasizes the importance of understanding internal mechanics by introducing a novel design approach to control geometry and topology arrangement in shaping lattice properties. By introducing a semi-controlled mechanism, the study seeks to improve the reliability and uniformity of mechanical responses in lattice structures. The findings highlighted the benefits of semi-controlled design approaches in achieving tailored mechanical properties. Specimens were compression tested in quasi-static uniaxial loading and showed that structures created with parabolic distribution dimensioned by hp=0.5hv originated the most reliable and most vital mechanical response compared with other design models, including typical Voronoi distribution. The improved mechanical response in between proposed design models constantly progressed by about 15% on average consecutively, starting from the parabolic distribution dimensioned by hp=1.0hv as the weakest ranked, up to the best one, dimensioned by hp=0.5hv , even better than the typical Voronoi distribution. The proposed design model has introduced an entirely novel approach that significantly enhances the product’s volume tessellation using routines that guarantee the validity of geometric and topologic entities. Uniaxial compression tests on lattice blocks highlighted the effect of the proposed approach on the mechanical properties of these structures, having shown particularly crucial repeatability and stability. Overall, the paper contributes to advancing the field of lightweight lattice structures through the novel design methodology and material characterization.
Ključne besede: Voronoi tessellation, lightweight design, node connectivity, mechanical response
Objavljeno v DKUM: 15.01.2026; Ogledov: 0; Prenosov: 0
Celotno besedilo (5,86 MB)
Gradivo ima več datotek! Več... |
3. Uporaba ogrodja Fastify in njegovih vtičnikov : diplomsko deloNik Damiš, 2025, diplomsko delo Opis: Fastify je lahkotno in zmogljivo Node.js ogrodje za razvoj strežniških aplikacij, ki združuje zmogljivo jedro z robustnim sistemom vtičnikov. Zaradi hitrosti, varnosti in enostavne razširljivosti predstavlja konkurenčno alternativo obsežnejšim ogrodjem. Namen diplomske naloge je podrobno predstaviti arhitekturo Fastifyja, vključno s sistemom vtičnikov, hierarhijo kontekstov, z življenjskim ciklom zahtev ter razložiti ključne koncepte, kot so dekoratorji, kljuke in JSON-sheme. Raziskava primerja Fastify z drugimi Node.js ogrodji glede na hitrost in odzivnost, kjer se najbolje izkažejo lahkotna ogrodja. Praktični del vključuje razvoj modularne aplikacije s avtentikacijo, avtorizacijo, podatkovno bazo, nalaganjem datotek in s Swagger dokumentacijo.
Ključne besede: Fastify, Node.js, Ogrodja, JavaScript
Objavljeno v DKUM: 23.09.2025; Ogledov: 0; Prenosov: 11
Celotno besedilo (1,51 MB) |
4. Razvoj platforme za upravljanje storitev avtomehanikovAnže Fric, 2025, diplomsko delo Opis: V diplomskem delu smo razvili platformo, ki poenostavlja usklajevanje terminov med strankami in avtomehaniki. Platformo sestavljajo mobilna aplikacija in zaledni del s serverjem in bazo. Uporablja se uveljavljen sistem avtentikacije, kar zagotavlja varno prijavo uporabnikov. Pred rezervacijo termina lahko uporabniki pregledajo ključne informacije o avtomehaniku, kot so kontaktni podatki, delovni čas in okvirne cene storitev. Dodatno imajo avtomehaniki možnost samostojnega vnosa podatkov o opravljenih servisih, kar zagotavlja celovito servisno zgodovino vozil. Vse skupaj je povezano s potisnimi sporočili, ki se pošiljajo po operacijah s termini. V zaključku dela se predstavijo še rezultati testiranja zmogljivosti.
Ključne besede: avtomehanik, mobilna aplikacija, React Native, Node.js, Supabase
Objavljeno v DKUM: 04.09.2025; Ogledov: 0; Prenosov: 26
Celotno besedilo (4,40 MB) |
5. Nadgradnja ogrevalnega sistema z internetom stvariBranko Zebec, 2024, magistrsko delo Opis: Magistrsko delo vsebuje analizo in razvoj prototipa za nadgradnjo ogrevalnega sistema, pri katerem smo sledili dvema ciljema: povezati ogrevalni sistem v internetno omrežje s pomočjo tehnologije interneta stvari in optimizirati ogrevalni sistema. Za dosego ciljev smo raziskali vse potrebne tehnologije, programska orodja in strojno opremo. Pri implementaciji smo uporabili programsko platformo Home Assistant, s katero smo razvili glavni krmilnik na strojni opremi Raspberry Pi 4. V okviru platforme Home Assistant smo integrirali dodatna programska orodja: Cloudflare za vzpostavitev varne povezave na oblačni strežnik in s tem ustvarjanje varnega tunela do glavnega krmilnika, Node-RED, s katerim smo razvili programsko kodo za povezavo z napravami Modbus in optimizacijo sistema vključno s PID krmilnikom, InfluxDB za bazo podatkov vseh spremenljivk našega ogrevalnega sistema, Grafano za vizualizacijo podatkov iz baze, Google SDK za integracijo Googlovega pomočnika na Home Assistant. Na krmilnik smo povezali naprave prek protokola Modbus TCP/IP (angl. Transmission Control Protocol/Internet Protocol): Daikin krmilnik za dostop do toplotne črpalke Daikin Altherma HT, Siemens sobni termostat. Za spremljanje porabe električne energije smo razvili merilnik porabe na modulu ESP32, ki smo ga povezali na kontakte S0 analognega števca. Programsko kodo C++ smo povzeli in prilagodili ter povezali prek protokola MQTT na Home Assistant. Razvili smo tudi simulacijski model, s pomočjo katerega smo preverjali nastavitve PID krmilnika. S statistično analizo smo preverili, ali so podatki po optimizaciji sistema resnično potrdili pričakovane rezultate glede zmanjšanja porabe električne energije, manjšega vtoka ogrevalne vode ter sprememb sobne temperature. Prototip smo preverjali v celotnem ogrevalnem obdobju, le-ta je deloval zanesljivo brez izpadov. Za oddaljeni dostop smo uporabljali Home Assistant pomočnika in Googlovega pomočnika. Optimizacija ogrevalnega sistema se je izkazala za zelo uspešno, kar smo potrdili z meritvami spremenljivk, predvsem porabe električne energije. Ugotovili smo, da se v prehodnem obdobju ogrevanja poraba znatno zmanjša. Na podlagi tega smo prepričani, da bomo na dolgi rok podaljšali življenjsko dobo toplotne črpalke ter prispevali k trajnostni naravnanosti našega sistema ogrevanja.
Ključne besede: internet stvari (IoT), pametni dom, ogrevanje, Home Assistant, Node-RED, PID.
Objavljeno v DKUM: 22.11.2024; Ogledov: 0; Prenosov: 70
Celotno besedilo (9,03 MB) |
6. Pregled in analiza orodij in ogrodij Node.js za podporo neprekinjene integracijeLuka Šnajder, 2024, diplomsko delo Opis: Neprekinjena integracija (CI) je ključnega pomena za sodoben razvoj programske opreme, saj omogoča pogosto združevanje in testiranje kode. Vendar pa implementacija neprekinjene integracije v različnih okoljih, kot je Node.js, predstavlja izziv zaradi raznolikosti razpoložljivih orodij in praks. Cilj diplomske naloge sta bila pregled in analiza učinkovitosti različnih orodij neprekinjene integracije pri podpori razvoja Node.js aplikacij. Uporabili smo primerjalno analizo na orodjih neprekinjene integracije, kot so Jenkins, GitHub Actions in CircleCI, analizirali izzive teh orodij v podjetjih ter izvedli optimizacijo cevovoda za izboljšanje učinkovitosti. Glavni rezultati raziskave so pokazali prednosti in slabosti posameznih orodij ter izpostavili pomen dobre pokritosti kode in testov za uspešnost procesov neprekinjene integracije.
Ključne besede: Neprekinjena integracija, Node.js, CI/CD orodja, avtomatizacija
Objavljeno v DKUM: 22.10.2024; Ogledov: 0; Prenosov: 40
Celotno besedilo (3,40 MB) |
7. Completely resected stage III melanoma controversy : 15 years of national tertiary centre experienceBarbara Perić, Sara Milićević, Andraž Perhavec, Marko Hočevar, Janez Žgajnar, 2021, izvirni znanstveni članek Opis: Background: Two prospective randomized studies analysing cutaneous melanoma (CM) patients with sentinel lymph node (SLN) metastases and rapid development of systemic adjuvant therapy have changed our approach to stage III CM treatment. The aim of this study was to compare results of retrospective survival analysis of stage III CM patients% treatment from Slovenian national CM register to leading international clinical guidelines.
Patients and methods: Since 2000, all Slovenian CM patients with primary tumour % TIb are treated at the Institute of Oncology Ljubljana and data are prospectively collected into a national CM registry. A retrospective analysis of 2426 sentinel lymph node (SLN) biopsies and 789 lymphadenectomies performed until 2015 was conducted using Kaplan-Meier survival curves and log-rank tests.
Results: Positive SLN was found in 519/2426 (21.4%) of patients and completion dissection (CLND) was performed in 455 patients. The 5-year overall survival (OS) of CLND group was 58% vs. 47% of metachronous metastases group (MLNM) (p = 0.003). The 5-year OS of patients with lymph node (LN) metastases and unknown primary site (UPM) was 45% vs. 21% of patients with synchronous LN metastasis. Patients with SLN tumour burden < 0.3 mm had 5-year OS similar to SLN negative patients (86% vs. 85%; p = 0.926). The 5-year OS of patients with burden > 1.0 mm was similar to the MLNM group (49% vs. 47%; p = 0.280).
Conclusions: Stage III melanoma patients is a heterogeneous group with significant OS differences. CLND after positive SLNB might still remain a method of treatment for selected patients with stage III.
Ključne besede: cutaneous melanoma, surgery treatment, sentinel node biopsy, completion lymph node dissection, overall surviva
Objavljeno v DKUM: 08.10.2024; Ogledov: 0; Prenosov: 8
Celotno besedilo (412,50 KB)
Gradivo ima več datotek! Več... |
8. Zasnova spletne aplikacije za naročanje hraneErik Cvetkov, 2024, diplomsko delo Opis: Diplomska naloga obravnava razvoj aplikacije za naročanje hrane. Pri izdelavi aplikacije smo uporabili različne tehnologije, kot so JavaScript, Node.js in Ubuntu Linux strežnik. Interaktivnost je bila realizirana s pomočjo spletnega vtičnika. Razvita aplikacija vključuje vse ključne funkcionalnosti, ki jih pričakujejo uporabniki, vključno s pregledom ponudbe, upravljanjem košarice, dodajanjem naslova dostave in oddajo naročila. Poleg tega aplikacija omogoča restavracijam sprejem naročil v realnem času preko tehnologije WebSocket. Intuitiven uporabniški vmesnik in funkcionalnost, kot je takojšnje posodabljanje stanja naročil, prispevata k izboljšani uporabniški izkušnji tako za stranke kot za restavracije, kar je bilo ključno vodilo pri razvoju aplikacije.
Ključne besede: dostava hrane, prototipni razvoj, Node.js, JavaScript, WebSocket
Objavljeno v DKUM: 30.08.2024; Ogledov: 114; Prenosov: 54
Celotno besedilo (4,02 MB) |
9. Graph Neural Network-Based Method of Spatiotemporal Land Cover Mapping Using Satellite ImageryDomen Kavran, Domen Mongus, Borut Žalik, Niko Lukač, 2023, izvirni znanstveni članek Ključne besede: multispectral, Sentinel-2, superpixel, node, EfficientNetV2, GraphSAGE
Objavljeno v DKUM: 23.05.2024; Ogledov: 158; Prenosov: 31
Celotno besedilo (34,47 MB)
Gradivo ima več datotek! Več... |
10. Density-based entropy centrality for community detection in complex networksKrista Rizman Žalik, Mitja Žalik, 2023, izvirni znanstveni članek Opis: One of the most important problems in complex networks is the location of nodes that are essential or play a main role in the network. Nodes with main local roles are the centers of real communities. Communities are sets of nodes of complex networks and are densely connected internally. Choosing the right nodes as seeds of the communities is crucial in determining real communities. We propose a new centrality measure named density-based entropy centrality for the local identification of the most important nodes. It measures the entropy of the sum of the sizes of the maximal cliques to which each node and its neighbor nodes belong. The proposed centrality is a local measure for explaining the local influence of each node, which provides an efficient way to locally identify the most important nodes and for community detection because communities are local structures. It can be computed independently for individual vertices, for large networks, and for not well-specified networks. The use of the proposed density-based entropy centrality for community seed selection and community detection outperforms other centrality measures.
Ključne besede: networks, undirected graphs, community detection, node centrality, label propagation
Objavljeno v DKUM: 06.02.2024; Ogledov: 335; Prenosov: 37
Celotno besedilo (707,65 KB)
Gradivo ima več datotek! Več... |
