Luleå University of Technology Publications
Not a member yet
39356 research outputs found
Sort by
A comparative life cycle study of window interventions: Impact of building characteristics and local context
No full textImproving building envelope components, such as windows, is a common renovation measure in cold climates to increase energy-efficiency. However, the life cycle (LC) benefit of maintaining or replacing windows is highly uncertain and depends on the building, its context and methodological choices made in LC assessments. This study aimed to expand knowledge on how buildings' characteristics and local context (location-related) influence the preferred window intervention option by investigating the LC climate impact and costs of various window maintenance and replacement scenarios for two typical residential buildings in subarctic Sweden. Using life cycle assessment and life cycle costing, along with sensitivity analyses of seven relevant parameters, the study explored the total climate impact and costs over a 60-year period for each window intervention's production and operational stages. Results indicate that building characteristics are important, as the two buildings show contrary LC results, even within the same local context. Swapping the buildings' location affects initial LC performance, and which sensitivity parameters appear critical. The parameters influence the LC results, but only a few alter the window interventions' rankings, with heating emission factor having most impact. Reducing LC climate impact and costs by replacing windows is not guaranteed, as it depends on building characteristics and local context. Instead, window maintenance may be preferred for its lower material use. The study's integrated LC approach provides insights for harmonizing renovation budgets with climate targets. Adapting renovation measures to each specific case is key for choosing the best option, balancing operational and embodied impacts.Validerad;2025;Nivå 2;2025-06-30 (u2);Full text: CC BY license;Funder: Lulebo;</p
Background orientated and shadowgraphy schlieren monitoring in laser-based additive manufacturing
No full textSchlieren monitoring techniques were applied to Laser Directed Energy Deposition (L-DED) and Laser Cladding (LC) processes to evaluate their performance in capturing thermal dynamics and spatter formation under varying process parameters. Two schlieren setups were compared based on their ability to detect changes in temperature, pressure and material state within the process zone. The first, using a background-oriented design, exhibited higher resolution and sensitivity to local thermal gradients and plume dynamics, while the second, with a broader field of view, demonstrated enhanced stability in monitoring cumulative thermal buildup. Optical flow analysis revealed a strong correlation between line energy and the magnitude of optical turbulence, particularly in regions where vaporization occurred, with a clear plateau observed between 110.6 J/mm and 243.9 J/mm, corresponding to optimal melt pool conditions. Beyond 243.9 J/mm, a significant increase in optical flow was observed, indicating plasma formation and enhanced turbulence. A dome-like schlieren structure consistently formed above the melt pool, expanding with higher energy input, offering insights into the balance between thermal buoyancy and vapor pressure. Additionally, the quadratic relationship between line energy and the schlieren dome volume of enclosed optical flow provided a means to identify energy-efficient and stable process conditions. The findings underscore the potential of schlieren-based monitoring for precise control and optimization of additive manufacturing processes, with implications for improving process stability and minimizing defects like spatter and porosity.Validerad;2025;Nivå 2;2025-07-07 (u2);Full text license: CC BY;This paper has previously been published as a manuscript in a thesis.</p
Advanced Cut-Edge Characterization Methods for Improved Sheared-Edge Damage Evaluation in High-Strength Sheet Steels
No full textThis study investigates shear cutting of high-strength steel sheets, a process known to negatively impact the forming and fatigue properties of the material. The localized deformation near the cut edges imposes sheared-edge damage, especially in advanced high-strength steels where severe shear deformation occurs in the very vicinity of the cut edge. In this work, an extensive experimental investigation was carried out on punched holes of thin sheets, using light optical microscopy and metallographic techniques for sheared-edge damage assessment. These methods provided detailed insights into the sheared-edge damage and offer a thorough understanding of the deformation behavior in the shear-affected zone. Advanced engineering cut-edge investigation methods have been developed based on 2D and 3D stereo light microscopy for non-destructive panoramic cut-edge parameters and cut-edge profile determination along cut-hole circumference. Such methods provide an efficient evaluation instrument for challenging close-cut holes, with the possibility of industrial in-line monitoring and machine learning applications for Industry 4.0 implementation. Additionally, the study compares grain shear angle measurement and Vickers indentation for deformation assessment of the cut edge. It concludes that grain shear angle offers higher resolution. This parameter is therefore postulated as relevant for assessing the sheared-edge zone. The findings contribute to a deeper understanding of sheared-edge damage and improve evaluation methods, potentially enhancing the use of high-strength steels in automotive and safety-critical applications.Validerad;2025;Nivå 2;2025-07-08 (u2);Funder: European Union RFCS (grant number 847213);Full text: CC BY license;</p
Effect of Particle Size on Magnetite Oxidation Behavior: A Modeling Approach Incorporating Ultra-Fine Particle Effects
No full textMagnetite concentrates, essential for pellet production, often contain a high proportion of fine particles. These fine particles significantly influence the induration process. Understanding their oxidation behavior is crucial for optimizing pellet quality. Previous research primarily focused on magnetite particles larger than 25 µm. This study extends the investigation to finer size fractions below 7 µm. Isothermal oxidations were conducted on three different size fractions from two different mines, using a thermogravimetric analyzer (TGA) at 973 K and 1073 K, followed by light optical microscopy to observe the structural evolution of hematite. The oxidation of magnetite exhibits a two-step phenomenon: an initial stage characterized by a high oxidation rate, followed by a second stage where the oxidation degree increases at a constant rate. The oxidation behavior of both studied concentrates follows a consistent pattern: finer particles exhibit faster oxidation than coarser particles, resulting in a higher oxidation degree in a specific duration. Particles in the finer size range (< 7 µm) undergo complete oxidation during the initial stage. A predictive model based on the Avrami kinetic equation was developed to assess the effect of particle size on magnetite oxidation. The model demonstrated a high validation (98 pct), indicating that particle size is a reliable predictor of magnetite oxidation behavior. Validerad;2025;Nivå 2;2025-11-06 (u5);Full text license: CC BY</p
Design för additiv tillverkning inom rymdindustrin : Mot en fördjupad förståelse av ytjämnhet och effektivt designstöd
No full textAs competition in the space industry grows, so does the demand for high-performance, lightweight, and cost-efficient space products. Additive Manufacturing (AM) has emerged as a promising solution, offering design freedom beyond the capabilities of traditional subtractive manufacturing. Laser Powder Bed Fusion (LPBF), in particular, enables the production of metal components with complex geometries and short lead times while still meeting the stringent requirements of the space industry. However, LPBF introduces new design constraints and process-specific challenges, such as surface roughness, that impact dimensional accuracy, material properties, and manufacturability. These issues contribute to design uncertainty during product development, necessitating careful consideration of design decisions and their knock-on impacts on product performance and the overall development process. Addressing these issues requires effective Design for AM (DfAM) support to help engineers balance innovative design potential with production feasibility. This thesis investigates how designers are supported in understanding and addressing AM-specific challenges during product development, with a focus on surface roughness in LPBF. Inspired by the Design Research Methodology, the research comprises five studies, combining systematic literature reviews, an industrial case study, experimental testing, and interviews with twenty AM aerospace professionals. The findings identify several surface roughness–related design considerations and explore how process knowledge can be embedded in design support. Eleven key characteristics of effective design support are identified and used to evaluate a design process for identifying, exploring, and mitigating AM design uncertainties through product-specific AM design artefacts (AMDAs). Leading to the development of the AMDA method, an enhanced framework for structured design uncertainty investigation. Interview insights reveal the state-of-the-art practices, challenges, and gaps in existing DfAM support. Further, models of the aerospace AM design approach are presented, capturing how AM affects the product development process. This thesis offers actionable insights for evaluating and improving DfAM support, helping engineers make better-informed design decisions. It highlights how AM alters the design process and the need to link buildability and performance more explicitly in design support. Overall, the thesis guides the development of AM design support that will aid the creation of easy-to-manufacture, qualifiable, and cost-effective AM product designs for the space industry
AI-drivet talangmatchning : Förutsägelse av medarbetarretention genom kandidatdataanalys
No full textEmployee retention remains a critical challenge in human resource management, as early attrition generates substantial financial and organizational costs. Traditional recruitment systems, particularly Applicant Tracking Systems (ATS), rely on keyword-based filtering and therefore optimize for immediate role fit while neglecting long-term stability. This thesis addresses this gap by designing and evaluating an AI-driven talent matching framework that integrates semantic similarity with retention-aware modelling. The framework combines transformer-based sentence embeddings (LaBSE) with retention predictors derived from employment history, educational background, and psychometric inference of Big Five personality traits. Using two anonymized real-world recruitment datasets, a multi-stage pipeline was developed encompassing preprocessing, modelling, and integration layers. A blended scoring function was designed to unify semantic similarity and retention signals, and an interactive dashboard was implemented to support human-in-the-loop recruitment. Results demonstrate that semantic embeddings alone achieve near-perfect ranking accuracy. Incorporating retention-aware scoring reshapes candidate shortlists by promoting individuals with stronger stability potential, without reducing semantic alignment. Both quantitative evaluation (NDCG and Precision) and qualitative case studies confirm the framework’s ability to balance immediate fit with long-term retention. This work contributes to predictive hiring research by confirming the value of employment history, education, and psychometric traits, and by showing how such attributes can be systematically structured and integrated into a practical hiring framework. Beyond its academic contribution, the approach provides practical benefits for organizations, including reduced early attrition, improved hiring quality, and greater transparency in recruitment decisions
En balansgång helt enkelt : En kvantitativ undersökning om mottagande lärares inställningar till elevöverlämningar inför högstadiet
No full textElever genomgår flera övergångar under sin skoltid. En av dessa är övergången från mellanstadiet till högstadiet där överlämningsaktivitet och information som ges vid en överlämning kan påverka elevernas fortsatta skolgång på högstadiet. Tidigare forskning visar på vikten av samsyn hos lärarna och ett gott samarbete på skolan som organisation för att skapa trygghet och tillit hos de nya eleverna. Mottagande lärare är nyckelspelare vid övergången. Denna studie syftar därför till att undersöka lärarnas upplevelse av vad en god överlämning är och vilken typ av information man önskar vid en överlämning för att kunna bemöta de nya eleverna. För att ta reda på detta har en kvantitativ metod använts genom enkäter där lärarna har fått besvara frågor gällande sin upplevelse, vad de gör och vad de önskar under en överlämning. För att arbeta fram frågorna och analysera enkätens resultat har Sundbergs (2022) tolkning av Läroplansteori använts. Vidare är enkätfrågorna formulerade delvis efter vad Skolverket beskriver om överlämningars innehåll och genomförande och delvis utifrån forskning om övergångsaktiviteter. Läroplansteorin ger en analysmodell för hur resultatet visar på hur information från Skolverket som myndighet (formuleringsarenan), konkretiseras i skolan (realiseringsarenan). Studien visar att det saknas en samsyn hos mottagande lärare på högstadiet gällande vad en god överlämning innebär. Vissa lärare önskar väldigt lite information och andra väldigt mycket information om eleverna. Det uttrycks även en viss brist i samverkan mellan lärare och elevhälsoteam. Detta indikerar ett behov av utvecklingsarbete på skolorna för att öka samsyn och förståelse för överlämningarnas viktiga funktion. Ett utvecklingsarbete skulle påverka möjligheterna att skapa en så likvärdig utbildning som möjligt för alla elever som börjar högstadiet.
Från rivning till resurs : En studie om byggavfall, återbruk och material
No full textDenna rapport undersöker utmaningar och möjligheter kopplade till materialinventering, återvinning och återbruk inom bygg- och rivningssektorn. Genom en kombination av litteraturstudier och intervjuer med relevanta aktörer identifieras centrala tekniska, logistiska och organisatoriska hinder för en effektiv hantering av rivningsmaterial, särskilt betong, glas och metall. Bland de mest framträdande utmaningarna återfinns bristfällig dokumentation av inbyggda material, kontamineringsrisker, brist på lagringsutrymmen samt tidsbrist och avsaknad av planering i projektens tidiga skeden. Rapporten påvisar även att dagens incitamentsstruktur, såväl ekonomiskt som miljömässigt, är otillräcklig för att driva ett storskaligt återbruk.För att främja en mer cirkulär byggsektor föreslås införande av lagkrav för återbruk, etablering av återbrukscentraler samt satsningar på digitala verktyg och utbildningsinsatser. Slutligen betonas vikten av att integrera materialinventering redan i projektens planeringsfas och att bedömningar av miljömässig samt ekonomisk nytta bör vara en självklar del av processen
Vattennivåers påverkan på järnvägens stabilitet—Modellering med Finita Elementmetoden
No full textKlimatförändringarna kommer att ge upphov till ökade regnmängder i framtiden, vilket kan leda till att infrastruktur utsätts för ökade vattenflöden och förhöjda grundvattennivåer. Förhöjda vattennivåer kan ge upphov till portrycksökningar i jord som sänker jordens hållfasthet, ökade laster från uppdämt vatten, vilket försämrar en konstruktions stabilitet mot brott eller skred (ras). Järnvägen, som är en kritisk infrastruktur i Sverige, kan vara känslig för den här typen av stabilitetsförsämrande vattennivåförändringar då de godståg som trafikerar järnvägen bidrar till stora brottmedverkande laster. Den första delen av rapporten, litteraturstudien, fokuserar på att samla information om och beskriva hur olika vattennivåer och vattenförhållanden kan påverka stabiliteten av järnvägen, samt hur järnvägen är uppbyggd, hur Trafikverkets regler kring avvattning på järnväg ser ut, och kortfattat om och hur klimatförändringar kan påverka vattennivåer och vattenförhållanden i jord i Sverige. Huvuddelen av litteraturstudien fokuserar på hur förändrade vattennivåer och portryck påverkar jordars hållfasthet, då järnvägskonstruktionen på bank som sådan i grunden är en geoteknisk konstruktion som kan utsättas för jordskred (ras). Detta gäller både för marken som järnvägen anläggs på och själva järnvägsbanken eller bankroppen i sig. Litteraturstudien avhandlar dock även andra sätt som förändrade vattennivåer och vattenförhållanden kan negativt påverka stabilitet hos järnväg, som erosion och finmaterialvandring (mud pumping). Finita elementmetoden, implementerad i programvaran PLAXIS 2D, valdes för att analysera effekten av förändrade vattennivåer på järnvägens stabilitet. Finita elementmetoden (FEM) är en generell numerisk metod inom strukturmekanik för att lösa partiella differentialekvationer, och PLAXIS 2D är ett program som tillämpar denna för att utföra beräkningar på geotekniska problem i två dimensioner, där modeller av det undersökta problemet skapas i programvaran. Fyra olika sektioner av järnväg på bank, uppdelat till totalt åtta fall/sektioner, har valts att modelleras. Tre av sektionerna är befintliga, och en är projekterad. Två av sektionerna är uppdelat i sektioner för med och utan tryckbankar, och de andra två sektionerna är uppdelade i reella befintliga eller planerade dubbelspår, samt fiktiva motsvarande enkelspår. Materialmodellen NGI-ADP har använts för de sektioner där data om den odränerade skjuvhållfasthetens anisotropi har funnits, Mohr-Coulomb modellen i övriga fall. Beräkningarna med oförändrade/långvariga/befintliga vattennivåer har dessutom verifierats med andra beräkningsmetoder, Morgenstern-Price metoden för släntstabilitet i Slope/W och Meyerhoffs bärighetsekvation. För fallen/sektionerna som har beräknats med NGI-ADP modellen i PLAXIS 2D har dessutom kontrollberäkningar skett med Mohr-Coulomb modellen i PLAXIS 2D. Resultaten från beräkningarna i PLAXIS 2D pekar på att en ökning av grundvattennivån mellan 0,5 och 1,0 meter har en liten inverkan på stabiliteten och är stabilitetshotande endast för sektioner mycket nära brottgränstillstånd. Uppdämt vatten upp till 1 meter under RUK (Rälunderkant) har också en relativt begränsad påverkan på stabiliteten, men kan bli problematiskt för sektioner med dålig stabilitet. Uppdämt vatten (hela vägen) upp till RUK kan bli stabilitetshotande för sektioner med relativt god stabilitet. Climate change will lead to increased rainfall in the future, which may result in higher flow rates and groundwater levels. In turn, this can negatively affect infrastructure, including the stability of geotechnical constructions, since the strength of soil materials depends on the effective stress state, which is influenced by factors such as pore pressure, which changes with the water levels in the soil. Geotechnical structures may also be impacted by static loads caused by standing water. One of the critical types of infrastructure in Sweden that could be negatively affected by changing water levels, even to the point of failure, is railway, particularly as it is subjected to large loads from trains that contribute to landslide or other failure. The first part of the report, the literature review, focuses on gathering information and describing how various water levels and water conditions can affect the stability of the railway embankment structure. It also covers Trafikverket's regulations on railway drainage, and briefly touches on if and how climate change can affect water levels and water conditions in soil in Sweden. The main part of the literature review focuses on how changes in water levels and pore pressure affect soil strength, as a railway embankment is fundamentally a geotechnical structure that can be susceptible to landslides. This applies to both the ground the railway is built on and the railway embankment itself. The literature review also discusses other ways that changes in water levels and water conditions can negatively affect railway stability, such as erosion and fine particle migration (mud pumping). The finite element method, in the PLAXIS 2D software, was selected to analyze the impact of changing water levels on railway stability. Four different sections, divided into a total of eight cases/sections, were modeled. Three of the sections were existing sections, whilst one was planned. Two of the sections are divided into cases with and without retaining embankments, the other two into cases with real (existing or planned) double tacks, and fictional corresponding single tracks. The NGI-ADP material model was used in cases where data on anisotropy in undrained shear strength existed; the Mohr-Coulomb model was used in other cases (and for comparative calculations of the cases modeled with the NGI-ADP model). The calculations for the initial cases (with existing water levels) were verified with other calculation methods, primarily slope stability using the Morgenstern-Price method in the software Slope/W. Rapid drawdown has not been studied. The results from the calculations in PLAXIS 2D indicate that an increase in the groundwater level between 0.5 and 1.0 meters has a small impact on stability and only threatens the stability for sections already very close to a state of failure. Impounded or dammed-up water up to 1 meter below RUK (Underside of Rail) also has a relatively limited effect on stability but can become problematic for sections with poor stability. Impounded water (all the way) up to RUK can become stability-threatening for sections with decent stability.
Handwritten Text Generation with Diffusion Models: Beyond Visual Quality
No full textHandwritten Text Generation (HTG) has emerged as a promising remedy for data scarcity that limits the training of Deep Learning (DL) models for Document Image Analysis and Recognition (DIAR) tasks. Current HTG pipelines are mostly based on adversarial training, which can face several issues, such as mode collapse, limiting variability, and in turn, reducing the practical usefulness of the generated data. Moreover, existing evaluation protocols prioritize visual realism rather than usefulness, obscuring the link between generative quality and downstream task performance. These limitations raise several questions related to which HTG methods truly benefit DIAR tasks, which real datasets should serve as a foundation for synthetic data generation, how Diffusion Models can be adapted for controllable and robust HTG, and whether current evaluation protocols capture downstream utility. Given the success of text-to-image Diffusion Models in generating realistic images from text prompts, their adaptation for handwriting generation offers the potential to produce high-quality, style-aware handwritten data that can directly enhance Handwritten Text Recognition (HTR). The overarching goal of this thesis is to bridge the gap between generation and task-aligned evaluation by showing how controllable diffusion-based HTG, guided by systematic data analysis and evaluated through practical utility, can address these challenges. The contributions of this thesis are organized along three complementary directions. First, to establish a solid basis for HTG, a comprehensive overview of modern, historical, and synthetic document image datasets and HTG methods is presented. This results in C1, a systematic overview of dataset resources, and C2, a detailed survey of generative paradigms and evaluation practices in HTG, identifying key data and methodological gaps that motivate the development of diffusion-based models. Second, to overcome the instability and limited variability of adversarial methods, three diffusion-based approaches are proposed. C3 (WordStylist), a latent diffusion model enabling verbatim text and style conditioning. C4 (DiffusionPen), a few-shot extension of WordStylist capable of generalizing to unseen writers through hybrid classification and metric-learning style embeddings. Moreover, C5 (Dual Orthogonal Guidance), a sampling-time mechanism that enhances stability while preserving stylistic diversity. These proposed HTG methods demonstrate that diffusion-based models can generate realistic, diverse, and style-consistent handwriting under controllable conditions. Third, recognizing that generative quality should translate into practical utility, C6 introduces a task-aligned evaluation framework that links generation metrics to recognition outcomes. This framework measures content preservation, style preservation, robustness to Out-of-Vocabulary (OOV) content, and variability, providing a practical assessment of whether synthetic handwriting improves DIAR performance. By integrating generation and evaluation, this contribution redefines how HTG success is measured. In summary, this thesis demonstrates that controllable diffusion-based HTG, grounded in systematic dataset analysis, enabled by robust generative modeling, and evaluated with task-aligned metrics, provides efficient synthetic handwriting pipelines that directly enhance HTR performance. On a broader level, this work bridges the gap between generative modeling and document analysis, setting the stage for future research in task-aware synthesis of document images