Luleå University of Technology Publications
Not a member yet
39356 research outputs found
Sort by
Proficiency of Physics Informed Machine Learning in Multi-component Fault Recognition of Rotational Machines under Different Speed Conditions
No full textUnderstanding the limitations of incorporating conventional machine learning synergy led to the inclusion of physics knowledge. This study presents the potency of physics-informed feature engineering for machine learning to enhance fault detection in gears, shafts, and bearings at three constant-speed running conditions. AI models such as Decision Tree (DT), Random Forest (RF), and Support Vector Machine-Radial Basis Function (SVM-RBF) are constructed to verify traditional statistical performance metric and physics-based signal descriptors. Additionally, time-frequency domain representation as spectrogram images is fed into the CNN-oriented ResNet-152 architecture to demonstrate the skillfulness of the model’s ability. Based on the results obtained, RF is observed to be supreme with 98.42% upon applying physics-centric parameters when compared with statistical variables. To make an inference, further comparison of the best classification model’s accuracy using physics expertise when accounted with ResNet image-based categorization, physics-grounded RF models have premier achievements. Thus, it is concluded that physical laws are expedient in offering exceptional outcomes for identifying various defects in complex industrial rotary machines in different operating modes.Validerad;2025;Nivå 1;2025-06-10 (u8);Full text license: CC BY</p
Vision Transformer & Brain Connectivity Patterns for Estimating Cognitive States
No full textThe transformer model is excellent at handling time series signals (such as electroencephalography: EEG) because it can extract information from long-term dependencies effectively. This work combines binarization of EEG connectivity features, cognitive state classification using the vision transformer (ViT), and identifying graphical connectivity patterns for each cognitive state of the mental arithmetic task. The common spatial pattern (CSP) filter coefficient-based channel selection method selects the optimum EEG channels from the input channel set. Then, the Singular Value Decomposition (SVD) method is applied to prepare the binarized connectivity feature matrices, eliminating noisy connections between the optimum channels. The binarized functional-effective connectivity features are passed to the ViT model for cognitive state classification. The ViT model achieves the maximum classification accuracy of 94.86% with the phase-based connectivity feature. The proposed model improves classification accuracy by 6.15% compared to the state-of-the-art studies. This study also suggests a robust brain connectivity network to build a graphical connectivity pattern for each cognitive state. My findings of the EEG-based graphical patterns will bring further understanding of the scalp-level EEG channel patterns among different brain regions for other cognitive tasks.Validerad;2025;Nivå 2;2025-05-12 (u5);Full text license: CC BY 4.0;</p
Thermodynamic Equilibrium Study of Ash Transformation during Entrained Flow Conversion of Agricultural Biomass Focusing on the Potential Extraction of Valuable Si and K-P Compounds via Condensation from the Gas Phase
No full textAgricultural biomass is today largely underutilized in combustion and gasification processes because of the abundant supply of other easier-to-process biomass fuels. These biomass types generally have a moderate to high ash content comprising valuable elements, such as Si, K, and P, which can lead to ash-related operational problems. The high share of Si, K, and P in agricultural biomass assortments also has a significant economic value. These elements are usually retained in the coarse or fly ash fractions. Extracting valuable Si- and K–P-containing compounds with high purity from these ash fractions often requires further postprocessing steps, which increases operational costs. Therefore, a potential novel design concept could be to control the combustion/gasification processes so that Si, K, and P can be extracted by condensation from the flue/hot gases at a quality that implies added value instead of extra costs. This work aims to identify the possibilities of extracting valuable Si and K–P compounds from the hot gases generated during entrained flow conditions via stepwise controlled condensation in the close-flame regions or heat exchanger zone. Thermodynamic equilibrium calculations were performed by employing the databases (GTOX and SGPS) in FactSage 8.0 software. The calculations were performed under varying conditions ,i.e., temperatures, atmospheres, and fuel compositions. The selected fuels were rice husks (Si-rich), brewer’s spent grains (P–Si-rich with moderate to minor amounts of Ca, Mg, and K), and grass (K–Si-rich with moderate amounts of Ca, Mg, and P). The results indicate that the high-temperature formation of the valuable Si compounds, such as SiC (s) and Si2N2O (s), would require an inert atmosphere during both the release and cooling stages. Moreover, a high Si/P molar ratio is needed to form valuable Si-containing compounds. The predicted K-bearing phosphates during the gas cooling near the burner zone were formed in the same temperature range as Ca-, Mg-, and Si-containing compounds with all of the fuels. The results obtained by this study can guide experimental research on the practical extraction of Si and K–P compounds from different types of agricultural biomass during thermochemical conversion in entrained flow conditions.Validerad;2025;Nivå 2;2025-05-12 (u5):Full text license: CC BY 4.0;Funder: Austrian Research Promotion Agency (892426);</p
Faktorer som påverkar företag i Software-as-a-Service-sektorn vid internationell expansion : En kvalitativ fallstudie i byggteknologisektorn
No full textThis thesis examines strategic expansion decisions of Software-as-a-Service (SaaS) firms in the construction technology sector, where the rise of digitization creates both new opportunities and challenges for international entry. Drawing on previous research on the SaaS model and international business development, the study identifies business-specific, market-related and organizational factors that are particularly relevant in this context. The study is based on a qualitative, exploratory case study of the company Infobric, where semi-structured interviews were conducted with key personnel to capture internal perspectives on expansion. The theoretical framework combines established models of market entry and strategic decisionmaking with SaaS-specific expansion prerequisites, divided into external and internal analysis. The analysis identifies three recurring themes, such as regulatory complexity, cultural distance and the need for organizational coordination when entering new markets. The study results in a framework operationalized in a questionnaire where the answers are visualized in a radar graph, providing guidance on which entry strategy is most appropriate in a particular market situation – or whether the company should wait with expansion until internal or external conditions are met. The work thus contributes both a practical tool and a theoretical framework for understanding how SaaS companies can reason about international growth.Detta examensarbete undersöker strategiska expansionsbeslut hos Software-as-a-Service (SaaS)-företag inom byggteknologisektorn, där digitaliseringens framfart skapar både nya möjligheter och utmaningar vid internationell etablering. Utifrån tidigare forskning om SaaSmodellen och internationell affärsutveckling identifieras affärsspecifika, marknadsrelaterade och organisatoriska faktorer som är särskilt relevanta i denna kontext. Studien bygger på en kvalitativ, explorativ fallstudie av företaget Infobric, där semistrukturerade intervjuer genomförts med nyckelpersoner i syfte att fånga interna perspektiv på expansion. Det teoretiska ramverket kombinerar etablerade modeller för marknadsinträde och strategiskt beslutsfattande med SaaS-specifika expansionsförutsättningar, uppdelat i extern och intern analys. I analysen identifieras ett antal återkommande teman, såsom regulatorisk komplexitet, kulturell distans och behovet av organisatorisk samordning vid inträde på nya marknader. Studien resulterar i ett ramverk som operationaliserats i ett frågeformulär där svaren visualiseras i en radargraf, vilket ger vägledning kring vilken inträdesstrategi som är mest lämplig i en viss marknadssituation – eller om företaget bör avvakta med expansion till dess att interna eller externa förutsättningar är uppfyllda. Arbetet bidrar därmed med både ett praktiskt verktyg och ett teoretiskt ramverk för att förstå hur SaaS-bolag kan resonera kring internationell tillväxt
Micellar Solubilization of Phenols With One or Two Hydroxyl Groups Using Biological Surfactant Rhamnolipid
No full textWe studied the solubilization of phenols with one and two hydroxyl groups (phenol, p-cresol, guaiacol, and pyrocatechol, resorcinol, hydroquinone) by micelles of the biological surfactant rhamnolipid using NMR diffusometry. We discuss the results within the framework of a model of two states of solubilizer molecules in solution: free in the aqueous phase and bound in surfactant micelles. The solubilization characteristics of rhamnolipid were calculated: the proportion of solubilized molecules р, micelle-water partition coefficient Km, and molar solubilization ratio (MSR) depending on the concentration of rhamnolipid in solutions.Validerad;2025;Nivå 2;2025-06-25 (u2);Funder: Kazan Federal University, FZSM-2023-0016;Full text: CC BY license;</p
Potential Techniques to Prevent Eutrophication Through Aluminum Sulfate Addition or Oxygen Injection : A case study in Lake Svartbyträsket, Boden, Sweden
No full textEutrophic lakes characterized by algal blooms and oxygen depleted bottoms symbolize a worldwide environmental challenge and a clear indicator of human impact on aquatic ecosystems. Eutrophication occurs when nutrient input in the forms of nitrogen (N) and phosphorus (P) is exceptionally high, contributing to excessive algal growth in the lake. The decomposition of dead algae consumes oxygen, resulting in degraded water quality and loss of habitat for aquatic organisms. Numerous efforts have been implemented to mitigate algal blooms and enhance lake water quality. However, there are few methods proven to be effective in achieving long-term recovery from eutrophication. The purpose of this study was to evaluate two different methods that aim to reduce phosphorus and permanently bind it in the sediments. The first method involved oxygen injection into the water, thereby forming iron hydroxides (Fe(OH)₃) under oxic conditions, which could adsorb phosphorus. The second method was performed by adding aluminum sulfate, Al2(SO₄)3 into the water that could form aluminum hydroxides (Al(OH)3 or AlOOH). These aluminum complexes also had the potential to absorb phosphorus in a way that immobilizes it, making phosphorus unavailable for nutrient uptake. This study focused on Svartbyträsket, a lake in Boden, Sweden which exhibits evident signs of nutrient enrichment and eutrophication. The main purpose of the research was to evaluate two eutrophication mitigation strategies and determine which one that could be best suited for application in Svartbyträsket. The experimental procedure involved column experiments conducted at laboratory scale using sediment and water from the deepest point of the lake (11 meters, SV26). Three column experiments were performed: (1) oxygen injection, (2) reference, and (3) addition of aluminum sulfate. In addition to the column experiments, probe measurements were carried out in April 2025 at the sampling location, (SV26) where pH, oxygen, electrical conductivity (EC), and temperature were evaluated at one-meter intervals. Ultrafiltrated water samples (<1 kD) and total concentrations of elements in the water were also collected during this occasion. To determine the composition of the sediments, SEM-EDS analyses were conducted both before and after the column experiments to observe how the sediments were affected by the experiment. The results show that the deepest part of the lake becomes oxygen-depleted at a depth of 4 meters during the winter, and that the water column becomes stratified. No major geochemical changes happened in column 1 and column 2 during the experiment. A significant pH decrease occurred in column 3, and the aluminum concentration in the water increased markedly after the addition of Al₂(SO₄)₃. Framboidal pyrite (FeS₂) had also formed in column 3 after the column experiments. The main conclusions of this study were that the method involving aluminum sulfate addition is not suitable for improving water quality in Svartbyträsket. That is primarily due to the high levels of aluminum, which contributed to acidic water and thus dissolution of aluminum hydroxide complexes. The oxygen injection method, on the other hand, could potentially be applicable in the lake. However, further investigations are required to obtain a deeper understanding of the lake.
A Statistical Bonded Particle Model Study on Laboratory Scale Rock Drilling
No full textRock drilling is a crucial process in many industries, one example being the mining industry, where it is used for exploration and blasting. In a typical rock drilling process, the rock surface is fractured by dynamic mechanical interaction with a drill bit, resulting in rock fragments detaching from the surface. These cuttings are then transported through the borehole via water or air, and the rock fragment size is important for efficient borehole flushing. In this work, a heterogeneous bonded particle model was calibrated and applied to a laboratory scale rock drilling process. The mineral grain structure was obtained from an electron microscope scan of the rock surface, and the average grain size, volume percentage and stiffness of the three most common minerals were represented in the model. The dynamic mechanical behaviour of the rock material was obtained by conducting uniaxial compression and Brazilian disc tests in a split-Hopkinson pressure bar configuration. The results were used to calibrate the model. After the heterogeneous model was shown to be able to capture the macroscopic strengths and fracture modes of the split-Hopkinson experiments, it was used to simulate the laboratory scale rock drilling experiment, where two tool indentation depths were investigated. Here, the simulation was compared to experimental results in terms of vertical load acting on the tool, machine compliance as well as rock-cutting size distributions. The results from the simulation were in good agreement with the experimental observations.Fulltext license: CC BY;This article has previously appeared as a manuscript in a thesis.DigiRoc
Affärsrelationer och digitala kommunikationsverktyg i utveckling : En studie om svenska SMF i internationell handel
No full textFöretagens omställning och anpassning till digitala kommunikationsverktyg har skett snabbt och ofta utan reflektion över dess långsiktiga effekter för att möta omedelbara behov. Studien syftar till att bidra till förståelsen av digitala kommunikationsverktyg samt dess inverkan på internationell handel och tillföra insikter som kan förbättra internationell extern kommunikation för små- till medelstora företag (SMF). Detta genom att besvara forskningsfrågan: Vilken roll digitala kommunikationsverktyg spelar i att initiera, vidareutveckla och avsluta affärsrelationer samt vilka utmaningar och möjligheter användningen av digitala kommunikationsverktyg medför för små- till medelstora företag som bedriver internationell handel? Forskningsfrågan besvarades med en kvalitativ metod som bygger på intervjuer med fem respondenter i inköps- såväl som försäljningspositioner på olika SMF, med olika lång erfarenhet av export och import från olika branscher. Studien bygger vidare på teorin kring processhantering av affärsrelationer av Mitrega och Pfajfar (2015). Sammanfattningsvis framkommer det i studien att vidareutveckling är det momentet i processhanteringen av dyadiska affärsrelationer som är starkt präglat av användandet av digitala kommunikationsverktyg. Däremot finns det inte en enhetlig strategi bland SMF vid valet av kommunikationsform vid inledandet och avslutandet av en dyadisk affärsrelation, eftersom valet påverkas istället av branschspecifika faktorer. I studien identifieras fyra utmaningar med digitala kommunikationsverktyg: misskommunikation, cybersäkerhet, informationsöverflöd och kulturella skillnader, samt tre möjligheter: effektivitet, attityder och tillit samt ökad exponering. Möjligheterna med digitala kommunikationsverktyg överväger utmaningarna då samtliga respondenter använder sig av digitala kommunikationsverktyg men på olika sätt och i olika delar av processhanteringen av affärsrelationerna.
Discrete element analysis on aggregate breakage of unbound granular road materials
No full textAdequate aggregate resistance to crushing and abrasion is crucial for good performance of unbound road layers, in particular when incorporating marginal aggregates in road construction. In this study, aggregate crushing in unbound granular materials (UGMs) is investigated experimentally with uniaxial compression tests, and numerically with discrete element method (DEM). A DEM model of UGMs subjected to uniaxial monotonic compression is implemented in a commercial DEM software, incorporating granular mechanics-based particle-contact and breakage laws. Based on comparison with the experimental findings, it is shown that the implemented model captures the UGMs behavior well including aggregate breakage characteristics. The model is used furthermore to evaluate aggregate crushing in UGMs subjected to cyclic loading representative for unbound road layers of low-volume roads. The feasibility of using the developed DEM approach for evaluating the implications of using marginal UGMs in unbound road layers and for optimizing structural design of those layers is discussed.Validerad;2025;Nivå 2;2025-07-02 (u5);Full text license: CC BY 4.0;</p