1. Predictive modelling of weld bead geometry in wire arc additive manufacturingKristijan Šket, Miran Brezočnik, Timi Karner, Rok Belšak, Mirko Ficko, Tomaž Vuherer, Janez Gotlih, 2025, original scientific article Abstract: This study investigates the predictive modelling of weld bead geometry in wire arc additive manufacturing (WAAM) through advanced machine learning methods. While WAAM is valued for its ability to produce large, complex metal parts with high deposition rates, precise control of the weld bead remains a critical challenge due to its influence on mechanical properties and dimensional accuracy. To address this problem, this study utilized machine learning approaches—Ridge regression, Lasso regression and Bayesian ridge regression, Random Forest and XGBoost—to predict the key weld bead characteristics, namely height, width and cross-sectional area. A Design of experiments (DOE) was used to systematically vary the welding current and travelling speed, with 3D weld bead geometries captured by laser scanning. Robust data pre-processing, including outlier detection and feature engineering, improved modelling accuracy. Among the models tested, XGBoost provided the highest prediction accuracy, emphasizing its potential for real-time control of WAAM processes. Overall, this study presents a comprehensive framework for predictive modelling and provides valuable insights for process optimization and the further development of intelligent manufacturing systems.
Keywords: wire arc additive manufacturing, WA AM, predictive modelling, machine learning, weld bead geometry, XGBoost
Published in DKUM: 13.03.2025; Views: 0; Downloads: 0
 Full text (3,54 MB) |
2. Application of machine learning to reduce casting defects from bentonite sand mixtureŽiga Breznikar, Marko Bojinović, Miran Brezočnik, 2024, original scientific article Abstract: One of the largest Slovenian foundries (referred to as Company X) primarily focuses on casting moulds for the glass industry. In collaboration with Pro Labor d.o.o., Company X has been systematically gathering defect data since 2021. The analysis revealed that the majority of scrap caused by technological issues is attributed to sand defects. The initial dataset included information on defect occurrences, technological parameters of sand mixture and chemical properties of the cast material. This raw data was refined using data science techniques and statistical methods to support classification. Multiple binary classification models were developed, using sand mixture parameters as inputs, to distinguish between good casting and scrap, with the k-nearest neighbours algorithm. Their performances were evaluated using various classification metrics. Additionally, recommendations were made for development of a real-time industrial application to optimize and regulate pouring temperature in the foundry process. This is based on simulating different pouring temperatures while keeping the other parameters fixed, selecting the temperature that maximizes the likelihood of successful casting
Keywords: gravity casting, machine learning, defects, classifier, data science
Published in DKUM: 11.03.2025; Views: 0; Downloads: 3
Full text (518,07 KB)
This document has many files! More... |
3. Enhancing manufacturing precision: Leveraging motor currents data of computer numerical control machines for geometrical accuracy prediction through machine learningLucijano Berus, Jernej Hernavs, David Potočnik, Kristijan Šket, Mirko Ficko, 2024, original scientific article Abstract: Direct verification of the geometric accuracy of machined parts cannot be performed simultaneously with active machining operations, as it usually requires subsequent inspection with measuring devices such as coordinate measuring machines (CMMs) or optical 3D scanners. This sequential approach increases production time and costs. In this study, we propose a novel indirect measurement method that utilizes motor current data from the controller of a Computer Numerical Control (CNC) machine in combination with machine learning algorithms to predict the geometric accuracy of machined parts in real-time. Different machine learning algorithms, such as Random Forest (RF), k-nearest neighbors (k-NN), and Decision Trees (DT), were used for predictive modeling. Feature extraction was performed using Tsfresh and ROCKET, which allowed us to capture the patterns in the motor current data corresponding to the geometric features of the machined parts. Our predictive models were trained and validated on a dataset that included motor current readings and corresponding geometric measurements of a mounting rail later used in an engine block. The results showed that the proposed approach enabled the prediction of three geometric features of the mounting rail with an accuracy (MAPE) below 0.61% during the learning phase and 0.64% during the testing phase. These results suggest that our method could reduce the need for post-machining inspections and measurements, thereby reducing production time and costs while maintaining required quality standards
Keywords: smart production machines, data-driven manufacturing, machine learning algorithms, CNC controller data, geometrical accuracy
Published in DKUM: 10.03.2025; Views: 0; Downloads: 3
Full text (4,44 MB)
This document has many files! More... |
4. Study of environmental impacts on overhead transmission lines using genetic algorithmsKristijan Šket, Mirko Ficko, Nenad Gubeljak, Miran Brezočnik, 2023, original scientific article Abstract: In our study, we explored the complexities of overhead transmission line (OTL) engineering, specifically focusing on their responses to varying atmospheric conditions (ambient temperature, ambient humidity, solar irradiance, ambient pressure, wind speed, wind direction), and electric current usage. Our goal was to comprehend how these independent variables impact critical responses (dependent variables) such as conductor temperature, conductor sag, tower leg stress, and vibrations – parameters crucial for electric distribution. We modelled the target output variable as a polynomial of a certain degree of the input variables. The precise forms of the polynomial were determined using the genetic algorithms (GA). Developed models are essential for quantifying the influence of each input parameter, enriching our understanding of essential system elements. They provide long-term predictions for assessing transmission line lifespan and structural stability, with particularly high precision in forecasting temperature and sag angle. It is important to note that certain engineering parameters, such as material properties and load considerations, were not included in our research, potentially influencing accuracy.
Keywords: Overhead Transmission Lines (OTL), machine learning, modelling, optimization, genetic algorithms (GA)
Published in DKUM: 10.03.2025; Views: 0; Downloads: 2
Full text (417,77 KB)
This document has many files! More... |
5. Identification of Lithium-Ion Battery Parameter Variations Across Cells using Artificial IntelligenceTine Lubej, 2025, master's thesis Abstract: This thesis focuses on improving the simulation, estimation, and accuracy of parameter identification in lithium-ion battery models. The key objective was to enhance a previously developed program by transitioning it to an object-oriented design, making it more efficient, user-friendly, and modular. Additionally, efforts were made to optimize the parameter estimation process by upgrading the cost function used during simulations and integrating real-world battery measurement data, specifically for the LGM50 battery type.
The first step in the thesis involved reworking the codebase to an object-oriented structure, which improved not only the code’s clarity but also its extensibility and efficiency. With this change, the program was better suited for future improvements and became more accessible for other users through simplified installation procedures. This was accompanied by the implementation of unit testing to ensure the reliability of the code.
Experiments were conducted across a range of discharge rates (from 0.05C to 1C) to evaluate the performance of the model under different conditions. These tests helped to identify trends in how the model responded to changes in operational parameters. Additionally, a dynamic pulse test was performed, which allowed for more precise estimation of the parameters. The results of these tests demonstrated the robustness of the methodology, especially under dynamic conditions.
A major innovation introduced in this thesis was the development of a new cost function, which led to noticeable improvements in parameter estimation accuracy, particularly under high discharge rates and when estimating multiple parameters simultaneously. This new cost function proved especially effective in more complex scenarios, where the original cost function struggled to maintain the same level of accuracy.
The program’s capabilities were further extended by incorporating real experimental data. Using a constant discharge profile for the LGM50 battery, the results showed some challenges when dealing with real-world data, particularly due to issues in measurement or data preprocessing. Nonetheless, the model consistently produced solutions, although the accuracy was influenced by the quality of the input data.
The thesis concludes by highlighting the success of the improvements made, both in terms of the program’s structure and the precision of its estimations. However, it also emphasizes the importance of improving the quality of real-world data to fully leverage the model’s potential in practical applications. This work lays a foundation for future developments in battery modeling, providing a framework that is adaptable for further research and practical use.
Keywords: Machine Learning, Lithium-Ion Batteries, Parameter Estimation, Uncertainty Quantification, Real-experimental data
Published in DKUM: 03.03.2025; Views: 0; Downloads: 10
Full text (5,99 MB) |
6. The use of artificial intelligence in building engineering for historic buildings build in the Austro-Hungarian monarchyDaniela Dvornik Perhavec, Rok Kamnik, 2025, original scientific article Abstract: Knowledge discovery from databases (KDD) and data mining (DM) belong to the field of artificial intelligence (AI). The integration of artificial intelligence into various segments of the construction industry is still in its infancy, but it is expected to be used more widely soon, driven by the development of databases and data warehouses. By using BIM (Building Information Modelling) technologies in the planning of new buildings, we will be able to obtain valuable data. The situation is different for old, existing buildings and the building engineering associated with these properties. Civil engineers, renovation planners and architects need knowledge of the building before renovation. This knowledge is much less than the possibilities that exist. Information about the building can be found in provincial archives. For historic buildings, 10- 15% of the plans, drawings, descriptions, or projects are available. The remaining 85% must be researched on site, which is a lengthy and costly process and hinders the construction process. The question arose as to how the findings from the study of buildings based on written and preserved sources can be applied to the 85% of buildings for which no data is available. This paper presents the use of the collected data as an idea for an initiative to develop a database and modelling using artificial intelligence algorithms. The research study investigates the feature “load-bearing wall” for residential buildings with basements and floors built between 1857 and 1948 in the former Austro-Hungarian Empire. The aim of this study is to create a model to predict the characteristics of a building for which no archival material is available. The study is based on the use of artificial intelligence in the creation of decision trees to help engineers improve their knowledge of historic buildings in the former Austro-Hungarian Empire and building engineering for historic objects.
Keywords: knowledge discovery from data, machine learning, Austro-Hungarian Monarchy buildings
Published in DKUM: 03.03.2025; Views: 0; Downloads: 0
Full text (6,35 MB)
This document has many files! More... |
7. Recent applications of explainable AI (XAI) : a systematic literature reviewMirka Saarela, Vili Podgorelec, 2024, review article Keywords: explainable artificial intelligence, applications, interpretable machine learning, convolutional neural network, deep learning, post-hoc explanations, model-agnostic explanations
Published in DKUM: 31.01.2025; Views: 0; Downloads: 3
Full text (1,42 MB) |
8. Wearable online freezing of gait detection and cueing systemJan Slemenšek, Jelka Geršak, Božidar Bratina, Vesna M. Van Midden, Zvezdan Pirtošek, Riko Šafarič, 2024, original scientific article Abstract: This paper presents a real-time wearable system designed to assist Parkinson’s disease patients experiencing freezing of gait episodes. The system utilizes advanced machine learning models, including convolutional and recurrent neural networks, enhanced with past sample data preprocessing to achieve high accuracy, efficiency, and robustness. By continuously monitoring gait patterns, the system provides timely interventions, improving mobility and reducing the impact of freezing episodes. This paper explores the implementation of a CNN+RNN+PS machine learning model on a microcontroller-based device. The device operates at a real-time processing rate of 40 Hz and is deployed in practical settings to provide ‘on demand’ vibratory stimulation to patients. This paper examines the system’s ability to operate with minimal latency, achieving an average detection delay of just 261 milliseconds and a freezing of gait detection accuracy of 95.1%. While patients received on-demand stimulation, the system’s effectiveness was assessed by decreasing the average duration of freezing of gait episodes by 45%. These preliminarily results underscore the potential of personalized, real-time feedback systems in enhancing the quality of life and rehabilitation outcomes for patients with movement disorders.
Keywords: Parkinson’s disease, freezing of gait, machine learning, real-time systems, wearable devices, on-demand stimulation
Published in DKUM: 31.01.2025; Views: 0; Downloads: 3
Full text (6,29 MB) |
9. |
10. Efficient compressed storage and fast reconstruction of large binary images using chain codesDamjan Strnad, Danijel Žlaus, Andrej Nerat, Borut Žalik, 2024, original scientific article Abstract: Large binary images are used in many modern applications of image processing. For instance, they serve as inputs or target masks for training machine learning (ML) models in computer vision and image segmentation. Storing large binary images in limited memory and loading them repeatedly on demand, which is common in ML, calls for efficient image encoding and decoding mechanisms. In the paper, we propose an encoding scheme for efficient compressed storage of large binary images based on chain codes, and introduce a new single-pass algorithm for fast parallel reconstruction of raster images from the encoded representation. We use three large real-life binary masks to test the efficiency of the proposed method, which were derived from vector layers of single-class objects – a building cadaster, a woody vegetation landscape feature map, and a road network map. We show that the masks encoded by the proposed method require significantly less storage space than standard lossless compression formats. We further compared the proposed method for mask reconstruction from chain codes with a recent state-of-the-art algorithm, and achieved between and faster reconstruction on test data
Keywords: binary mask, machine learning, chain code, binary encoding, bitmap reconstruction
Published in DKUM: 29.01.2025; Views: 0; Downloads: 7
Full text (1,45 MB) |
