Linköping Electronic Conference Proceedings
Not a member yet
1113 research outputs found
Sort by
Nonlinearity analysis of variables for modelling and control
Full text linkNonlinearities become essential in various systems when the operating area widens. The linear models are special cases for narrow areas. The behaviour is often asymmetric and can become gradually steeper or flatter depending on the case. These nonlinear effects can be analysed from data distributions for chosen operating areas. Further extensions require recursive analysis. The widely used Gaussian distribution is seldom valid for a wide area. The variable specific scaling can be presented with two second order polynomial defined by five parameters interpreted as the operating point and four corner points of the feasible range. These parameters define the shape factors which may require adjusting to fill the only requirement that the functions need to be monotonously increasing. Alternative constraints provide good solutions for combining expert knowledge with the data-based analysis. If the nonlinear behaviour is analysed correctly, only linear interactions are needed in the models. As the analysis is based on the same methodology, different applications can be combined by using appropriate process data. The smooth operation and high quality of products is the main goal of all these applications, and this can be achieved by combining these indicators with process control in the same way as it has been one for smaller indicators used in lime kiln control and water treatment. Different parts of the methodology have been tested in versatile applications. The main benefit is that the same structures can be used in various applications since the scaling functions take care of linking to the real world
A Gaussian Mixture Model Approach for Characterizing Playing Styles of Ice Hockey Players
Full text linkPlayer categorization based on playing style is a highly important task in professional ice hockey, aiding scouting, player development, and strategic decision-making. Traditional methods often rely on simple metrics like goals or assists, which fail to capture the full complexity of a player’s style and contributions. Motivated by the increasing availability of detailed event data and advances in machine learning based modeling techniques, this paper explores a richer, data-driven approach to player categorization. We build on recent work in player vector representations and apply Gaussian Mixture Models (GMMs) to cluster forwards and defenders based on event data from five seasons of the Swedish Hockey League (SHL). Our contributions are threefold: (1) we construct detailed player vectors that summarize a wide range of offensive and defensive skills, (2) we apply GMMs to identify soft clusters of players, allowing for nuanced overlapping playing styles, and (3) we analyze the resulting clusters to interpret distinct player profiles and provide concrete examples. Our results offer a more flexible and realistic view of player roles, reflecting the continuous and multi-dimensional nature of playing styles. The approach helps enhance talent evaluation and roster building, and offers an efficient framework for future analyses across leagues and seasons
Sally Jones + the Chief = A Relationship Counteracting Species Boundaries
Full text linkThis study analyses the interspecies relationship between Sally Jones, a gorilla, and her human companion, the Chief, in three novels by Jakob Wegelius, focusing on this dynamic relationship and how it reframes the narrative. The research explores three main questions: the modes of interspecies communication, the core elements of their companionship and how their mutual dependency subverts notions of human supremacy. The theoretical framework integrates perspectives from human-animal studies, including Donovan’s critique of speciesist ideologies that separate human and animal communication (2017), Haraway’s concepts of mutual dependence (2003) and contact zones (2008) in interspecies relations and Derrida's carnofallogocentrism, which interrogates anthropocentric hierarchies (2002). Methodologically, the study employs close reading of the texts paying attention to details in the texts, based on the research questions, theories and previous research, analyzing verbal and non-verbal communication between Sally and the Chief, as well as their interdependent relationship. Findings reveal that their companionship is rooted in mutual respect and understanding, transcending human-animal hierarchies. This relationship critiques notions of human supremacy, as the Chief and Sally navigate their lives as equals, disrupting conventional ideas of ownership and superiority, as well as species and societal boundaries and fostering empathy for ‘the Other’. The study contributes to the broader discourse on interspecies relationships, showing how narratives can shape ethical understandings of human-animal relations and potentially reshape perceptions of human-animal relationships in children’s literature. 
Pipeline-Based Automated Integration and Delivery Testing of Simulation Assets with FMI/SSP in a Railway Digital Twin
Full text linkRailway infrastructure systems have recently been enhanced through the use of the digital twin (DT) concept, enabling visualization and control in a virtual environment while effectively mitigating life cycle costs. This work provides insights into the development and operations (DevOps) of a railway DT platform and highlights the automation and management of asset integration and processing based on the FMI and SSP interface standards through the use of the Continuous Integration / Continuous Delivery pipeline technology. This offers long-term durability, pausability, remote triggering, open-source and workflow design capabilities, and connectivity to other tools such as version control systems and code analysis tools. In this research paper, we present an anti-slip cosimulation model of a railway vehicle as a use case example to demonstrate the pipeline-oriented automation and management in combination with a version control system and code analysis tool within the platform
Simulation and Cost Estimation of CO2 Capture with alternatives for doubled capacity
Full text linkThis study presents a techno-economic assessment of an amine-based carbon capture technology. The aim is to compare different methods to evaluate the cost effect of doubling the capacity. A base case was established in Aspen HYSYS with 15 m absorber packing height, 6 m desorber packing height, removal efficiency of 85 % and a heat exchanger minimum temperature approach (ΔTmin) of 10 °C. Then dimensioning and cost estimation was carried out using Aspen HYSYS spreadsheets to automatically calculate CAPEX, OPEX and carbon capture cost per ton CO2 captured. To estimate the Bare Erected Cost (BEC), the Enhanced Detailed Factor (EDF) and the Aspen Process Economic Analyzer (APEA) were employed. The EDF method determines the installation cost of each piece of equipment, while the Nazir-Amini method only offers the Total Plant Cost (TPC). Applying the EDF method, the TPC for the base case, the doubled feed gas case and the two-absorber case were calculated to 76, 141 and 150 MEuro respectively. The estimated annual OPEX for the base case was 42.5 MEuro, while for the two alternatives the OPEX was very close to the double of the base case. The estimated carbon capture cost for the base case, two-absorber case, and double feed gas scenario were 52.4 €/ton, 51.8 €/ton, and 50.5 €/ton, respectively. The study demonstrates that a combination of Aspen HYSYS simulation, APEA and the EDF method is an effective method to evaluate different alternatives for increasing the capacity
Design of electrified fluidized bed calciner for direct capture of CO2 from cement raw meal
Full text linkUsing green electricity to calcine the raw materials and combining this with storage of the pure CO2 generated in the calcination process can significantly reduce CO2 emissions in the cement industry, which generates around 7 % of the global CO2 emissions.In this study, a lab-scale electrically heated fluidized bed calciner, operating with a mixture of fine meal particles and coarse inert particles, is simulated using CPFD software. The electrification of the reactor is done using several horizontal cylinders, which are electrically heated to provide energy both for heating the raw meal (with 77% CaCO3) up to the calcination temperature and for calcination (CaCO3 CaO + CO2). The reactor design is done based on a specified electrical energy input, the gas velocity required for fluidization of coarse inert particles and the velocity required for entrainment of the fine calcined particles. A fluidization velocity of 0.3 m/s appears to be optimal for the reactor, whereas 0.8 m/s resulted in complete entrainment of the bed. The maximum calcination degree achieved was 90% when operating with preheated meal. The average meal residence time was found to be 24-26 s
Phase Transformations in Steelmaking Slags: A Thermodynamic Approach
Full text linkIn addition to solidification, steelmaking slags may undergo phase transformations in solid state during their cooling process. The mineralogy of these oxide slags is significantly influenced by the chemical composition and cooling rate. For the phases forming, two distinct solidification modes can be assumed, depending on the cooling rate: equilibrium cooling and Scheil–Gulliver cooling. Characterization methods, such as scanning electron microscopy (SEM) and electron probe microanalyzer (EPMA) allow analyzing the elemental composition of individual phases. Here, computational thermodynamics were applied in phase identification of crystallized electric arc furnace (EAF) slags. FactSage 8.3 thermodynamic calculation software was used to estimate the composition of stable phases as a function of temperature. Solid solutions with varying compositions were considered in this study. The calculation results from two solidification modes, i.e., equilibrium cooling and Scheil-Gulliver cooling, were saved in Excel spreadsheets. A MATLAB script was developed to go through the results and find the phase with a composition closest to the input values. For both solidification modes, the composition and temperature best fitting the input analysis was determined. The input is the elemental composition of the phase of interest, acquired using EPMA. After the data processing, the results are visualized in graphs, illustrating the analyzed and estimated compositions of the identified solid solution phase and its occurrence temperature
Machine Learning -based Optimization of Biomass Drying Process: Application of Utilizing Data Center Excess Heat
Full text linkThe utilization of biomass as a renewable energy source holds significant promise for climate mitigation efforts. Excess heat from Nordic data centers offers opportunities for sustainable energy utilization. This research explores the feasibility of using data center excess heat for biomass drying to enhance the biomass energy value. In this study, the challenge of predicting biomass moisture under demanding measurement conditions is addressed by developing a predictive model for exhaust air humidity from the dryer. This model indirectly describes biomass moisture and employs machine learning methods such as linear regression model (LM), gradient boosting machines (GBM), eXtreme gradient boosting (XGBoost), random forest (RF), and multilayer perceptron (MLP), while enhancing transparency through explainable artificial intelligence (XAI) techniques for analyzing and visualizing humidity fluctuations. Based on this study, it can be demonstrated that tree-based ensemble methods GBM, RF, and XGBoost can accurately predict the humidity of air exiting the dryer with coefficient of determination from 0.88 to 0.89. Weather conditions, supply air humidity, and dryer fan speed emerged as key factors affecting drying efficiency, providing actionable insights for process optimization. Specific thresholds for these features can be defined to facilitate process settings. Moreover, improving system air tightness enhances drying efficiency and mitigates weather effects. The model shows promising predictive capabilities for exhaust air humidity, enabling future dynamic modeling to indirectly predict biomass end moisture, enabling adaptive control of drying processes, optimizing production capacities, and advancing sustainable energy through AI-driven solutions
Ice Hockey Action Recognition via Contextual Priors
Full text linkSkeleton-based action recognition models, which are developed for generic human-pose data, struggle with ice-hockey broadcasts player action recognition, where the players appear smaller, move abruptly, and wield sticks that are invisible to standard skeleton models. To address these issues, we propose CP-Hockey, a context-aware pipeline that incorporates two domain-specific priors. First, a temporal player’s boundingbox normalization stabilizes player scale across the player tracklet, raising top-1 accuracy from 31 % to 57 % on a six-class NHL dataset. Second, we design hockey-specific skeletons that include stick end-points and optional detailed head landmarks. A 15-keypoint body-plus-stick model improves the accuracy to 64 %, while our full 20-keypoint configuration reaches 65 %. Experimental results with STGCN++ and 2s-AGCN show that both contextual priors are necessary: scale normalization reduces spatial jitter, and stick keypoints disambiguate visually similar movements such as stickwork versus striking a puck with a stick. CP-Hockey establishes a strong baseline for fine-grained ice-hockey analytics and provides a blueprint for adapting skeleton pipelines to other equipment-centric sports
Individual Puck Possessions Part I: Frequency, Duration, and Distance Travelled
Full text linkIn this paper we use puck and player tracking data from the 2023- 24 NHL season to study individual player possessions (focusing on 5v5 situations). We study metrics such as possession count, average and total possession duration, average and total distance travelled with the puck, and examine relationships between these metrics and traditional measures of success (i.e., goals, assists and points). A key finding is that individual offensive zone possession is strongly correlated with points (r = 0.70) and is moderately correlated with goals (r = 0.64), assists (r = 0.54), and shots on goal (r = 0.69). We also observe differences in individual possessions based on position (forwards versus defence), zone of play, and strength and large and statistically significant differences between top ranked players and league averages (across most possession metrics). Finally, we examine the benefits of our individual possession metrics and find that they are highly stable (so they are useful for predictions), able to differentiate players, and provide information not captured by existing metrics