Search

This study aims to experimentally assess repeatability and reproducibility of direct tensile strength (DTS) tests with deformability measurements on two types of rocks: Blanco Mera granite (Spain) and Cotta sandstone (Germany). The tests were conducted in four rock mechanics laboratories located in different countries (Canada, Germany, Spain and Sweden). A total of 51 tests were performed on cylindrical specimens of the two rocks, using different test equipment and measuring devices. Mean and standard deviation DTS values were determined in the four laboratories for the granite (5.70 ± 0.32, 6.06 ± 0.11, 3.84 ± 0.50 and 6.76 ± 0.10 MPa) and for the sandstone (1.88 ± 0.07, 1.96 ± 0.06, 1.15 ± 0.32 and 1.74 ± 0.19 MPa), together with Young’s moduli and Poisson’s ratios in tension, being statistically analysed to evaluate the variability and compare the main results obtained from the participating laboratories. The findings indicate that the DTS test with deformability measurements on cylindrical rock specimens is operationally feasible. However, certain shortcomings have been identified during the course of the experiments with the existing methodologies, such as the one suggested by the ISRM for DTS tests. The results have also shown to be sensitive to appropriate test and strain measurement configurations. The objective of this study was to shed light on these issues and provide new insights for potential future improvements of the existing testing methods.

This study aims to experimentally assess repeatability and reproducibility of direct tensile strength (DTS) tests with deformability measurements on two types of rocks: Blanco Mera granite (Spain) and Cotta sandstone (Germany). The tests were conducted in four rock mechanics laboratories located in different countries (Canada, Germany, Spain and Sweden). A total of 51 tests were performed on cylindrical specimens of the two rocks, using different test equipment and measuring devices. Mean and standard deviation DTS values were determined in the four laboratories for the granite (5.70 ± 0.32, 6.06 ± 0.11, 3.84 ± 0.50 and 6.76 ± 0.10 MPa) and for the sandstone (1.88 ± 0.07, 1.96 ± 0.06, 1.15 ± 0.32 and 1.74 ± 0.19 MPa), together with Young’s moduli and Poisson’s ratios in tension, being statistically analysed to evaluate the variability and compare the main results obtained from the participating laboratories. The findings indicate that the DTS test with deformability measurements on cylindrical rock specimens is operationally feasible. However, certain shortcomings have been identified during the course of the experiments with the existing methodologies, such as the one suggested by the ISRM for DTS tests. The results have also shown to be sensitive to appropriate test and strain measurement configurations. The objective of this study was to shed light on these issues and provide new insights for potential future improvements of the existing testing methods.

Solar energy forecasting is essential for the effective integration of solar power into electricity grids and the optimal management of renewable energy resources. Distinguishing itself from the existing literature, this review study provides a nuanced contribution by centering on advancements in forecasting techniques. While preceding reviews have examined factors such as meteorological input parameters, time horizons, the preprocessing methodology, optimization, and sample size, our study uniquely delves into a diverse spectrum of time horizons, spanning ultrashort intervals (1 min to 1 h) to more extended durations (up to 24 h). This temporal diversity equips decision makers in the renewable energy sector with tools for enhanced resource allocation and refined operational planning. Our investigation highlights the prominence of Artificial Intelligence (AI) techniques, specifically focusing on Neural Networks in solar energy forecasting, and we review supervised learning, regression, ensembles, and physics-based methods. This showcases a multifaceted approach to address the intricate challenges associated with solar energy predictions. The integration of Satellite Imagery, weather predictions, and historical data further augments precision in forecasting. In assessing forecasting models, our study describes various error metrics. While the existing literature discusses the importance of metrics, our emphasis lies on the significance of standardized datasets and benchmark methods to ensure accurate evaluations and facilitate meaningful comparisons with naive forecasts. This study stands as a significant advancement in the field, fostering the development of accurate models crucial for effective renewable energy planning and emphasizing the imperative for standardization, thus addressing key gaps in the existing research landscape.

In an effort to optimize laptop performance for gaming and high-demand applications without costly hardware upgrades, this research investigates the impact of CPU undervoltage using the AMD Ryzen Mobile 4600H processor. Undervolting, the process of reducing the CPU's voltage supply, is proposed as a strategy to enhance performance by lowering operational temperatures, potentially allowing for more efficient processing. This study uses the AMD APU Tuning Utility to adjust voltage settings and assesses performance changes using a series of benchmarks. Initial findings indicate that undervoltage can indeed have beneficial effects. The most significant data point from the research is the comparison of Cinebench R23 scores before and after applying undervolting settings. From a baseline score of 6835 points, system performance increased to 7880 points in the optimal undervolting scenario, an improvement of 1045 points. This shows a noticeable enhancement in processing efficiency. However, the study also reveals some complexities in undervolting, such as an initial drop in performance in the first configuration before gains are realized in subsequent adjustments. Efficiency values varied across different settings, starting with a decrease (-0.41) and culminating in a substantial gain (+1.54) by the fourth configuration. These results suggest that while undervolting can improve performance, the outcomes depend significantly on finding the right voltage balance, highlighting the nuanced nature of CPU voltage manipulation for performance optimization.

Introduction: A textbook outcome patient is one in which the operative course passes uneventful, without complications, readmission or mortality. There is a lack of publications in terms of TO on acute cholecystitis. Objetive: The objective of this study is to analyze the achievement of TO in patients with urgent early cholecystectomy (UEC) for Acute Cholecystitis. and to identify which factors are related to achieving TO. Materials and methods: This is a post hoc study of the SPRiMACC study. It ́s a prospective multicenter observational study run by WSES. The criteria to define TO in urgent early cholecystectomy (TOUEC) were no 30-day mortality, no 30-day postoperative complications, no readmission within 30 days, and hospital stay ≤ 7 days (75th percentile), and full laparoscopic surgery. Patients who met all these conditions were taken as presenting a TOUEC. Outcomes: 1246 urgent early cholecystectomies for ACC were included. In all, 789 patients (63.3%) achieved all TOUEC parameters, while 457 (36.6%) failed to achieve one or more parameters and were considered non-TOUEC. The patients who achieved TOUEC were younger had significantly lower scores on all the risk scales analyzed. In the serological tests, TOUEC patients had lower values for in a lot of variables than non-TOUEC patients. The TOUEC group had lower rates of complicated cholecystitis. Considering operative time, a shorter duration was also associated with a higher probability of reaching TOUEC. Conclusion: Knowledge of the factors that influence the TOUEC can allow us to improve our results in terms of textbook outcome.

OpenFOAM is a CFD software widely used in both industry and academia. The exaFOAM project aims at enhancing the HPC scalability of OpenFOAM, while identifying its current bottlenecks and proposing ways to overcome them. For the assessment of the software components and the code profiling during the code development, lightweight but significant benchmarks should be used. The answer was to develop microbenchmarks, with a small memory footprint and short runtime. The name microbenchmark does not mean that they have been prepared to be the smallest possible test cases, as they have been developed to fit in a compute node, which usually has dozens of compute cores. The microbenchmarks cover a broad band of applications: incompressible and compressible flow, combustion, viscoelastic flow and adjoint optimization. All benchmarks are part of the OpenFOAM HPC Technical Committee repository and are fully accessible. The performance using HPC systems with Intel and AMD processors (x86_64 architecture) and Arm processors (aarch64 architecture) have been benchmarked. For the workloads in this study, the mean performance with the AMD CPU is 62% higher than with Arm and 42% higher than with Intel. The AMD processor seems particularly suited resulting in an overall shorter time-to-solution., This work is carried out in the scope of the exaFOAM project, which has received funding from the German Federal Ministry of Education and Research and the European High-Performance Computing Joint Undertaking (JU) under grant agreements No. 16HPC022K and No. 956416, respectively. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and France, Germany, Spain, Italy, Croatia, Greece, and Portugal., Peer Reviewed, "Article signat per 13 autors/es: Flavio C. C. Galeazzo, Marta Garcia-Gasulla,Elisabetta Boella, Josep Pocurull, Sergey Lesnik, Henrik Rusche , Simone Bnà, Matteo Cerminara , Federico Brogi, Filippo Marchetti, Daniele Gregori, R. Gregor Weiß, andAndreas Ruopp" ", Postprint (published version)

Artiklen beskriver hvordan vi har udviklet en række datasæt – et såkaldt benchmark – til at evaluere forskellige aspekter af sprogforståelse i danske sprogmodeller. Vores antagelse er at den viden der allerede er beskrevet i en række eksisterende danske ordbøger, kan opfattes som ’ground truth’ for semantikken i det danske ordforråd. Vores metode går derfor ud på at ’vende’ de semantiske ordbøger om og bruge dem til at generere et benchmark der afprøver modellernes evne til at forstå dansk. Mere specifikt undersøger vi hvor godt modellerne i) forstår synonymi, nærsynonymi, og hvornår noget er semantisk associeret, ii) skaber inferens i relation til begrebsmæssig viden og nedarvning af egenskaber fra overbegreb til underbegreb, iii) laver korrekte følgeslutninger i forbindelse med specifikke handlinger og hændelser, iv) skelner mellem centrale betydninger af et ord i kontekst og v) håndterer positiv og negativ konnotation eller ’sentiment’ i løbende tekst. Vi afprøver vores datasæt på ChatGPT 3.5 turbo og på ChatGPT 4.0 og kan se at datasættene har en passende sværhedsgrad i forhold til hvad modellerne er i stand til at håndtere, om end ChatGPT 4.0 opnår særdeles gode resultater for flere af datasættene.

Artiklen beskriver hvordan vi har udviklet en række datasæt – et såkaldt benchmark – til at evaluere forskellige aspekter af sprogforståelse i danske sprogmodeller. Vores antagelse er at den viden der allerede er beskrevet i en række eksisterende danske ordbøger, kan opfattes som ’ground truth’ for semantikken i det danske ordforråd. Vores metode går derfor ud på at ’vende’ de semantiske ordbøger om og bruge dem til at generere et benchmark der afprøver modellernes evne til at forstå dansk. Mere specifikt undersøger vi hvor godt modellerne i) forstår synonymi, nærsynonymi, og hvornår noget er semantisk associeret, ii) skaber inferens i relation til begrebsmæssig viden og nedarvning af egenskaber fra overbegreb til underbegreb, iii) laver korrekte følgeslutninger i forbindelse med specifikke handlinger og hændelser, iv) skelner mellem centrale betydninger af et ord i kontekst og v) håndterer positiv og negativ konnotation eller ’sentiment’ i løbende tekst. Vi afprøver vores datasæt på ChatGPT 3.5 turbo og på ChatGPT 4.0 og kan se at datasættene har en passende sværhedsgrad i forhold til hvad modellerne er i stand til at håndtere, om end ChatGPT 4.0 opnår særdeles gode resultater for flere af datasættene.

Large language models (LLMs) are gaining increasing popularity in both academia and industry, owing to their unprecedented performance in various applications. As LLMs continue to play a vital role in both research and daily use, their evaluation becomes increasingly critical, not only at the task level, but also at the society level for better understanding of their potential risks. Over the past years, significant efforts have been made to examine LLMs from various perspectives. This paper presents a comprehensive review of these evaluation methods for LLMs, focusing on three key dimensions: what to evaluate, where to evaluate, and how to evaluate. Firstly, we provide an overview from the perspective of evaluation tasks, encompassing general natural language processing tasks, reasoning, medical usage, ethics, education, natural and social sciences, agent applications, and other areas. Secondly, we answer the 'where' and 'how' questions by diving into the evaluation methods and benchmarks, which serve as crucial components in assessing the performance of LLMs. Then, we summarize the success and failure cases of LLMs in different tasks. Finally, we shed light on several future challenges that lie ahead in LLMs evaluation. Our aim is to offer invaluable insights to researchers in the realm of LLMs evaluation, thereby aiding the development of more proficient LLMs. Our key point is that evaluation should be treated as an essential discipline to better assist the development of LLMs. We consistently maintain the related open-source materials at: https://github.com/MLGroupJLU/LLM-eval-survey © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

Traditional methods for object detection typically necessitate a substantial amount of training data, and creating high-quality training data is time-consuming. We propose a novel Few-Shot Object Detection network (FSODv2) in this paper that aims to detect objects from previously unseen categories using only a few annotated examples. Attention RPN, Multi-Relation Detector, and Contrastive Training strategy are central to our method (Fan et al., in: CVPR, 2020), which exploit similarity between few shot support set and query set to detect novel objects while suppressing false detection in the background. We also contribute a new dataset, FSOD-1k, which contains 1000 categories of various objects with high-quality annotations to train our network. To the best of our knowledge, this is one of the first datasets designed for few-shot object detection. This paper improves our FSOD model through well-designed model calibration in three areas: (1) we propose an improved FPN with multi-scale support inputs to calibrate the multi-scale support-query feature matching by exploiting multi-scale features from the same support image with different input scales; (2) we introduce a support classification supervision branch to calibrate the support feature supervision, aligning to the query feature training supervision; (3) we propose backbone calibration to preserve prior knowledge while alleviating backbone bias toward base classes by employing classification dataset to help our model calibration procedure, where such dataset has previously only been used for pre-training in other related works. Besides, we propose a Fast Attention RPN to improve evaluation speed and save computational memory during inference. Once trained, our few-shot network can detect objects from previously unseen categories without further training or fine-tuning, resulting in new state-of-the-art performance on different datasets in the few-shot setting. Our method is general in scope and has numerous poten

Thermal-related problems in electric machinery are complex problems with different physics. Most of the problems are related to the motion of fluid due to cooling or motions of the machinery; thus, numerical methods such as computational fluid dynamics (CFD) are commonly used. However, due to the complexity of the problems, it is costly to utilize. Thus, this thesis investigated an alternative method to the most common method today, the finite volume method (FVM). The alternative method uses a lagrangian approach or a mesh-free approach, as it can also be known. The method is moving particle simulation (MPS), which is also based on the lagrangian approach of smooth particle hydrodynamics (SPH) but with modifications to improve stability, convergence, and more. Thus, the aim of the thesis is to evaluate the feasibility, accuracy, and modelling aspects of the MPS method irrelevant problems. This is done by investigating three canonical cases and comparing the results to the finite volume method (FVM) and existing literature. The canonical cases relate to the thermal issues of an electric motor, namely external cooling, the rotor-stator gap, and oil cooling, and are related to the backwards facing step, Taylor-Couette flow, and a free round single impinging jet, respectively. In all these cases, the MPS method was used and a sensitivity analysis was performed to estimate the discretization error. The FVM method was also done by the author using large eddy simulation (LES)for two of the cases, where an estimation of the discretization error was also performed. The results indicated that the MPS method did not perform well in the first two canonical cases, exhibiting low accuracy and high computational cost, thus rendering it unsuitable for internal flows. However, in the last case, the MPS method aligned with the literature and showed potential for reduced computational cost. Nevertheless, obtaining crucial thermal parameters such as the Nusselt number with a satisfacto

Denna rapport undersöker huruvida det befintliga nätverket av publika laddstationer i Hallands län är tillräckligt för att möta den växande efterfrågan i linje med elektrifieringen av vägtrafiken, som i sin tur kan hjälpa Sverige nå sina långsiktiga klimatmål. Vad som är att betrakta som tillräckligt definieras i denna rapports kontext utifrån riktmärken angivna av Europeiska unionen (EU) och European Automobile Manufacturers’ Association (ACEA). För att genomföra analysen skapades en karta med publika laddningsstationer i programmet ArcMap, till vilken geodata över administrativa länsgränser och vägar lades till. Enligt Europaparlamentets direktiv 2014/94 krävs en laddningspunkt per tio elbilar - ett förhållande på 1:10. ACEA tillhandahåller en liknande, men inte identisk, riktlinje där de rekommenderar ett förhållande på 1:7 i Sverige. Båda dessa riktlinjer överträffas av den nuvarande infrastrukturen enligt resultaten som redovisats i denna rapport, som uppnår ett förhållande på 1:4,3. ACEA specificerar dock ytterligare en riktlinje - att 134 laddningspunkter krävs per 100 km väg längs huvudkorridorerna i TEN-T-nätverket. I detta sammanhang visar resultatet från rapporten att det befintliga nätverket av publika laddstationer endast uppgår till 118 per 100 km. Även om det är betryggande att den befintliga infrastrukturen uppfyller två av de tre nämnda riktlinjerna, bör den uppfylla alla för att anses vara tillräcklig. Slutsatsen i denna rapport är därför att det nuvarande nätverket av offentliga laddstationer inte är tillräckligt. Tilläggningsvis har det även i rapporten observerats att endast 57% av länets befolkning har en eller fler publika laddningsstationer inom 2 km - vilket ytterligare antyder att den befintliga infrastrukturen ej erbjuder tillräcklig täckning för invånarna. Fortsatt utveckling och utbyggnad av laddinfrastrukturen är nödvändig för att stödja den ökande elektrifieringen av vägtrafiken och bidra till att uppnå Sveriges långsiktiga klimatmål., This report examines whether the existing network of public charging stations in Halland County is sufficient to meet the growing demand in line with the increasing number of electric cars, which in turn would help Sweden reach its long-term climate goals. In the context of this report, what is considered sufficient is defined using benchmarks from the European Union (EU) and the European Automobile Manufacturers’ Association (ACEA). To carry out the analysis, a map of public charging stations was created in ArcMap, to which geodata of county boundaries and road networks was added. According to the European Parliament's directive 2014/94, one charging point is required per ten electric cars - a ratio of 1:10. The ACEA provides a similar, but not identical, benchmark where they recommend a ratio of 1:7 in Sweden. Both of these benchmarks are exceeded by the current infrastructure according to results shown in this report, which achieves a ratio of 1:4.3. However, the ACEA also specify another benchmark - that 134 charging points are required per 100 km of road along the main corridors of the TEN-T network. In this context, the results of this report show that the existing network of public charging stations only number 118 per 100 km. Although it is reassuring that the existing infrastructure achieves two of the three outlined benchmarks, it should achieve all of them in order for it to be considered sufficient. The conclusion of this report is therefore that the current network of public charging stations is not sufficient. Additionally, it has in this report been observed that only 57% of the county’s population have one or more public charging stations within 2 km - further indicating that the current infrastructure does not provide sufficient coverage for the populace. Continued development and expansion of the charging infrastructure is necessary to support the increasing electrification of road traffic and contribute to achieving Sweden's long-term climate goal

Third-party dependency updates can cause a build to fail if the new dependency version introduces a change that is incompatible with the usage: this is called a breaking dependency update. Research on breaking dependency updates is active, with works on characterization, understanding, automatic repair of breaking updates, and other software engineering aspects. All such research projects require a benchmark of breaking updates that has the following properties: 1) it contains real-world breaking updates; 2) the breaking updates can be executed; 3) the benchmark provides stable scientific artifacts of breaking updates over time, a property we call 'reproducibility'. To the best of our knowledge, such a benchmark is missing. To address this problem, we present BUMP, a new benchmark that contains reproducible breaking dependency updates in the context of Java projects built with the Maven build system. BUMP contains 571 breaking dependency updates collected from 153 Java projects. BUMP ensures long-term reproducibility of dependency updates on different platforms, guaranteeing consistent build failures. We categorize the different causes of build breakage in BUMP, providing novel insights for future work on breaking update engineering. To our knowledge, BUMP is the first of its kind, providing hundreds of real-world breaking updates that have all been made reproducible., Part of ISBN 9798350330663QC 20240823

Taking a better look at subjects of interest helps humans to improve confidence in their age estimation. Unlike still images, sequences offer spatio-temporal dynamic information that contains many cues related to age progression. A review of previous work on video-based age estimation indicates that this is an underexplored field of research. This may be caused by a lack of well-defined and publicly accessible video benchmark protocol, as well as the absence of video-oriented training data. To address the former issue, we propose a carefully designed video age estimation benchmark protocol and make it publicly available. To address the latter issue, we design a video-specific age estimation method that leverages pseudo-labeling and semi-supervised learning. Our results show that the proposed method outperforms image-based baselines on both offline and online benchmark protocols, while the online estimation stability is improved by more than 50%.

Choosing and developing performant database solutions helps organizations optimize their operational practices and decision-making. Since graph data is becoming more common, it is crucial to develop and use them in big data with complex relationships with high and consistent performance. However, legacy database technologies such as MySQL are tailored to store relational databases and need to perform more complex queries to retrieve graph data. Previous research has dealt with performance aspects such as CPU and memory usage. In contrast, energy usage and temperature of the servers are lacking. Thus, this paper evaluates and compares state-of-the-art graphs and relational databases from the performance aspects to allow a more informed selection of technologies. Graph-based big data applications benefit from informed selection database technologies for data retrieval and analytics problems. The results show that Neo4j performs faster in querying connected data than MySQL and ArangoDB, and energy, CPU, and memory usage performances are reported in this paper., Norwegian ranking - 1 point https://kanalregister.hkdir.no/publiseringskanaler/KanalTidsskriftInfo.action?id=469469

In the 2023 phase of the international collaborative DECOVALEX modeling project, Task E focused on understanding thermal, hydrological, and mechanical (THM) processes related to predicting brine migration in the excavation damaged zone around a heated excavation in salt. Salt is attractive as a disposal medium for radioactive waste because it is self-healing and is essentially impermeable and non-porous in the far field. Investigation of the short-term, near-field behavior is important for radioactive waste disposal because this early period strongly controls the amount of inflowing brine. Brine leads to corrosion of waste forms and waste packages, and possible dissolution of radionuclides with brine transport being a potential transport vector to the accessible environment.The Task was divided into steps. Step 0 included matching unheated brine inflow data from boreholes at the Waste Isolation Pilot Plant (WIPP) and matching temperature observations during a Brine Availability Test in Salt (BATS) heater test. Step 1 included validation of models against a thermo-poroelastic analytical solution, and two-phase flow around an excavation. Finally, Step 2 required all the individual components covered in steps 0 and 1 to come together to match observed brine inflow behavior during the same BATS heater test.There were a range of approaches from the teams, from mechanistic to prescriptive. Given the uncertainties in the problem, some teams used one- or two-dimensional models of the processes, while other teams included more geometrical complexity in three-dimensional models. Task E was a learning experience for the teams involved, and feedback from the modeling teams has led to changes in follow-on BATS experiments at WIPP. The primary Task E lessons learned were the impact of hydrologic initialization methods (wetting up vs. drying down), the difference between confined and unconfined thermal expansion, and the large changes in permeability associated with heating and co

Taking a better look at subjects of interest helps humans to improve confidence in their age estimation. Unlike still images, sequences offer spatio-temporal dynamic information that contains many cues related to age progression. A review of previous work on video-based age estimation indicates that this is an underexplored field of research. This may be caused by a lack of well-defined and publicly accessible video benchmark protocol, as well as the absence of video-oriented training data. To address the former issue, we propose a carefully designed video age estimation benchmark protocol and make it publicly available. To address the latter issue, we design a video-specific age estimation method that leverages pseudo-labeling and semi-supervised learning. Our results show that the proposed method outperforms image-based baselines on both offline and online benchmark protocols, while the online estimation stability is improved by more than 50%.

The evolution of distributed systems, microservices, and IoT applications has elevated the importance of efficient communication protocols and architectures. While gRPC, with its bidirectional streaming and protocol buffers, presents a compelling alternative to traditional REST and other communication protocols, a systematic performance comparison is needed. Previous published comparative studies exist where the targeted formats have been gRPC versus REST, or where more protocols were included, but the review was limited. A more in-depth empirical study involving scalability across various types of data transfers as well as the impact of security on performance is lacking. This thesis addresses the aforementioned gap by thoroughly analyzing gRPC's latency, throughput, and scalability performance against widely used communication protocols in a local setting. Through benchmarking and empirical evaluations, the study aims to provide actionable insights into the strengths and limitations of gRPC contra REST and SOAP, aiding developers and organizations in making informed decisions when selecting communication protocols for a microservice system architecture. By simulating real-world clients interacting with microservices implemented with SOAP, REST, and gRPC and assessing their request throughput and latency across different message shapes, the results show that gRPC outperforms both SOAP and REST. The performance advantage exhibited by gRPC is notably pronounced for smaller messages, although for larger messages, the improvement over REST is only marginally better. SOAP, on the other hand, consistently displayed the poorest performance among the three protocols across all experiments. The results also show that gRPC handles scaling clients the best and its performance is least impacted by enabling the security protocol TLS., Utvecklingen av distribuerade system, mikrotjänster och IoT-applikationer har ökat betydelsen av effektiva kommunikationsarkitekturer och -protokoll. Medan gRPC, med sin tvåvägsströmning och Protocol Buffers, presenterar ett övertygande alternativ till traditionella REST och andra kommunikationsprotokoll, behövs en systematisk prestandajämförelse. Tidigare publicerade jämförande studier finns där de utvlada formaten har varit gRPC kontra REST, eller där fler protokoll ingick, men där granskningen varit begränsad. En mer djupgående empirisk studie som involverar skalbarhet över olika typer av dataöverföringar samt påverkan av säkerhet på prestanda saknas. Denna avhandling adresserar det tidigare nämnda gapet genom att noggrant analysera gRPC:s latens-, genomströmnings- och skalbarhetsprestanda mot väletablereade kommunikationsprotokoll. Genom prestandamätning och empiriska utvärderingar i en lokal miljö syftar studien till att tillhandahålla handlingsbara insikter i gRPC:s styrkor och begränsningar jämfört med REST och SOAP, vilket hjälper utvecklare och organisationer att fatta informerade beslut vid val av kommunikationsprotokoll för en mikrotjänstarkitektur. Genom att simulera verkliga klienter som interagerar med mikrotjänster implementerade med SOAP, REST och gRPC och bedöma deras förfrågningsgenomströmning och latens över olika meddelandeformer, visar resultaten att gRPC presterar bättre än både SOAP och REST. Prestandafördelen som visas av gRPC är tydligt märkbar för mindre meddelanden, medan för större meddelanden är förbättringen över REST bara marginellt bättre. SOAP, å andra sidan, visade konsekvent sämst prestanda bland de tre protokollen över alla experiment. Resultaten visar också att gRPC hanterar uppskalning av klienter bäst och dess prestanda påverkas minst av att säkerhetsprotokollet TLS är aktiverat.

Thermal-related problems in electric machinery are complex problems with different physics. Most of the problems are related to the motion of fluid due to cooling or motions of the machinery; thus, numerical methods such as computational fluid dynamics (CFD) are commonly used. However, due to the complexity of the problems, it is costly to utilize. Thus, this thesis investigated an alternative method to the most common method today, the finite volume method (FVM). The alternative method uses a lagrangian approach or a mesh-free approach, as it can also be known. The method is moving particle simulation (MPS), which is also based on the lagrangian approach of smooth particle hydrodynamics (SPH) but with modifications to improve stability, convergence, and more. Thus, the aim of the thesis is to evaluate the feasibility, accuracy, and modelling aspects of the MPS method irrelevant problems. This is done by investigating three canonical cases and comparing the results to the finite volume method (FVM) and existing literature. The canonical cases relate to the thermal issues of an electric motor, namely external cooling, the rotor-stator gap, and oil cooling, and are related to the backwards facing step, Taylor-Couette flow, and a free round single impinging jet, respectively. In all these cases, the MPS method was used and a sensitivity analysis was performed to estimate the discretization error. The FVM method was also done by the author using large eddy simulation (LES)for two of the cases, where an estimation of the discretization error was also performed. The results indicated that the MPS method did not perform well in the first two canonical cases, exhibiting low accuracy and high computational cost, thus rendering it unsuitable for internal flows. However, in the last case, the MPS method aligned with the literature and showed potential for reduced computational cost. Nevertheless, obtaining crucial thermal parameters such as the Nusselt number with a satisfacto

Denna rapport undersöker huruvida det befintliga nätverket av publika laddstationer i Hallands län är tillräckligt för att möta den växande efterfrågan i linje med elektrifieringen av vägtrafiken, som i sin tur kan hjälpa Sverige nå sina långsiktiga klimatmål. Vad som är att betrakta som tillräckligt definieras i denna rapports kontext utifrån riktmärken angivna av Europeiska unionen (EU) och European Automobile Manufacturers’ Association (ACEA). För att genomföra analysen skapades en karta med publika laddningsstationer i programmet ArcMap, till vilken geodata över administrativa länsgränser och vägar lades till. Enligt Europaparlamentets direktiv 2014/94 krävs en laddningspunkt per tio elbilar - ett förhållande på 1:10. ACEA tillhandahåller en liknande, men inte identisk, riktlinje där de rekommenderar ett förhållande på 1:7 i Sverige. Båda dessa riktlinjer överträffas av den nuvarande infrastrukturen enligt resultaten som redovisats i denna rapport, som uppnår ett förhållande på 1:4,3. ACEA specificerar dock ytterligare en riktlinje - att 134 laddningspunkter krävs per 100 km väg längs huvudkorridorerna i TEN-T-nätverket. I detta sammanhang visar resultatet från rapporten att det befintliga nätverket av publika laddstationer endast uppgår till 118 per 100 km. Även om det är betryggande att den befintliga infrastrukturen uppfyller två av de tre nämnda riktlinjerna, bör den uppfylla alla för att anses vara tillräcklig. Slutsatsen i denna rapport är därför att det nuvarande nätverket av offentliga laddstationer inte är tillräckligt. Tilläggningsvis har det även i rapporten observerats att endast 57% av länets befolkning har en eller fler publika laddningsstationer inom 2 km - vilket ytterligare antyder att den befintliga infrastrukturen ej erbjuder tillräcklig täckning för invånarna. Fortsatt utveckling och utbyggnad av laddinfrastrukturen är nödvändig för att stödja den ökande elektrifieringen av vägtrafiken och bidra till att uppnå Sveriges långsiktiga klimatmål., This report examines whether the existing network of public charging stations in Halland County is sufficient to meet the growing demand in line with the increasing number of electric cars, which in turn would help Sweden reach its long-term climate goals. In the context of this report, what is considered sufficient is defined using benchmarks from the European Union (EU) and the European Automobile Manufacturers’ Association (ACEA). To carry out the analysis, a map of public charging stations was created in ArcMap, to which geodata of county boundaries and road networks was added. According to the European Parliament's directive 2014/94, one charging point is required per ten electric cars - a ratio of 1:10. The ACEA provides a similar, but not identical, benchmark where they recommend a ratio of 1:7 in Sweden. Both of these benchmarks are exceeded by the current infrastructure according to results shown in this report, which achieves a ratio of 1:4.3. However, the ACEA also specify another benchmark - that 134 charging points are required per 100 km of road along the main corridors of the TEN-T network. In this context, the results of this report show that the existing network of public charging stations only number 118 per 100 km. Although it is reassuring that the existing infrastructure achieves two of the three outlined benchmarks, it should achieve all of them in order for it to be considered sufficient. The conclusion of this report is therefore that the current network of public charging stations is not sufficient. Additionally, it has in this report been observed that only 57% of the county’s population have one or more public charging stations within 2 km - further indicating that the current infrastructure does not provide sufficient coverage for the populace. Continued development and expansion of the charging infrastructure is necessary to support the increasing electrification of road traffic and contribute to achieving Sweden's long-term climate goal

This paper examines and compares the efficiency of storing and retrieving similar radar track reports in a vector database, as well as the error resistance of the database, using various indexing techniques available in Pgvector and PostgreSQL's native indexing methods. By using generated dummy data and open-source tools such as Pgvector, PostgreSQL, and the Rust programming language, the study demonstrates the implementation's high level of error resistance and rapid performance. The results highlight the potential of vector databases in handling radar signal data, showcasing their robustness and speed in various scenarios, but fails to conclusively answer whether vector databases can reliably identify radar signals from similar sources. This suggests that further research and optimization are necessary to fully understand and enhance the capabilities of vector databases in this application., Denna uppsats undersöker och jämför hur effektivt en vektordatabas kan spara och hämta liknande data från signalrapporter, samt hur resistent vektordatabasen är mot felaktigheter i datan, genom att använda olika indexeringstekniker som finns tillgängliga i PostgreSQL och Pgvector. Via användning av genererad data och verktyg med öppen källkod så som PostgreSQL, Pgvector och programmeringsspråket Rust demonstrerar studien implementationens höga nivå av felresistens och höga prestanda. Resultaten påvisar implementationens stabilitet och hastighet i olika scenarion men lyckas inte get ett slutgiltigt svar på om vektordatabaser kan pålitligt indentifiera radarsignaler från liknande källor. Detta tyder på att det krävs fortsatt forksning och optimisering för att skapa gedigen förståelse i ämnet och förbättra vektordatabasers förmågor inom området.

Today fund matching is a natural part of Swedes finance and is a popular form of savings that includes a large number of investors in the Swedish fund market. This in turn generates an increased interest in how portfolio managers should locate and acquire knowledge in portfolio selection. This gives a greater interest in how different investment strategies can be affected and generate an investors wealth to an increased level within the stock market, which gives an increased focus to be able to generate as high risk-adjusted return as possible. The study partly presents traditional theory and background on modern portfolio theory and the efficient market hypothesis. Empirical studies also present within the financial market that demonstrate the differences of opinion between how actively versus passively managed funds have performed and which investment strategy is most beneficial for investment. The purpose of the study is to compare realized return on active versus passive funds during long term, short term and specific time periods that had a lot of economic fluctuations, like bear markets. Within the study 10 actively managed funds and two index measures are selected to be studied and compared based on their respective performance, both within its rise and fall in the Swedish fund market. The performance measures will then be applied to be able to produce the results of the study and to be able to answer whether the active fund’s have any statistically significant over- and underperformance. After conducting single index models and t-test on the 10 active funds, the result of the study shows that despite using two benchmarks index, ten different active funds, long time period, short time period or specific time periods defined by market imbalance , we still resulted in many P-values that was not statistically significant. Active funds failed to overperform against passive funds, but passive funds also failed to outperform our selection of active funds.

Taking a better look at subjects of interest helps humans to improve confidence in their age estimation. Unlike still images, sequences offer spatio-temporal dynamic information that contains many cues related to age progression. A review of previous work on video-based age estimation indicates that this is an underexplored field of research. This may be caused by a lack of well-defined and publicly accessible video benchmark protocol, as well as the absence of video-oriented training data. To address the former issue, we propose a carefully designed video age estimation benchmark protocol and make it publicly available. To address the latter issue, we design a video-specific age estimation method that leverages pseudo-labeling and semi-supervised learning. Our results show that the proposed method outperforms image-based baselines on both offline and online benchmark protocols, while the online estimation stability is improved by more than 50%.

The evolution of distributed systems, microservices, and IoT applications has elevated the importance of efficient communication protocols and architectures. While gRPC, with its bidirectional streaming and protocol buffers, presents a compelling alternative to traditional REST and other communication protocols, a systematic performance comparison is needed. Previous published comparative studies exist where the targeted formats have been gRPC versus REST, or where more protocols were included, but the review was limited. A more in-depth empirical study involving scalability across various types of data transfers as well as the impact of security on performance is lacking. This thesis addresses the aforementioned gap by thoroughly analyzing gRPC's latency, throughput, and scalability performance against widely used communication protocols in a local setting. Through benchmarking and empirical evaluations, the study aims to provide actionable insights into the strengths and limitations of gRPC contra REST and SOAP, aiding developers and organizations in making informed decisions when selecting communication protocols for a microservice system architecture. By simulating real-world clients interacting with microservices implemented with SOAP, REST, and gRPC and assessing their request throughput and latency across different message shapes, the results show that gRPC outperforms both SOAP and REST. The performance advantage exhibited by gRPC is notably pronounced for smaller messages, although for larger messages, the improvement over REST is only marginally better. SOAP, on the other hand, consistently displayed the poorest performance among the three protocols across all experiments. The results also show that gRPC handles scaling clients the best and its performance is least impacted by enabling the security protocol TLS., Utvecklingen av distribuerade system, mikrotjänster och IoT-applikationer har ökat betydelsen av effektiva kommunikationsarkitekturer och -protokoll. Medan gRPC, med sin tvåvägsströmning och Protocol Buffers, presenterar ett övertygande alternativ till traditionella REST och andra kommunikationsprotokoll, behövs en systematisk prestandajämförelse. Tidigare publicerade jämförande studier finns där de utvlada formaten har varit gRPC kontra REST, eller där fler protokoll ingick, men där granskningen varit begränsad. En mer djupgående empirisk studie som involverar skalbarhet över olika typer av dataöverföringar samt påverkan av säkerhet på prestanda saknas. Denna avhandling adresserar det tidigare nämnda gapet genom att noggrant analysera gRPC:s latens-, genomströmnings- och skalbarhetsprestanda mot väletablereade kommunikationsprotokoll. Genom prestandamätning och empiriska utvärderingar i en lokal miljö syftar studien till att tillhandahålla handlingsbara insikter i gRPC:s styrkor och begränsningar jämfört med REST och SOAP, vilket hjälper utvecklare och organisationer att fatta informerade beslut vid val av kommunikationsprotokoll för en mikrotjänstarkitektur. Genom att simulera verkliga klienter som interagerar med mikrotjänster implementerade med SOAP, REST och gRPC och bedöma deras förfrågningsgenomströmning och latens över olika meddelandeformer, visar resultaten att gRPC presterar bättre än både SOAP och REST. Prestandafördelen som visas av gRPC är tydligt märkbar för mindre meddelanden, medan för större meddelanden är förbättringen över REST bara marginellt bättre. SOAP, å andra sidan, visade konsekvent sämst prestanda bland de tre protokollen över alla experiment. Resultaten visar också att gRPC hanterar uppskalning av klienter bäst och dess prestanda påverkas minst av att säkerhetsprotokollet TLS är aktiverat.

This paper examines and compares the efficiency of storing and retrieving similar radar track reports in a vector database, as well as the error resistance of the database, using various indexing techniques available in Pgvector and PostgreSQL's native indexing methods. By using generated dummy data and open-source tools such as Pgvector, PostgreSQL, and the Rust programming language, the study demonstrates the implementation's high level of error resistance and rapid performance. The results highlight the potential of vector databases in handling radar signal data, showcasing their robustness and speed in various scenarios, but fails to conclusively answer whether vector databases can reliably identify radar signals from similar sources. This suggests that further research and optimization are necessary to fully understand and enhance the capabilities of vector databases in this application., Denna uppsats undersöker och jämför hur effektivt en vektordatabas kan spara och hämta liknande data från signalrapporter, samt hur resistent vektordatabasen är mot felaktigheter i datan, genom att använda olika indexeringstekniker som finns tillgängliga i PostgreSQL och Pgvector. Via användning av genererad data och verktyg med öppen källkod så som PostgreSQL, Pgvector och programmeringsspråket Rust demonstrerar studien implementationens höga nivå av felresistens och höga prestanda. Resultaten påvisar implementationens stabilitet och hastighet i olika scenarion men lyckas inte get ett slutgiltigt svar på om vektordatabaser kan pålitligt indentifiera radarsignaler från liknande källor. Detta tyder på att det krävs fortsatt forksning och optimisering för att skapa gedigen förståelse i ämnet och förbättra vektordatabasers förmågor inom området.

When it comes to in-vehicle networks (IVNs), the controller area network (CAN) bus dominates the market; automobiles manufactured and sold worldwide depend on the CAN bus for safety-critical communications between various components of the vehicle (e.g., the engine, the transmission, the steering column). Unfortunately, the CAN bus is inherently insecure; in fact, it completely lacks controls such as authentication, authorization, and confidentiality (i.e., encryption). Therefore, researchers have travailed to develop automotive security enhancements. The automotive intrusion detection system (IDS) is especially popular in the literature—due to its relatively low cost in terms of money, resource utilization, and implementation effort. That said, developing and evaluating an automotive IDS is often challenging; if researchers do not have access to a test vehicle, then they are forced to depend on publicly available CAN data—which is not without limitations. Lack of access to adequate CAN data, then, becomes a barrier to entry into automotive security research. We seek to lower that barrier to entry by introducing a new CAN dataset to facilitate the development and evaluation of automotive IDSs. Our datasets—dubbed can-dataset, can-log, can-csv, can-ml, and can-train-and-test—provide CAN data from four different vehicles produced by two different manufacturers. The attack captures for each vehicle model are equivalent, enabling researchers to assess the ability of a given IDS to generalize to different vehicle models and even different vehicle manufacturers. Our datasets contain replayable .log files as well as labeled and unlabeled .csv files, thereby meeting a variety of development and evaluation needs. In particular, the can-train-and-test dataset offers nine unique attacks, ranging from denial of service (DoS) to gear spoofing to standstill; as such, researchers can select a subset of the attacks for training and save the remainder for testing in order to ass

Multiple software frameworks based on the Discrete Element Method (DEM) are available for simulating granular materials. All of them employ the same principles of explicit time integration, with each time step consisting of three main steps: contact detection, calculation of interactions, and integration of the equations of motion. However, there exist significant algorithmic differences, such as the choice of contact models, particle and wall shapes, and data analysis methods. Further differences can be observed in the practical implementation, including data structures, architecture, parallelization and domain decomposition techniques, user interaction, and the documentation of resources. This study compares, verifies, and benchmarks nine widely-used software frameworks. Only open-source packages were considered, as these are freely available and their underlying algorithms can be reviewed, edited, and tested. The benchmark consists of three common bulk processes: silo emptying, drum mixing, and particle impact. To keep it simple and comparable, only standard features were used, such as spherical particles and the Hertz-Mindlin model for dry contacts. Scripts for running the benchmarks in each software are provided as a dataset.

Accurately predicting weather and climate in cities is critical for safeguarding human health and strengthening urban resilience. Multimodel evaluations can lead to model improvements; however, there have been no major intercomparisons of urban-focussed land surface models in over a decade. Here, in Phase 1 of the Urban-PLUMBER project, we evaluate the ability of 30 land surface models to simulate surface energy fluxes critical to atmospheric meteorological and air quality simulations. We establish minimum and upper performance expectations for participating models using simple information-limited models as benchmarks. Compared with the last major model intercomparison at the same site, we find broad improvement in the current cohort's predictions of short-wave radiation, sensible and latent heat fluxes, but little or no improvement in long-wave radiation and momentum fluxes. Models with a simple urban representation (e.g., ‘slab’ schemes) generally perform well, particularly when combined with sophisticated hydrological/vegetation models. Some mid-complexity models (e.g., ‘canyon’ schemes) also perform well, indicating efforts to integrate vegetation and hydrology processes have paid dividends. The most complex models that resolve three-dimensional interactions between buildings in general did not perform as well as other categories. However, these models also tended to have the simplest representations of hydrology and vegetation. Models without any urban representation (i.e., vegetation-only land surface models) performed poorly for latent heat fluxes, and reasonably for other energy fluxes at this suburban site. Our analysis identified widespread human errors in initial submissions that substantially affected model performances. Although significant efforts are applied to correct these errors, we conclude that human factors are likely to influence results in this (or any) model intercomparison, particularly where participating scientists have varying experience

Today fund matching is a natural part of Swedes finance and is a popular form of savings that includes a large number of investors in the Swedish fund market. This in turn generates an increased interest in how portfolio managers should locate and acquire knowledge in portfolio selection. This gives a greater interest in how different investment strategies can be affected and generate an investors wealth to an increased level within the stock market, which gives an increased focus to be able to generate as high risk-adjusted return as possible. The study partly presents traditional theory and background on modern portfolio theory and the efficient market hypothesis. Empirical studies also present within the financial market that demonstrate the differences of opinion between how actively versus passively managed funds have performed and which investment strategy is most beneficial for investment. The purpose of the study is to compare realized return on active versus passive funds during long term, short term and specific time periods that had a lot of economic fluctuations, like bear markets. Within the study 10 actively managed funds and two index measures are selected to be studied and compared based on their respective performance, both within its rise and fall in the Swedish fund market. The performance measures will then be applied to be able to produce the results of the study and to be able to answer whether the active fund’s have any statistically significant over- and underperformance. After conducting single index models and t-test on the 10 active funds, the result of the study shows that despite using two benchmarks index, ten different active funds, long time period, short time period or specific time periods defined by market imbalance , we still resulted in many P-values that was not statistically significant. Active funds failed to overperform against passive funds, but passive funds also failed to outperform our selection of active funds.

Taking a better look at subjects of interest helps humans to improve confidence in their age estimation. Unlike still images, sequences offer spatio-temporal dynamic information that contains many cues related to age progression. A review of previous work on video-based age estimation indicates that this is an underexplored field of research. This may be caused by a lack of well-defined and publicly accessible video benchmark protocol, as well as the absence of video-oriented training data. To address the former issue, we propose a carefully designed video age estimation benchmark protocol and make it publicly available. To address the latter issue, we design a video-specific age estimation method that leverages pseudo-labeling and semi-supervised learning. Our results show that the proposed method outperforms image-based baselines on both offline and online benchmark protocols, while the online estimation stability is improved by more than 50%.

Background and Objectives: Finite element simulations are widely employed as a non-invasive and cost-effective approach for predicting outcomes in biomechanical simulations. However, traditional finite element software, primarily designed for engineering materials, often encountered limitations in contact detection and enforcement, leading to simulation failure when dealing with complex biomechanical configurations. Currently, a lot of model tuning is required to get physically accurate finite element simulations without failures. This adds significant human interaction to each iteration of a biomechanical model. This study addressed these issues by introducing PolyFEM, a novel finite element solver that guarantees inversion- and intersection-free solutions with completely automatic collision detection. The objective of this research is to validate PolyFEM's capabilities by comparing its results with those obtained from a well-established finite element solver, FEBio. Methods: To achieve this goal, five comparison scenarios were formulated to assess and validate PolyFEM's performance. The simulations were reproduced using both PolyFEM and FEBio, and the final results were compared. The five comparison scenarios included: (1) reproducing simulations from the FEBio test suite, consisting of static, dynamic, and contact-driven simulations; (2) replicating simulations from the verification paper published alongside the original release of FEBio; (3) a biomechanically based contact problem; (4) creating a custom simulation involving high-energy collisions between soft materials to highlight the difference in collision methods between the two solvers; and (5) performing biomechanical simulations of biting and quasi-stance. Results: We found that PolyFEM was capable of replicating all simulations previously conducted in FEBio. Particularly noteworthy is PolyFEM's superiority in high-energy contact simulations, where FEBio fell short, unable to complete over half of the

Governments are increasingly dependent on GovTech, which is the technology that facilitates processes in the public sector. Benchmarking the state of GovTech is done by governments and yields indispensable insights, which are used for optimising resource utilisation, identifying areas for improvement, and facilitating evidence-based policy prioritisation. Current benchmarking efforts are resource-intensive, time-intensive, and have limited scope, resulting in an inefficient assessment of GovTech innovation. This thesis explores the potential of Large Language Models (LLMs) to overcome the practical limitations of existing GovTech benchmarking methods, analysing the GovTech Maturity Index of the World Bank as a case study. Using an LLM and leveraging state-of-the-art techniques, including fine-tuning, Retrieval Augmented Generation, and Prompt Engineering, the usability of these models as an artefact for GovTech benchmarking is assessed. The results show that the best-performing model outperforms the random chance accuracy, indicating that the LLM not only understands the question and data format, but also contains the information to correctly answer the benchmark questions. The research concludes that the created artefact has the potential to improve GovTech benchmarking, resulting in more informed policy decision making. The thesis contributes valuable insights to the field of the GovTech research field by making GovTech benchmarking more efficient, leading to a better analysis of the current GovTech market, and contributing to the academic debate on GovTech market analysis. Furthermore, by streamlining the Dutch benchmarking process, resulting in more accurate insights, the study contributes to the advancement of GovTech solutions within the Netherlands, ultimately benefiting society as a whole. The artefact created in this research has significant policy relevance, as more efficient benchmarking allows policymakers to better o, Engineering and Policy Analysis

Aquest estudi ofereix una exploració integral sobre l'orientació d'hipergràfs dins de les vies metabòliques. El nostre focus va estar en la formulació d'un model AMPL, permetent als investigadors obtenir solucions efectives mitjançant l'optimització de ILP. Les principals contribucions del nostre treball inclouen: - Desenvolupament d'un model lògicament robust per resoldre el problema de l'orientació d'hiperarestes en vies metabòliques. - Avaluació comparativa del model respecte alternatives, identificant punts forts, punts febles i escenaris d'aplicació òptims. - Introducció de restriccions addicionals considerades significatives en el camp de les vies metabòliques, incloentí: - La prevenció de la inversió en reaccions específiques. - La identificació de certs vèrtexs com interns. - La identificació de certs vèrtexs com externs. A més, aquest article presenta una eina d'interfície gràfica d'usuari (GUI) amigable que facilita l'exploració del model, l'anàlisi de resultats, l'addició de restriccions i l'orientació de vies metabòliques en escenaris del món real., This study offers a comprehensive exploration into the orientation of hypergraphs within metabolistic pathways. Our focus was on formulating an AMPL model, enabling researchers to attain effective solutions through ILP optimization. Key contributions of our work encompass: - Development of a logically robust model for solving hyperedge orientation in metabolistic pathways. - Benchmarking the model against alternatives, identifying strengths, weaknesses, and optimal application scenarios. - Introduction of additional constraints deemed significant in metabolistic pathways, including: - Prevention of inversion in specific reactions. - Identification of certain vertices as internal. - Identification of certain vertices as external. Additionally, this paper introduces a user-friendly GUI tool that facilitates model exploration, result analysis, constraint addition, and pathway orientation in real-world scenarios.

In the framework of the IAEA Coordinated Research Project on In-Vessel Melt Retention, a benchmark of CFD simulations, devoted to thermal-hydraulic behavior of the light metal layer involves several research organizations: KTH of Sweden, SSTC NRS of Ukraine, ÚJV Řež of Czech Republic and CEA of France. This work aims at better simulating the focusing effect phenomenon leading to a heat flux peak along the height of the light metal layer, which is formed above the oxide layer in a stratified corium pool configuration during a PWR severe accident. This is a known safety issue compromising the reactor vessel integrity. The first benchmark step provides a solid foundation to the CFD schemes (physical models, meshes) by comparing the results of CFD simulations with thermal-hydraulic experimental data obtained using water as simulating fluid in a representative and quite laminar configuration. Then a similar but highly turbulent case, of higher height, is considered for more complex validation of the numerical simulation approach. Results with different turbulent models are compared against experimental data. On the strength of this encouraging work, a simulation of the same height configuration but considering steel fluid under severe accident conditions is foreseen at the final stage of this benchmark., Part of ISBN 9784888982566QC 20231214

Deep Learning for Computer Vision has great potential in intralogistics, for example for applications such as mobile robots or autonomous forklifts. However, the availability of labelled image datasets within this area is limited. To address this problem, we benchmarked two different datasets, LOCO (Logistics Objects in Context) and the TOMIE framework (Tracking Of Multiple Industrial Entities), to figure out, if these datasets can be combined to a single one. Therefore, we examine the usability of these datasets for Object Detection tasks using the YOLOv7 framework. For this we trained several Networks and compared them with each other. A deep analysis between these two datasets shows that they are very different and only suitable for specific tasks which are not interchangeable, despite having the same domain. Deeper Investigations are done to find the reasons for this. To close the Gap between LOCO and TOMIE, a synthetic data generation pipeline for Pallets is developed and 18 000 images are rendered. Furthermore, models are trained based on the synthetic data and compared with the models trained on real data. The synthetic data generation pipeline successfully closes the reality gap, and the performance on TOMIE is increased, but the performance on LOCO is significantly weaker. To develop a deeper understanding of this behavior we examine the underlying datasets and the reasons for the performance difference are identified.

MOTIVATION: Gaining structural insights into the protein-protein interactome is essential to understand biological phenomena and extract knowledge for rational drug design or protein engineering. We have previously developed DeepRank, a deep-learning framework to facilitate pattern learning from protein-protein interfaces using convolutional neural network (CNN) approaches. However, CNN is not rotation invariant and data augmentation is required to desensitize the network to the input data orientation which dramatically impairs the computation performance. Representing protein-protein complexes as atomic- or residue-scale rotation invariant graphs instead enables using graph neural networks (GNN) approaches, bypassing those limitations. RESULTS: We have developed DeepRank-GNN, a framework that converts protein-protein interfaces from PDB 3D coordinates files into graphs that are further provided to a pre-defined or user-defined GNN architecture to learn problem-specific interaction patterns. DeepRank-GNN is designed to be highly modularizable, easily customized and is wrapped into a user-friendly python3 package. Here, we showcase DeepRank-GNN's performance on two applications using a dedicated graph interaction neural network: (i) the scoring of docking poses and (ii) the discriminating of biological and crystal interfaces. In addition to the highly competitive performance obtained in those tasks as compared to state-of-the-art methods, we show a significant improvement in speed and storage requirement using DeepRank-GNN as compared to DeepRank. AVAILABILITY AND IMPLEMENTATION: DeepRank-GNN is freely available from https://github.com/DeepRank/DeepRank-GNN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Autoencoders are the backbones of many imputation methods that aim to relieve the sparsity issue in single-cell RNA sequencing (scRNA-seq) data. The imputation performance of an autoencoder relies on both the neural network architecture and the hyperparameter choice. So far, literature in the single-cell field lacks a formal discussion on how to design the neural network and choose the hyperparameters. Here, we conducted an empirical study to answer this question. Our study used many real and simulated scRNA-seq datasets to examine the impacts of the neural network architecture, the activation function, and the regularization strategy on imputation accuracy and downstream analyses. Our results show that (i) deeper and narrower autoencoders generally lead to better imputation performance; (ii) the sigmoid and tanh activation functions consistently outperform other commonly used functions including ReLU; (iii) regularization improves the accuracy of imputation and downstream cell clustering and DE gene analyses. Notably, our results differ from common practices in the computer vision field regarding the activation function and the regularization strategy. Overall, our study offers practical guidance on how to optimize the autoencoder design for scRNA-seq data imputation.

The Mean Opinion Score (MOS) serves as a subjective measure for assessing the quality of synthesized speech. Nevertheless, the conventional approach to MOS evaluations can be resource-intensive in terms of both time and cost. This article unveils an automatic MOS scoring toolkit that builds upon our success in securing the top position for some metrics in VoiceMos2022 Challenge and emerging as champions in some tracks of the VoiceMos2023 Challenge. We offer a pre-trained MOS scoring tool for English and provide training code for other languages. Our documentation, examples, and source code are available at https://github.com/superphysics/KyotoMOS.

In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094., Bjoern Menze is supported through the DFG funding (SFB 824, subproject B12) and a Helmut-Horten-Professorship for Biomedical Informatics by the Helmut-Horten-Foundation. Florian Kofler is Supported by Deutsche Forschungsgemeinschaft (DFG) through TUM International Graduate School of Science and Engineering (IGSSE), GSC 81. An Tang was supported by the Fonds de recherche du Québec en Santé and Fondation de l’association des radiologistes du Québec (FRQS- ARQ 34939 Clinical Research Scholarship – Junior 2 Salary Award). Hongwei Bran Li is supported by Forschungskredit (Grant NO. FK-21- 125) from University of Zurich., Peer Reviewed, Article signat per 109 autors/es: Patrick Bilic 1,a,b, Patrick Christ 1,a,b, Hongwei Bran Li 1,2,∗,b, Eugene Vorontsov 3,a,b, Avi Ben-Cohen 5,a, Georgios Kaissis 10,12,15,a, Adi Szeskin 18,a, Colin Jacobs 4,a, Gabriel Efrain Humpire Mamani 4,a, Gabriel Chartrand 26,a, Fabian Lohöfer 12,a, Julian Walter Holch 29,30,69,a, Wieland Sommer 32,a, Felix Hofmann 31,32,a, Alexandre Hostettler 36,a, Naama Lev-Cohain 38,a, Michal Drozdzal 34,a, Michal Marianne Amitai 35,a, Refael Vivanti 37,a, Jacob Sosna 38,a, Ivan Ezhov 1, Anjany Sekuboyina 1,2, Fernando Navarro 1,76,78, Florian Kofler 1,13,57,78, Johannes C. Paetzold 15,16, Suprosanna Shit 1, Xiaobin Hu 1, Jana Lipková 17, Markus Rempfler 1, Marie Piraud 57,1, Jan Kirschke 13, Benedikt Wiestler 13, Zhiheng Zhang 14, Christian Hülsemeyer 1, Marcel Beetz 1, Florian Ettlinger 1, Michela Antonelli 9, Woong Bae 73, Míriam Bellver 43, Lei Bi 61, Hao Chen 39, Grzegorz Chlebus 62,64, Erik B. Dam 72, Qi Dou 41, Chi-Wing Fu 41, Bogdan Georgescu 60, Xavier Giró-i-Nieto 45, Felix Gruen 28, Xu Han 77, Pheng-Ann Heng 41, Jürgen Hesser 48,49,50, Jan Hendrik Moltz 62, Christian Igel 72, Fabian Isensee 69,70, Paul Jäger 69,70, Fucang Jia 75, Krishna Chaitanya Kaluva 21, Mahendra Khened 21, Ildoo Kim 73, Jae-Hun Kim 53, Sungwoong Kim 73, Simon Kohl 69, Tomasz Konopczynski 49, Avinash Kori 21, Ganapathy Krishnamurthi 21, Fan Li 22, Hongchao Li 11, Junbo Li 8, Xiaomeng Li 40, John Lowengrub 66,67,68, Jun Ma 54, Klaus Maier-Hein 69,70,7, Kevis-Kokitsi Maninis 44, Hans Meine 62,65, Dorit Merhof 74, Akshay Pai 72, Mathias Perslev 72, Jens Petersen 69, Jordi Pont-Tuset 44, Jin Qi 56, Xiaojuan Qi 40, Oliver Rippel 74, Karsten Roth 47, Ignacio Sarasua 51,12, Andrea Schenk 62,63, Zengming Shen 59,60, Jordi Torres 46,43, Christian Wachinger 51,12,1, Chunliang Wang 42, Leon Weninger 74, Jianrong Wu 25, Daguang Xu 71, Xiaoping Yang 55, Simon Chun-Ho Yu 58, Yading Yuan 52, Miao Yue 20, Liping Zhang 58, Jorge Cardoso 9, Spyridon Bakas 19,23,24, Rickmer Br, Postprint (published version)

A state machine model can turn a complex behavioural system into a more accessible graphical model, and can improve the way people work with system design by making it easier to communicate and understand the system. The clear structure of a state machine model enables automatic generation of well structured, and consequently readable, and maintainable code. There are many known implementations and even plenty of commercial software available for working with state machines, and the goal of this project is to compare and discuss some of these implementations in the context of Ericsson's demands for run time, memory usage, scalability, readability, and maintainability. More specifically, the project focuses on the state machine models specified in the UML (the unified modeling language [15]), utilizing UML's associated markup language, XMI, to go from graphical model to generated C code. The resulting C code is primarily a code skeleton which only provides the basic behaviour of transitioning between states given a specific event, it is expected that the developers manually implement additional features themselves. The examined implementations are: Nested Switch, Array of Structs, Function Pointers, Basic State Pattern, State-Table Pattern, and Hierarchical State Pattern. Additionally, the project investigates how multiple state machines can communicate and work together as interacting state machines. And finally, to showcase how a state machine implementation can be maintainable, we develop an iterative code editor that can edit already operating and manually modified state machine implementations. The implementations are tested on a case study example provided by Ericsson, aimed to represent a sort of typical state machine design when it comes to number of states and events. The implementations are further tested with randomly generated state machines, to examine their scalability properties. In our opinion the results favour the Array of Structs and Basic State Pa

The fabrication of ship hulls involves welding operations, one typical operation is fillet welding of stiffeners to plates (panels) a so-called T-joint. The residual stresses from the welding process will influence the fatigue life of the structure, and residual deformations will increase the risk for buckling of the panel and give poor fabrication tolerances which may complicate the assembly of a built-up structure. / Though this weld geometry has been frequently studied in the literature both experimentally and numerically, the focus has been on the residual stress field and if residual deformations are studied, the angular distortion and the typical V-shape seen in the welded plate (panel) are reported. This benchmark focusses on the residual shape after two-pass fillet welding of a stiffener to a plate (T-joint and the influence of the constraints used during welding. The information of the full welding residual deformations may be useful for the fabrication of larger complex welded structures. / The benchmark study clearly shows that careful consideration is needed when defining a reference plane for comparison of resulting deformations from FE-simulations and from experiments as well as good documentation of the actual boundary conditions and steps carried after the completion of the welding and before measuring residual deformations. The FE-simulations have shown that it is difficult to obtain the full antisymmetric, twisted shape of the plate as measured in the experiments, although a trend similar to the experiments can be seen as the plate tends to move upwards at the end of each weld pass, i.e a slight influence of the weld pass direction.

El mercado de activos y criptomonedas ha crecido significativamente en los últimos años al igual que el número de jóvenes interesados en invertir y ganar dinero "fácil". Este aumento, sin embargo, no se ha acompañado de un incremento en formación, resultando en un gran número de errores evitables y la pérdida de grandes cantidades de dinero. Una metodología comúnmente utilizada para minimizar dichos riesgos consiste en consultar las redes sociales en busca de expertos que orienten o de identificar tendencias. El uso de la inteligencia artificial puede ser muy beneficioso para automatizar los pasos seguidos en esta metodología, en combinación con métricas como la cantidad de menciones o el sentimiento asociado a los comentarios de las redes sociales. En este proyecto se pretende desarrollar una aplicación capaz de obtener y procesar datos de redes sociales en tiempo real con la que fomentar la inversión consciente y no especulativa, valorando también el rendimiento de la base de datos utilizada y creando una página web donde los usuarios puedan consultar las distintas métricas generadas., The asset market has grown significantly in the last few years as well as the number of young people interested in investing and making "easy" money. This increase in demand, however, has not been met with an increase in formation or education, resulting in lots of avoidable mistakes and the loss of considerable amounts of money. A common approach on minimizing losses consists of using social media to look for the guidance of specialists or try to identify the latest trends. The use of artificial intelligence can be highly beneficial to automate the steps followed by this approach along with the use of metrics such as the number of mentions or the sentiment of the comments. The purpose of this project is to develop an app capable of obtaining and processing data from social media in real time, automating the process of identifying trends and promoting conscious investment against speculation. Besides the AI engine behind the app, this project also focuses on the efficiency and performance of the chosen database and the implementation of a webpage where users can see the processed data and generated metrics.

Kubernetes is a container orchestration system that has been rising in popularity in recent years. The modular nature of Kubernetes allows the usage of different storage solutions, and for cloud environments, cloud-native distributed storage solutions maybe attractive due to their redundant nature. There are many tools for cloud-native distributed storage available on the market today with differing features and performance. Choosing the correct one for an organisation can be difficult. Organisations utilising Kubernetes in cloud environments would like to be as performance efficient as possible to save on costs and resources. This study aims to offer a benchmark and analysis for some of the most popular tools, to help organisations choose the ‘best’ solution for their operational needs, from a performance perspective. The benchmarks compare three cloud-native distributed storage solutions, OpenEBS, Portworx, and Rook-Ceph on both Amazon Elastic Kubernetes Service (EKS) and Azure Kubernetes Service (AKS). For a baseline comparison, the study will also benchmark the cloud providers own solutions; Azure Disk Storage, and Amazon Elastic Block Storage. The study compares these solutions from three key metrics; bandwidth, latency, and IOPS, in both read and write performance., Det finns övrigt digitalt material (t.ex. film-, bild- eller ljudfiler) eller modeller/artefakter tillhörande examensarbetet som ska skickas till arkivet.There are other digital material (eg film, image or audio files) or models/artifacts that belongs to the thesis and need to be archived.

A state machine model can turn a complex behavioural system into a more accessible graphical model, and can improve the way people work with system design by making it easier to communicate and understand the system. The clear structure of a state machine model enables automatic generation of well structured, and consequently readable, and maintainable code. There are many known implementations and even plenty of commercial software available for working with state machines, and the goal of this project is to compare and discuss some of these implementations in the context of Ericsson's demands for run time, memory usage, scalability, readability, and maintainability. More specifically, the project focuses on the state machine models specified in the UML (the unified modeling language [15]), utilizing UML's associated markup language, XMI, to go from graphical model to generated C code. The resulting C code is primarily a code skeleton which only provides the basic behaviour of transitioning between states given a specific event, it is expected that the developers manually implement additional features themselves. The examined implementations are: Nested Switch, Array of Structs, Function Pointers, Basic State Pattern, State-Table Pattern, and Hierarchical State Pattern. Additionally, the project investigates how multiple state machines can communicate and work together as interacting state machines. And finally, to showcase how a state machine implementation can be maintainable, we develop an iterative code editor that can edit already operating and manually modified state machine implementations. The implementations are tested on a case study example provided by Ericsson, aimed to represent a sort of typical state machine design when it comes to number of states and events. The implementations are further tested with randomly generated state machines, to examine their scalability properties. In our opinion the results favour the Array of Structs and Basic State Pa

A profound understanding of the physicochemical properties of organic crystals is crucial for topics from material science to drug discovery. Using molecular dynamics (MD) simulations with a sufficiently accurate force field, microscopic insight into structure and dynamics can be obtained of materials, including liquids and biomolecules. They are a valuable complement to experimental investigations that are used routinely in drug design, but not very often for studies of organic crystals. Indeed, the often delicate interactions in organic crystals act as a sensitive probe to investigate the accuracy of force fields. Here, we study the structural, dynamic, and thermodynamic properties of 30 organic crystals using the popular general AMBER force field (GAFF). In particular, we investigate both solid-solid and solid-liquid phase transitions. Melting points were determined using extensive solid-liquid coexistence simulations. For many compounds, we detect a phase transition from an ordered to a plastic crystal in the simulations. Based on the translational and rotational dynamics of the compounds, we can rationalize the properties of the plastic crystal phase. MD simulations can therefore help to answer the important question of whether or not organic crystals have a plastic crystal phase, and if so, what are the underlying factors in the molecular structure determining that.

A profound understanding of the physicochemical properties of organic crystals is crucial for topics from material science to drug discovery. Using molecular dynamics (MD) simulations with a sufficiently accurate force field, microscopic insight into structure and dynamics can be obtained of materials, including liquids and biomolecules. They are a valuable complement to experimental investigations that are used routinely in drug design, but not very often for studies of organic crystals. Indeed, the often delicate interactions in organic crystals act as a sensitive probe to investigate the accuracy of force fields. Here, we study the structural, dynamic, and thermodynamic properties of 30 organic crystals using the popular general AMBER force field (GAFF). In particular, we investigate both solid-solid and solid-liquid phase transitions. Melting points were determined using extensive solid-liquid coexistence simulations. For many compounds, we detect a phase transition from an ordered to a plastic crystal in the simulations. Based on the translational and rotational dynamics of the compounds, we can rationalize the properties of the plastic crystal phase. MD simulations can therefore help to answer the important question of whether or not organic crystals have a plastic crystal phase, and if so, what are the underlying factors in the molecular structure determining that.

A profound understanding of the physicochemical properties of organic crystals is crucial for topics from material science to drug discovery. Using molecular dynamics (MD) simulations with a sufficiently accurate force field, microscopic insight into structure and dynamics can be obtained of materials, including liquids and biomolecules. They are a valuable complement to experimental investigations that are used routinely in drug design, but not very often for studies of organic crystals. Indeed, the often delicate interactions in organic crystals act as a sensitive probe to investigate the accuracy of force fields. Here, we study the structural, dynamic, and thermodynamic properties of 30 organic crystals using the popular general AMBER force field (GAFF). In particular, we investigate both solid-solid and solid-liquid phase transitions. Melting points were determined using extensive solid-liquid coexistence simulations. For many compounds, we detect a phase transition from an ordered to a plastic crystal in the simulations. Based on the translational and rotational dynamics of the compounds, we can rationalize the properties of the plastic crystal phase. MD simulations can therefore help to answer the important question of whether or not organic crystals have a plastic crystal phase, and if so, what are the underlying factors in the molecular structure determining that.