Alkali-kiselsyrareaktionen (AKR) är en kemisk reaktion som medför att spänningar uppstår i betongen då den bildade silikatgelen expanderar. Reaktionen kräver tillräcklig hög alkalihalt, reaktiv ballast samt vatten. På grund av de AKR-inducerade spänningarna är det av intresse att få kunskap i hur reaktionen påverkar betongens böjmomentkapacitet, förankring samt skjuvnings- och genomstansningskapacitet. För att besvara frågeställningen har därför en omfattande litteraturinventering tillsammans med tre kompletterande intervjuer utförts. Resultat som har erhållits, påvisar att två huvudsakliga effekter fås av AKR. Utöver en reducerad hållfasthet på grund av expansioner och sprickbildningar, erhålls även en gynnsam förspänningseffekt när armering motverkar expansioner. Med hänsyn till dessa huvudeffekter påverkas betongs bärighet i olika stor utsträckning beroende på expansionens storlek samt armeringens läge och typ i tvärsnittet. Resultaten tyder på att draghållfastheten reduceras till 40 % medan tryckhållfastheten reduceras till 60 % vid 5 mm/m expansion. För böjkapaciteten har ingen större reduktion uppmäts då expansioner understiger 6 mm/m. Emellertid har en reduktion på 25 % observerats vid expansioner större än 6 mm/m. Skjuvkapaciteten i AKR-skadad betong beror till stor del av byglars närvaro samt expansionens omfattning. I balkar utan byglar reduceras skjuvkapaciteten med 15-25 % för slät armering och 20-30 % för räfflad armering. Genomstansningskapaciteten i ett dubbelarmerat betongtvärsnitt reduceras obetydande till dess att expansionen överstiger 6 mm/m varpå en mer påtaglig reduktion fås. Hållfastheten för vidhäftning minskar med ca 40 % då täckande betongskikt understiger 1.5Æ och/eller att inga byglar är närvarande. Om byglar är närvarande samt om täckande betongskikt överstiger 4Æ visas inga tecken på försämrad vidhäftningshållfasthet. Utöver en minskad bärighet, öppnar AKR upp betong och skapar gynnsammare förutsättningar för rost-och frostangrepp. / Alkali-silica reaction (ASR) is a chemical reaction that causes stresses in concrete when the produced alkali silica gel expands. The reaction requires sufficiently high alkali content, reactive aggregates and water. Due to ASR-induced stresses it is of interest to gain insight in how ASR affects the concrete’s bending capacity, anchoring together with shear- and punching shear capacity. An extensive literature review was therefore carried out together with three complementary interviews in order to answer the question at issue. Obtained results show two main effects of ASR. In addition to a reduced strength because of cracking and expansion, an advantageous pre-stress is gained due to restraint to expansion. Thus, the load-carrying capacity of concrete is affected in various extents depending on the size of expansion and location and type of the reinforcement. The results indicate that the tensile strength is reduced to 40 % whereas the compressive strength is reduced to 60 % at 5 mm/m expansions. While expansions lesser than 6 mm/m has not shown a greater reduction of the bending capacity, a reduction of 25 % has been observed in concrete with expansions greater than 6 mm/m. The shear capacity of an ASR-affected concrete structure depends mainly on the presence of links and the extent of expansions. In beams without links, shear capacity is reduced to 15-25 % for smooth bars and 20-30 % for ribbed bars. The punching strength of a concrete structure reinforced in both faces is not reduced until expansions exceed 6 mm/m, whereas a more significant reduction is obtained. The bond strength decreases by about 40 % if the concrete cover is less than 1.5 Æ and/or if no links are present. Meanwhile if links are present and if concrete cover is more than 4Æ, no signs of reduction of the bond strength has been observed. Additionally, to a reduced load-carrying capacity, ASR also opens up the concrete and consequently creates beneficial circumstances for corrosion and frost attacks.

Over the last decades, researchers have proposed a number of tools for the condition assessment of concrete infrastructure affected by alkali-silica reaction (ASR). Amongst those, increasing attention has been given to the Stiffness Damage Test (SDT), Damage Rating Index (DRI), and Residual Expansion (RE) laboratory test procedures that aim to determine the cause and extent (i.e., diagnosis) of damage along with the potential of further deterioration (i.e., prognosis) of affected concrete. Yet, most of the data gathered so far while using the aforementioned tools has been obtained on laboratory test specimens presenting distinct conditions from affected structural members in the field, especially regarding restraint effects. This work aims to understand the impact of restraint on ASR-induced expansion and damage. Thirty-two 450 mm by 450 mm by 675 mm concrete blocks with various reinforcement configurations (i.e., unreinforced, 1D and 2D reinforcement) and incorporating highly reactive coarse and fine aggregates (i.e., Springhill coarse and Texas sand) were manufactured and stored in conditions enabling ASR-induced development (i.e., 38°C and 100 R.H). Two expansion levels were selected for analysis (i.e., 0.08% and 0.15%); once reached, cores were extracted from three different directions (i.e., longitudinal, transversal and vertical) of all blocks and mechanical (i.e., SDT and compressive strength), microscopic (i.e., DRI, scanning electron microscope, etc.) and expansion (i.e., RE) test procedures were conducted on the concrete cores. Results suggest that the presence of restraint influences the induced expansion, resulting in an anisotropic response of the specimens. Furthermore, similar to the expansion behavior, an anisotropic distribution of induced damage and mechanical properties reduction are observed for the restrained concrete blocks in which the restraint configuration seems to significantly affect ASR-induced damage development and features. This led to the observation of a higher number of damage features, ASR development and mechanical properties reduction in cores obtained from unrestrained directions. Yet, some anticipated results from the current research will be studied in detail in the near future where the reliability of the existing techniques (i.e., residual expansion and soluble alkalis) for appraising ASR potential for further induced development and distress (i.e., prognosis) in affected concrete presenting distinct restraint scenarios will be evaluated.

Patientnämndens förvaltning i Stockholm tar årligen emot tusentals samtal som önskar framföra klagomål på vården i Region Stockholm. Syftet med arbetet är att undersöka hur en NLP-robot för klassificering av inkomna klagomål skulle kunna bidra till en ökad effektivitet av verksamheten. Klassificeringen av klagomålen har utförts med hjälp av en metod baserad på Support Vector Machines. För att optimera modellens korrekthet undersöktes hur längden av ordvektorerna påverkar korrektheten. Modellen gav en slutgiltig korrekthet 53,10 %. Detta resultat analyserades sedan med målsättningen att identifiera potentiella förbättringsmöjligheter hos modellen. För framtida arbeten kan det därför vara intressant att undersöka hur antalet samtal, antalet personer som spelar in samtal och klassfördelningen i datamängden påverkar korrektheten. För att undersöka hur effektiviteten hos Patientnämndens förvaltning i Stockholm skulle påverkas av implementeringen av en NLP-robot användes en SWOT-analys. Denna analys visade på tydliga fördelar med automatisering av klagomålshanteringen, men att en sådan implementation måste ske med försiktighet där det säkerställs att tillgången på kompetens är tillräcklig för att förebygga potentiella hot. / Every year Patientnämnden recieves thousands of phone calls from patients wishing to make complaints about the health care in Stockholm. The aim of this work is to investigate how an NLP-robot for classification of recieved phone calls would contribute to an increased efficiency of the operation. The classification of the complaints has been made using a method based on Support Vector Machines. In order to optimize the accuracy of the model the impact of the length of the word vector has been investigated. The final result was an accuracy of 53.10%. The result was analyzed with the goal to identify potential opportunities of improvement of the model. For future work it could be interesting to investigate in how the number of calls, the number of people recording the calls and the distribution between the classes affect the accuracy A SWOT-analysis was performed in order to investigate in how the efficiency of Patientnämnden would be affected by the implementation of an NLP-robot. The analysis showed apparent benefits of automation of complaint management, but also that such an implementation must be done with great caution in order to be able to ensure that the available competence is high enough to prevent potential threats.

The importance of Automatic Speech Recognition (ASR) Systems, whose job is to generate text from audio, is increasing as the number of applications of these systems is rapidly going up. However, when it comes to training ASR systems, the process is difficult and rather tedious, and that could be attributed to the lack of training data. ASRs require huge amounts of annotated training data containing the audio files and the corresponding accurately written transcript files. This annotated (labeled) training data is very difficult to find for most of the languages, it usually requires people to perform the annotation manually which, apart from the monetary price it costs, is error-prone. A supervised training task is impractical for this scenario.  The Arabic language is one of the languages that do not have an abundance of labeled data, which makes its ASR system's accuracy very low compared to other resource-rich languages such as English, French, or Spanish. In this research, we take advantage of unlabeled voice data by learning general data representations from unlabeled training data (only audio files) in a self-supervised task or pre-training phase. This phase is done by using wav2vec 2.0 framework which masks out input in the latent space and solves a contrastive task. The model is then fine-tuned on a few amounts of labeled data. We also exploit models that have been pre-trained on different languages, by using wav2vec 2.0, for the purpose of fine-tuning them on Arabic language by using annotated Arabic data.   We show that using wav2vec 2.0 framework for pre-training on Arabic is considerably time and resource-consuming. It took the model 21.5 days (about 3 weeks) to complete 662 epochs and get a validation accuracy of 58%.  Arabic is a right-to-left (rtl) language with many diacritics that indicate how letters should be pronounced, these two features make it difficult for Arabic to fit into these models, as it requires heavy pre-processing for the transcript files. We demonstrate that we can fine-tune a cross-lingual model, that is trained on raw waveforms of speech in multiple languages, on Arabic data and get a low word error rate 36.53%. We also prove that by fine-tuning the model parameters we can increase the accuracy, thus, decrease the word error rate from 54.00% to 36.69%.

Pressure is mounting in the concrete industry to adopt eco-efficient methods to reduce CO₂ emissions. Portland cement (PC), an essential concrete ingredient, is responsible for over two-thirds of the embodied energy of the concrete, generating about 8% of global greenhouse gas emissions. Extraction and transportation of aggregates and raw materials that comprise concrete mixes are also directly linked to their embodied energy; thus, recycled concrete aggregates (RCA) have been proposed as a promising alternative to increase sustainability in new construction. In this context, many studies have been conducted over the past decades on the properties of RCA concrete. Recent studies have shown that suitable fresh (i.e., flowability) and short-term hardened (i.e., compressive strength) properties might be achieved when the unique microstructural features of RCA are accounted for in the mix-design process of the recycled concrete. However, manufacturing RCA from construction demolition waste (CDW) or returned concrete (RC) presents its unique challenges. Amongst others, the variation in the source of RCA and the presence of damage due to several deterioration mechanisms causes major concern. Due to the presence of reactive aggregates in many quarries in Canada, alkali-silica reaction (ASR) is one of the most common deterioration mechanisms. The durability and long-term performance of RCA concrete are not fully understood and should be further investigated, especially in regards to a) the potential of further (secondary) deterioration of recycled concrete bearing coarse and fine alkali-silica reactive aggregates b) the impact of the severity of the initial reaction on mechanical properties and kinetics of expansion in recycled concrete and c) the impact of using sound and alkali-silica reaction (ASR) affected RCA on the chloride diffusivity (and thus corrosion initiation) of concrete. This work aims to appraise the durability performance of RCA concrete made of 100% coarse RCA, particularly two families of RCA selected (i.e., returned concrete RCA, demolished concrete RCA) to represent waste currently being generated. Furthermore, two types of reactive aggregates are selected to investigate the impact of the source of the reaction (i.e. reactive coarse aggregate as original virgin aggregate - OVA and reactive sand within the residual mortar - RM) within the RCA. ASR is the distress mechanism used to introduce damage to the manufactured RCA. A new mix design technique was used to produce recycled concrete mixtures to increase eco-efficiency, improve fresh-state properties, and reduce cement use in RCA concrete. In conclusion, the initial reaction's location and severity significantly impact the compressive strength, SDT parameters, chloride diffusion rate, and shear strength of concrete specimens. Specifically, the location of the initial reaction can influence the distribution and extension of damage within the various parts of recycled concrete, while the severity of the initial reaction can affect the overall integrity of the aggregates as well as the availability of silica and alkalis for secondary reaction. These results demonstrate the importance of assessing the severity of the initial reaction and its source in order to ensure the durability and long-term performance of recycled concrete made with reactive RCA.

No description available.

No description available.

No description available.

No description available.

<p>This study investigates the feasibility of using waste glass as a supplementary cementitious material (SCM). By further defining some of the parameters by which waste glass may be incorporated into concrete as a cement replacement, the environmental, economical, and engineering benefits of this material may be realized. Past observations, including the production of alkali silica reaction (ASR) gel, and the lack of pozzolanic reactivity, have limited the acceptance of waste glass as a SCM,</p> <p>Mechanical treatment was used to improve reactivity and provide a particle size at which waste glass performs comparably to ground granulated blast furnace slag and nearly as well as silica fume. At 6.6 µm, the pozzolanic reactivity of waste glass was demonstrated through consumption of Ca(OH)₂ and heat of hydration. Waste glass at a larger particle size (16.5 µm) was as reactive as slag. Use of waste glass at 10% replacement of Portland cement by mass and at a particle size below 100 µm proved useful as a SCM.</p> <p>A relationship between pozzolanic and alkali silica reaction (ASR) was identified with intermediate phases of the reaction present. Calcium silicate hydrates (C-S-H) from the pozzolanic reaction have a Ca/Si ratio of 1.5-2. ASR products generally have a Ca/Si ratio of 0.01-1. The products observed with agglomeration of glass particles had a Ca/Si ratio from 0.5-2. The affects of silica concentration and alkalinity of the solution on the reaction products were explored.</p> <p>A reaction rim was identified around glass agglomerates where fluorescence was observed. The results indicate that ASR can be induced even in low alkalinity cement, and the rate of reaction influences both the characteristics and composition of the reaction product.</p> / Doctor of Philosophy (PhD)