• 2019-2023close
• Open Access close
• Publications close
• Research data close
• Research software close
• SE close
• Publikationer från KTH close

This report is about a novel approach to attenuation of fan noise in aerial vehicles, by way of implementing a ducted fan in the chassis of a four meter blended wing body plane. Three different one meter PVC pipes were used and their performances as silencers were tested by measuring the sound power level and calculating the insulation loss compared to a fan by itself. The ducts were either empty or lined with acoustic absorbents and micro perforated panels. Experiments were carried out in the reverberation room at KTH Marcus Wallenberg laboratory for sound and vibration research using the guidelines in ISO 3741 (2010). The results showed that the empty duct lead to a 15.3 dB(A) insulation loss with no decrease in thrust from the fan. The absorbent and micro perforated panel, however, lead to a 22.7 dB(A) insulation loss while giving a major decrease in thrust of more than one order of magnitude. The results show the failure of implementation of the latter two silencers due to choking, but also the success of the empty duct. This shows that there is room for improvement and perhaps even a future possibility of a successful implementation in a real vehicle.

The Examination Timetabling Problem (ETP) is the problem of scheduling a number of exams during a set time period so that no students are required to sit two exams simultaneously. Despite the complexity of the problem, universities all over the world solve ETPs several times each year. Two known methods for solving ETPs is using either heuristics or Hopfield Neural Networks (HNN). This thesis compares the performance of a heuristic algorithm implemented with Local Search, Simulated Annealing and Tabu Search to the performance of a HNN algorithm. Both algorithms were executed on ten different ETPs reduced to Graph Colouring Problems (GCP). The results show that the heuristic algorithm always generated more satisfactory solutions to the ETPs than the HNN. The HNN was, however, implemented as software in this thesis. It is intended to be implemented as hardware and if this method were to have been used instead the HNN algorithm might have produced other results. At this stage the heuristic algorithm is more suitable than the HNN algorithm for solving ETPs. Schemaläggningsproblem för examinationer (ETP) syftar på problemet att schemalägga ett antal examinationer under ett bestämt tidsintervall så att ingen student behöver närvara på flera examinationer samtidigt. Det är ett komplext problem som universitet världen över behöver lösa flera gånger per år. Två kända metoder för att lösa ETPs är användning av antingen heuristiker eller Hopfield Neural Networks (HNN). Den här uppsatsen jämför prestandan av en heuristik implementerad med Lokal Sökning, Simulerad Härdning och Tabusökning med presentandan av en HNN-algoritm då båda metoderna exekveras på tio ETPs reducerade till Graffärgningsproblem (GCP). Resultaten visar att heuristiken alltid genererade mer tillfredsställande lösningar till schemaläggningsproblemen än HNN-algoritmen gjorde. HNN-algoritmen, som egentligen bör implementeras som hårdvara, implementerades dock som mjukvara i den här avhandlingen. Hade den implementerats som hårdvara istället hade kanske andra resultat producerats. För tillfället lämpar sig heuristiken bättre än HNN-algoritmen för att lösa ETPs.

QC 20210525

This thesis report revolved around the cloning process of plasmids. Attempts of cloning the red fluorescent protein mCherry, and the green fluorescent protein EGFP from various plasmids, into other plasmids containing different cell-junction/cytoskeleton plasmids were made. These plasmids were first amplified using PCR, and then cloned using Gibson-Assembly, and then transfected into live HEK293T or MDCK-II cells. After the transfection, the cells were examined in a microscope. The results showed no signal or localization for the cloned plasmids in their respective corresponding channel, 561 nm for the red fluorescent protein mCherry or 488 nm for the green fluorescent protein EGFP. The step that went wrong was the PCR step in the cloning process, since the backbone vector was not successfully amplified. The reasons for this was either that the backbone vector was too long, the primers regions were to rich with Guanine and Cytoseine, or the primers being too long. Den här tesen kretsade kring kloningsprocessen för plasmider. Det gjordes försök att från plasmider klona in det röda fluorescerande proteinet mCherry, samt det gröna fluorescerande proteinet EGFP in i andra plasmider som innehöll olika cell-junction proteiner. Både det fluorescerande fragmenten och plasmid-vektorerna innehållande cell-junction proteinerna amplifierades med PCR. Sedan gjordes Gibson-Assembly som var själva kloningsmetoden. Efter det transfekterades HEK293T, samt MDCK-II celler med lösningen från Gibson-Assembly kloningen. Dessa celler undersöktes sedan i mikroskop. Resultatet visade inga tydliga signaler varken i 561 nm kanalen (mCherry), eller i 488 nm kanalen (EGFP), vilket betyder att kloningen inte fungerade. Steget som gick fel var PCR-steget i själva kloningsprocessen, då plasmid-vektorerna inte amplifierades. Anledningen till detta var antingen att själva plasmid-vektorerna var för långa, primer regionerna hade för mycket Guanin och Cytosin, eller att alla primers själva var för långa.

We present an online API to access a number of Natural Language Processing services developed at KTH. The services work on Swedish text. They include tokenization, part-of-speech tagging, shallow parsing, compound word analysis, word inflection, lemmatization, spelling error detection and correction, grammar checking, and more. The services can be accessed in several ways, including a RESTful interface, direct socket communication, and premade Web forms. The services are open to anyone. The source code is also freely available making it possible to set up another server or run the tools locally. We have also evaluated the performance of several of the services and compared them to other available systems. Both the precision and the recall for the Granska grammar checker are higher than for both Microsoft Word and Google Docs. The evaluation also shows that the recall is greatly improved when combining all the grammar checking services in the API, compared to any one method, and combining services is made easy by the API.

This thesis focuses mainly on nonlinear dimensionality reduction. A comparison of a number of different extensions of the Isomap-algorithm found in the literature is presented as well as an introduction to the underlying theory. Further, two novel extensions of the algorithm are introduced. Residual variance is used as a means of quantifying performance and optimizing parameters. One of the novel approaches, NNL-Isomap, is applied to financial data sets such as the S&P 500, the results are compared with those from PCA and, when appropriate, well-known academic models. The thesis further introduces a selection of concepts from topological data analysis together with corresponding short explanatory examples as well as applies these concepts to some simple financial examples. Denna uppsats huvudfokus kretsar kring ickelinjär dimensionsreducering. En jämförelse av ett antal olika kända utvidgningar av Isomap-algoritmen ges, samt en introduktion till den bakomliggande teorin. Vidare presenteras två egenutvecklade utvidgningar av algoritmen. För att mäta prestanda samt optimera parametrar används residual variance-metoden. En av de egenutvecklade utvidgningarna, NNL-Isomap, appliceras på finansiella dataset såsom från S&P 500, och dess resultat jämförs med både det av PCA samt välkända akademiska modeller. Ett urval av begrepp härstammandes från topologisk dataanalys introduceras tillsammans med ett par korta exempel, dessa begrepp appliceras sedan på enklare finansiella exempel.

Envisioning an inner-city student housing project, adding life to an office-dominated area, and offering a publicly accessible roof terrace and open landscape park. We are proposing a student housing complex, composed of two 355-metres-long buildings with two and three floor levels, respectively, lifted up on pilotis, with corridor rooms and student apartments accomodating a total of 336 students and thus creating a bustling hub of human interactivity. For the proposal, we claim a large strip of land in the southern part of the property “Kvarteret Garnisonen 3” in Östermalm, Stockholm; the southern-most 15 % of the 84 000-square-metre property, previously used as a launching space for ceremonial marches, but currently used mostly for parking spaces and dog walking. Designing the large-scale structure to house many students, with each contributing to the added life to the space, and encouraging human interactions architecturally, by retaining the accessibility of the ground level by way of pilotis, as well as making the roof terrace open to the public, and offering several alternative routes between the levels of the buildings.

Introduction: Concussions in American Football are of a major concern due to highly reported injury rates. The importance of properly designed helmets have shown effect in reducing the risk of injuries, such as skull fractures. However, they are not as effective in reducing the risk of concussion. Helmets designed are required to pass standards and regulations for them to be allowed within the football leagues. The current test methods evaluate linear impacts, but lack evaluations of oblique impacts which are believed to cause concussions. Several test methods have been suggested, but little is known regarding how they compare. Objective: The purpose of this study was to compare three different test methods for evaluating helmet performance, utilizing finite element simulation. Three different helmet models were used for comparison, evaluating head kinematics. The helmet models were additionally ranked from best to worst based on their performances. Method: Three test methods, linear impactor, 45° angled linear impactor, and a drop test onto a 45° angled plate were simulated with three different open source helmet models. Simulations were conducted with one impact velocity at three impact locations. The influence of the interaction between helmet and head was also evaluated by altering the friction coefficient. Results: The test methods showed different results depending on helmet models, impact locations, and kinematics evaluated. Similarly, rankings of the helmets were varied based on methods and impact location. Little difference was observed after lowering the friction coefficient in majority of cases. The linear and angular acceleration for the drop side impact were mostly affected. Conclusion: Further evaluations of the test methods and comparison to real impacts is required to evaluate what method resembles head impacts best. Lowered friction coefficient had an effect for the drop impacts, but minor effect for other test methods

I takt med att tekniken utvecklas gör även underhållsstrategierna det. Trots detta är det många företag som fortfarande använder samma underhållsstrategier som man alltid gjort. Olika underhållsstrategier ser behovet av underhåll på olika sätt, allt från att vänta tills felet uppstår till att förebygga fel genom avancerad dataanalys och det kan därför vara svårt för underhållsorganisationer att välja rätt strategi. Detta arbete undersöker hur ett byte av underhållsstrategi kan påverka effektiviteten av underhållet i en större organisation, med fokus på kostnader, personalresurser och tillgänglighet på utrustning. PostNord har valts som fallstudie. De strategier som jämförs och utvärderas är; avhjälpande underhåll, förebyggande underhåll, tillståndsbaserat underhåll samt prediktivt underhåll. Resultatet påvisar ökad effektivitet för tre av fyra strategier, som också ser till företagets behov. Den efterföljande diskussionen av resultatet visar att det framtida behovet är oklart där främst två faktorer som påverkar; årliga volymminskningar samt morgondagens utdelningsmodell. Slutsatsen blir att det finns två olika rekommendationer beroende på hur det framtida behovet ter sig. Om det är ett fortsatt behov av hög tillgänglighet på maskinerna så rekommenderas en kombination av prediktivt underhåll och förebyggande underhåll på drifttid. Om det framtida behovet däremot påvisar att tillgängligheten inte kommer att påverka beslutet i samma grad rekommenderas en kombination av förebyggande underhåll på drifttid och tillståndsbaserat underhåll där tillståndsmätningar görs mot den data som samlas in idag. I båda rekommendationerna räknas fortfarande avhjälpande underhåll in som strategi för hantering av de akuta stopp som kan uppstå. Att titta på andra maskintyper där det framtida behovet framgår tydligare föreslås som fortsatt arbete. As technology evolves, so do maintenance strategies. Despite this, many companies tend not to change maintenance strategy. Maintenance strategies differ significantly in the way they address the need for maintenance, for some the strategy is to provide maintenance after a fault has occurred while others focus on preventing faults through advanced data analysis. With this diversity in approach, it may be difficult for maintenance organizations to choose the right strategy. This thesis examines how a change of maintenance strategy can affect the efficiency of maintenance in a larger organization, focusing on costs, personal resources and availability of the equipment. As a case study, PostNord has been the selected company for the study. The study compares and evaluates four maintenance strategies; corrective maintenance; preventive maintenance; condition-based maintenance; and predictive maintenance. The results show increased efficiency for three out of four strategies, which also meet the needs of the company. In the discussing the results in the context of PostNord, annual volume reductions and future distribution model, are identified as the two main factors that will affect the future validity and application of results. Subsequently, the conclusion consists of two recommendations depending on distribution model and annual volumes. If there is a continued need for high availability in the machines, then the combination of predictive maintenance and preventative maintenance on operating time is recommended. If the future need does impact on the need of availability, then a combination of preventive maintenance on operating time and condition-based maintenance where measurements are made against the data that are accessible today is recommended. Both recommendations still include corrective maintenance as a strategy for managing the emergency stops that may occur. To focus on machines with a clearer future need are recommended for continued work.