Det har av olika anledningar blivit mer populärt att bygga flervåningshus i trä vilket har lett till en utveckling av material i trä. Korslimmat trä även kallat KL-trä är en konsekvens av denna utveckling i Sverige och precis som för andra stommaterial används datorprogram för att rita hus i KL-trä. Syftet med denna rapport är att få lärdom om KL-trä samt undersöka hur programmet FEM-Design kan användas för att dimensionera ett flervåningshus i KL-trä. Därtill, har FEM-Design också använts för att räkna materialmängden för att kunna göra en klimatdeklaration, för båda materialen betong och trä. Resultatet av klimatdeklarationen visar att träbyggnaden får ett lägre utsläpp i byggskedet jämfört med betongbyggnaden i byggskedet, men Boverkets databas som har använts i denna rapport är något konservativ. Värdena i Boverkets databas i framför allt skede A1-A3 avviker något jämfört med EPD:er (miljövarudeklaration) från olika företag inom betongindustrin. KL-träbyggnaden är lättare än betongbyggnaden vilket gör att betongfundamenten till träbyggnaden borde kunna göras mindre för att klara av lasterna jämfört med betong. Däremot får träbyggnaden något tjockare bjälklag för att klara av ljud-, brand- och brukskraven som då medför en lägre rumshöjd om inte förändring görs av den totala byggnadshöjden. För att uppnå samma rumshöjd på 2,6 m i betongbyggnaden skulle det krävas att träbyggnaden byggs 1,2 m högre vilket leder till större materialåtgång. / For various reasons, it has become more popular to build multi-storey houses in wood, which has led to the development of materials in wood. Cross laminated timber also called CLT is a consequence of this development in Sweden and just like other frame materials, computer programs are used to draw houses in CLT. The aim of this report is to get knowledge of CLT and investigate how the software FEM-Design can be used to design a multi-storey building in CLT. In addition, FEM-Design has also been used to calculate the amount of material necessary to make a climate declaration, for both materials concrete and wood. The result based on the climate declaration shows that the wooden building has a lower emission in the construction phase compared with the concrete building, but the National Board of Housing, Building and Planning's database that has been used in this report is somewhat conservative. The values in the National Board of Housing, Building and Planning's database, primarily in stages A1-A3, differ slightly compared with EPDs (environmental product declaration) from different companies in the concrete industry. The CLT building is lighter than the concrete building, which means that the concrete foundations for the wooden building may be made smaller to handle the loads compared to concrete. On the other hand, the wooden building gets a slightly thicker floor to cope with the sound, fire and use requirements, which then results in a lower room height if the total height of the building still is the same. In order to achieve the same room height of 2.6 m in the concrete building, it would be required that the wooden building be built 1.2 m higher, which leads to greater material consumption.

De senaste siffrorna från Boverkets klimatindikatorer visade att byggbranschen, år 2016, stod för ett utsläpp av 21 Mton CO2e (koldioxidekvivalenter). Det motsvarade 18 % av de totala nationella utsläppen. I mätningar ett par år tidigare (2012) framgick att privata lokaler utgjorde 20 % av Sveriges totala bygginvesteringar vilket var i samma storleksordning som investeringarna i flerbostadshus. Ett stort fokus har de senaste åren legat på att effektivisera flerbostadshusens produktions- och bruksskede medan marknaden för hallbyggnader fortfarande domineras (ca 80 % marknadsandel) av klimatbelastande stålstommar och stora mängder mineralull utan att få ett liknande fokus.   I en förstudie som baserades på intervjuer och en enkät riktade till branschkontakter med erfarenheter av uppförda hallbyggnader framgick att beställare och entreprenörer upplevde svårigheter att dels uppfylla BBR:s krav på ljud, brand och fukt mm. med en träkonstruktion men även svårigheter att motivera trä som stommaterial ekonomiskt. Förstudien visade även på att beställare och entreprenörer viktar kostnadsrelaterade parametrar primärt och klimatrelaterade parametrar sekundärt.   Med utgångspunkt i att använda mindre klimatbelastande material i byggprocessen syftar studien dels till att utreda om och hur trä kan konkurrera med stål som stommaterial vid nyproduktion av en enplans hallbyggnad med lagerverksamhet samt att ge en helhetsbild av hur ett byte från en stålstomme till en trästomme kan påverka byggnaden med systemgränserna satta på utsidan av klimatskalet inkl. grundkonstruktionen. Till följd det ökande intresset av KL-trä (korslaminerat trä) som stommaterial syftar studien även till att utreda vilka effekter ett inslag av KL-trä i stomkonstruktionen kan ha och hur materialet lämpligast kan implementeras.   Med utgångspunkt i ett reellt referensobjekt, som utgjordes av en enplans hallbyggnad om 1008 m2 med lagerverksamhet och en lätt stomme av stål, utfördes i studien en jämförelse mot en fiktiv byggnad med trästomme som hade samma konstruktionsförutsättningar men där referensobjektets fribärande högprofilsplåt i takkonstruktionen ersattes med element av KL-trä i jämförelseobjektet. Jämförelseobjektet modellerades, dimensionerades och ritningar projekterades enligt rådande normer. Byggnadernas ingående material mängdades och offerter erhölls från olika leverantörer. Byggnaderna bröts ner i sina beståndsdelar där varje enskilt materials klimatpåverkan analyserades med leverantörsspecifika EPD:er (environmental product declaration).   De parametrar som jämfördes var fukt, stadga, beständighet, bärighet, egentyngd, utförandetid, klimatpåverkan, tidskonstant, brand, försäkringsbolagens inställning, projekterbarhet, transporter & fria spännvidder samt olika kostnadsposter.   De två byggnaderna jämfördes under en kalkylperiod om 50 år och hänsyn togs till produktionsskedet, bruksskedet och återvinningsskedet men där energianvändning i bruksskedet avgränsades bort.   Resultatet visade att trä som stommaterial minskade klimatpåverkan från stommen med 40 432 kgCO2e (85 %) vilket motsvarade 27 % av hela byggnadens utsläpp. Resultatet visade även på att jämförelseobjektet kunde uppföras 15 dagar (36 %) snabbare än referensobjektet samt att arbetstiden kunde minskas med 725 timmar. Trästommen var även att föredra avseende beständighet, brand, energihushållning och totala kostnader.   Resultatet visade att ett byte från högprofilsplåt till KL-trä i takkonstruktionen ökade byggnadens egentyngd och kostnader avsevärt men också att träets massa förbättrade byggnadens energihushållning och minskade dess klimatpåverkan.   Resultatet visade att stål som stommaterial medförde en lägre egenvikt för referensobjektet med 22,4 ton (34 %) exkl. grundkonstruktionen och var mycket enklare att projektera. Stålstommen renderade även avsevärt lägre kostnader för försäkring 161 850 kr (21 %) och projektering 235 000 kr (65 %).   Analysen visade att grundkonstruktionen medförde en så pass stor klimatbelastning för bägge byggnaderna att klimatbelastningen från stommaterialet blev sekundärt. Analysen visade även att mängden klimatbelastande mineralull kunde minskas med 10 % i takkonstruktionen till följd av KL-träts låga värmeledningsförmåga. Vidare visade analysen att stålstommens klimatbelastning främst skedde i byggnadens produktionsskede medan trästommens klimatbelastning främst skedde i byggnadens slutskede.   Analysen visade slutligen att jämförelseobjektet, under sin livscykel, medförde såväl en minskad klimatpåverkan per investerad krona om 9 kgCO2e/Tkr (24,9 %) relativt referensobjektet som en minskad klimatpåverkan per kvadratmeter om 40,2 kgCO2e/m2 (27,3 %) och lägre kostnader om 0,09 Tkr/m2 (2,2 %).   Studiens slutsatser är att det varken är svårare att uppnå kraven i BBR som ställs på en enplans hallbyggnad med lagerverksamhet med en trästomme eller att det medför högre kostnader. Slutligen så hade trä som stommaterial varit att föredra för det studerade objektet. / The latest figures from Boverket's climate indicators showed that the construction industry, in 2016, accounted for 21 Mton CO2e (carbon dioxide equivalents). This represented 18% of the total national emissions. In measurements a few years earlier (2012), it appeared that private premises accounted for 20% of Sweden's total construction investments, which was in the same order of magnitude as the investments in multi-dwelling buildings. In recent years, a major focus has been on streamlining the production and use phase of multi-dwelling buildings, while the market for hall buildings is still dominated (about 80% market share) of climate-impacting steel frames and large amounts of mineral wool. In a preliminary study based on interviews and a questionnaire aimed at industry contacts with experience of erected hall buildings, it appeared that clients and entrepreneurs experienced difficulties partly fulfilling BBR's demands for sound, fire and moisture with a wooden construction, but also difficulties in justifying wood as frame material economically. The preliminary study also showed that clients and contractors weighted cost-related parameters primarily and climate-related parameters secondary. Based on the use of less climate-impacting materials in the construction process, the study aimed at investigating whether and how wood could compete with steel as frame material in the new production of a single-storey hall building with warehouse operations and to provide an overall picture of how a change from a steel frame to a wooden frame could affect the building with the system boundaries set on the outside of the climate shell incl. the buildings foundation. As a result of the increasing interest in KL-wood (cross-laminated wood) as backing material, the study also aimed to investigate what effects an element of KL-wood could have in the frame construction and how the material could be implemented most appropriately. Based on a real reference object, which consisted of a single-storey hall building of 1008 m2 with warehouse operations and a lightweight steel frame, a comparison was made in the study with a fictitious building with wooden frames which had the same design conditions but where the reference object's cantilevered high-profile plate in the roof construction was replaced by elements of KL wood in the comparison object. The comparison building was modeled, designed and laid out in drawing according to prevailing standards. The building materials priceoffers were obtained from various suppliers. The buildings were broken down into their constituents, where the climate impact of each individual material was analyzed with supplier-specific EPDs (environmental product declaration). The parameters that were compared were humidity, stability, durability, bearing capacity, self-weight, building- and performancetime, climate impact, time constant, fire, insurance companies' attitude, projectability, transport & free ranges and costs.   The two buildings were compared during a calculation period of 50 years and took into account the production stage, the use stage and the recovery stage of limited energy use in the use stage. The result showed that wood as backing material reduced the climate impact from the body with 40 432 kgCO2e (85%), which corresponded to 27% of the entire building's emissions. The result also showed that the comparison object could be built 15 days (36%) faster than the reference object and that the performance time could be reduced by 725 hours. The wooden frame was also preferable for durability, fire, energy conservation and total costs.   The result showed that a change from high-profile steelplate to KL-wood in the roof construction increased the building's self-weight and costs considerably, but also that the wood's mass improved the building's energy conservation and reduced its climate impact. The result showed that steel as backing material resulted in a lower self-weight for the reference object with 22.4 tonnes (34%) excluding the foundation and was much easier to project planning. The steel frame also rendered significantly lower costs for insurance SEK 161,850 (21%) and planning SEK 235,000 (65%).   The analysis showed that the foundation resulted in such a large climate impact for both buildings that the climate impact from the body material became secondary. The analysis also showed that the amount of climate-impacting mineral wool could be reduced by 10% in the roof construction as a result of the low heat conductivity of KL-wood. Furthermore, the analysis showed that the climate impact of the steel frame mainly took place in the building's production stage, while the climate impact of the wooden frame occurred mainly in the final stage of the buildings lifecycle.   Finally, the analysis showed that the comparison object, during its life cycle, resulted in both a climate saving per invested krona of 9 kgCO2e / Tkr (24.9%) relative to the reference object as well as a climate saving per square meter of 40.2 kgCO2e / m2 (27.3%) and cost savings of   0.09 Tkr / m2 (2.2%). The study's conclusion is that it is neither more difficult to achieve the requirements of BBR that are set on a single-storey hall building with warehouse operations with a wooden frame or that it entails higher costs. Finally, wood as a frame material had been preferred for the studied object.

Apesar da ampla diversidade de línguas indígenas no Brasil, poucas pesquisas estudam estas línguas e suas relações. Inúmeros esforços têm sido dedicados a procurar similaridades entre as palavras das línguas indígenas e classificá-las em famílias de línguas. Seguindo a classificação mais aceita das línguas indígenas do Brasil, esta pesquisa propõe comparar palavras de 10 línguas indígenas brasileiras. Para isso, considera-se que estas palavras são sinais de fala e estima-se a função de distribuição de probabilidade (PDF) de cada palavra, usando um modelo de mistura de gaussianas (GMM). A PDF foi considerada um modelo para representar as palavras. Os modelos foram comparados utilizando medidas de distância para construir estruturas hierárquicas que evidenciaram possíveis relações entre as palavras. Seguindo esta linha, a hipótese levantada nesta pesquisa é que as PDFs baseadas em GMM conseguem caracterizar as palavras das línguas indígenas, permitindo o emprego de medidas de distância entre elas para estabelecer relações entre as palavras, de forma que tais relações confirmem algumas das classificações. Os parâmetros do GMM foram calculados utilizando o algoritmo Maximização da Expectância (em inglês, Expectation Maximization (EM)). A divergência Kullback Leibler (KL) foi empregada para medir semelhança entre as PDFs. Esta divergência serve de base para estabelecer as estruturas hierárquicas que ilustram as relações entre os modelos. A estimativa da PDF, baseada em GMM foi testada com o auxílio de sinais simulados, sendo possível confirmar que os parâmetros obtidos são próximos dos originais. Foram implementadas várias medidas de distância para avaliar se a semelhança entre os modelos estavam determinadas pelos modelos e não pelas medidas adotadas neste estudo. Os resultados de todas as medidas foram similares, somente foi observada alguma diferença nos agrupamentos realizados pela distância C2, por isso foi proposta como complemento da divergência KL. Estes resultados sugerem que as relações entre os modelos dependem das suas características, não das métricas de distância selecionadas no estudo e que as PDFs baseadas em GMM, conseguem fazer uma caracterização adequada das palavras. Em geral, foram observados agrupamentos entre palavras que pertenciam a línguas de um mesmo tronco linguístico, assim como se observou uma tendência a incluir línguas isoladas nos agrupamentos dos troncos linguísticos. Palavras que pertenciam a determinada língua apresentaram um comportamento padrão, sendo identificadas por esse tipo de comportamento. Embora os resultados para as palavras das línguas indígenas sejam inconclusivos, considera-se que o estudo foi útil para aumentar o conhecimento destas 10 línguas estudadas, propondo novas linhas de pesquisas dedicadas à análise destas palavras. / Although there exists a large diversity of indigenous languages in Brazil, there are few researches on these languages and their relationships. Numerous efforts have been dedicated to search for similarities among words of indigenous languages to classify them into families. Following the most accepted classification of Brazilian indigenous languages, this research proposes to compare words of 10 Brazilian indigenous languages. The words of the indigenous languages are considered speech signals and the Probability Distribution Function (PDF) of each word was estimated using the Gaussian Mixture Models (GMM). This estimation was considered a model to represent each word. The models were compared using distance measures to construct hierarchical structures that illustrate possible relationships among words. The hypothesis in this research is that the estimation of the PDF, based on GMM can characterize the words of indigenous languages, allowing the use of distance measures between the PDFs to establish relationships among the words and confirm some of the classifications. The Expectation Maximization algorithm (EM) was implemented to estimate the parameters that describe the GMM. The Kullback Leibler (KL) divergence was used to measure similarities between two PDFs. This divergence is the basis to establish the hierarchical structures that show the relationships among the models. The PDF estimation, based on GMM was tested using simulated signals, allowing confirming the useful approximation of the original parameters. Several distance measures were implemented to prove that the similarities among the models depended on the model of each word, and not on the distance measure adopted in this study. The results of all measures were similar, however, as the clustering results of the C2 distances showed some differences from the other clusters, C2 distance was proposed to complement the KL divergence. The results suggest that the relationships between models depend on their characteristics, and not on the distance measures selected in this study, and the PDFs based on GMM can properly characterize the words. In general, relations among languages that belong to the same linguistic branch were illustrated, showing a tendency to include isolated languages in groups of languages that belong to the same linguistic branches. As the GMM of some language families presents a standard behavior, it allows identifying each family. Although the results of the words of indigenous languages are inconclusive, this study is considered very useful to increase the knowledge of these types of languages and to propose new research lines directed to analyze this type of signals.

Apesar da ampla diversidade de línguas indígenas no Brasil, poucas pesquisas estudam estas línguas e suas relações. Inúmeros esforços têm sido dedicados a procurar similaridades entre as palavras das línguas indígenas e classificá-las em famílias de línguas. Seguindo a classificação mais aceita das línguas indígenas do Brasil, esta pesquisa propõe comparar palavras de 10 línguas indígenas brasileiras. Para isso, considera-se que estas palavras são sinais de fala e estima-se a função de distribuição de probabilidade (PDF) de cada palavra, usando um modelo de mistura de gaussianas (GMM). A PDF foi considerada um modelo para representar as palavras. Os modelos foram comparados utilizando medidas de distância para construir estruturas hierárquicas que evidenciaram possíveis relações entre as palavras. Seguindo esta linha, a hipótese levantada nesta pesquisa é que as PDFs baseadas em GMM conseguem caracterizar as palavras das línguas indígenas, permitindo o emprego de medidas de distância entre elas para estabelecer relações entre as palavras, de forma que tais relações confirmem algumas das classificações. Os parâmetros do GMM foram calculados utilizando o algoritmo Maximização da Expectância (em inglês, Expectation Maximization (EM)). A divergência Kullback Leibler (KL) foi empregada para medir semelhança entre as PDFs. Esta divergência serve de base para estabelecer as estruturas hierárquicas que ilustram as relações entre os modelos. A estimativa da PDF, baseada em GMM foi testada com o auxílio de sinais simulados, sendo possível confirmar que os parâmetros obtidos são próximos dos originais. Foram implementadas várias medidas de distância para avaliar se a semelhança entre os modelos estavam determinadas pelos modelos e não pelas medidas adotadas neste estudo. Os resultados de todas as medidas foram similares, somente foi observada alguma diferença nos agrupamentos realizados pela distância C2, por isso foi proposta como complemento da divergência KL. Estes resultados sugerem que as relações entre os modelos dependem das suas características, não das métricas de distância selecionadas no estudo e que as PDFs baseadas em GMM, conseguem fazer uma caracterização adequada das palavras. Em geral, foram observados agrupamentos entre palavras que pertenciam a línguas de um mesmo tronco linguístico, assim como se observou uma tendência a incluir línguas isoladas nos agrupamentos dos troncos linguísticos. Palavras que pertenciam a determinada língua apresentaram um comportamento padrão, sendo identificadas por esse tipo de comportamento. Embora os resultados para as palavras das línguas indígenas sejam inconclusivos, considera-se que o estudo foi útil para aumentar o conhecimento destas 10 línguas estudadas, propondo novas linhas de pesquisas dedicadas à análise destas palavras. / Although there exists a large diversity of indigenous languages in Brazil, there are few researches on these languages and their relationships. Numerous efforts have been dedicated to search for similarities among words of indigenous languages to classify them into families. Following the most accepted classification of Brazilian indigenous languages, this research proposes to compare words of 10 Brazilian indigenous languages. The words of the indigenous languages are considered speech signals and the Probability Distribution Function (PDF) of each word was estimated using the Gaussian Mixture Models (GMM). This estimation was considered a model to represent each word. The models were compared using distance measures to construct hierarchical structures that illustrate possible relationships among words. The hypothesis in this research is that the estimation of the PDF, based on GMM can characterize the words of indigenous languages, allowing the use of distance measures between the PDFs to establish relationships among the words and confirm some of the classifications. The Expectation Maximization algorithm (EM) was implemented to estimate the parameters that describe the GMM. The Kullback Leibler (KL) divergence was used to measure similarities between two PDFs. This divergence is the basis to establish the hierarchical structures that show the relationships among the models. The PDF estimation, based on GMM was tested using simulated signals, allowing confirming the useful approximation of the original parameters. Several distance measures were implemented to prove that the similarities among the models depended on the model of each word, and not on the distance measure adopted in this study. The results of all measures were similar, however, as the clustering results of the C2 distances showed some differences from the other clusters, C2 distance was proposed to complement the KL divergence. The results suggest that the relationships between models depend on their characteristics, and not on the distance measures selected in this study, and the PDFs based on GMM can properly characterize the words. In general, relations among languages that belong to the same linguistic branch were illustrated, showing a tendency to include isolated languages in groups of languages that belong to the same linguistic branches. As the GMM of some language families presents a standard behavior, it allows identifying each family. Although the results of the words of indigenous languages are inconclusive, this study is considered very useful to increase the knowledge of these types of languages and to propose new research lines directed to analyze this type of signals.

KL-träelement används som stommaterial i allt fler flerbostadshusprojekt i Sverige, då det fortfarande är en relativt ny metod är det intressant att undersöka den närmare. Vid val av stomme är byggtiden ofta en viktig faktor och därför ligger montagetiden till grund för jämförelsen i studien. Denna rapports syfte är att ge mer information inför valet genom att jämföra en KL-trästomme med en platsgjuten betongstomme utförd med plattbärlag och skalväggar med hänsyn till tidsåtgång av montage. En litteraturundersökning har genomförts för att analysera vilka faktorer som påverkar tidsåtgången vid resning av de två stomalternativen. Planeringsverktyget Powerproject har använts för att ta fram en tidsplanering för vardera stomme. Tidsplaneringarna har sedan jämförts för att se vilket stomalternativ som ger minst tidsåtgång. För att stommarna ska bli så jämförbara som möjligt har gemensamma egenskapskrav ställts på bärighet, brandmotstånd och ljudisolering samt U-värde. När stommen uppfyller dessa krav, betongen torkats till 90 % relativ fuktighet och efterföljande aktiviteter kan påbörjas bedöms stomarbetet klart. Simuleringarna av tidplanerna har gjorts för kvarteret Norrstjärnan i Örebro som har använts som referensobjekt. Tiderna i rapporten är baserade på underlag tillhandahållet av NCC. Utifrån tidsplaneringarna för referensobjektet konstateras att den totala tidsåtgången för en KL-trästomme är 125 arbetsdagar och 169 arbetsdagar för en betongstomme. Det är en skillnad på 44 arbetsdagar vilket ger en procentuell skillnad på ca 35 %. / CLT-elements are used as frame material in an increasing number of multi-dwelling buildings projects in Sweden, as it is still a relatively new method, it is interesting to investigate it more closely. In the choice of frame, construction time is often an important factor and therefore the assembly time is the basis for the comparison in the study. The purpose of this report is to provide more information prior to the choice by comparing a CLT-element frame with a cast concrete frame made with lattice girder system and half shell precast walls with regard to the time required for assembly. A literature survey has been carried out to analyze which factors influence the time spent on raising the two frame options. The planning tool Powerproject has been used to develop a time schedule for each frame. The time schedules have then been compared to see which framework alternative gives the least amount of production time. In order for the frames to be as comparable as possible, common property requirements have been set for bearing capacity, fire resistance, sound insulation and U-value. When the framework meets these requirements, the concrete is dried to 90 % relative humidity and subsequent activities can be started, the work with the frame is considered finished. The simulations of the timetables have been made for the project Norrstjärnan in Örebro, which has been used as a reference object. The times in the report are based on data provided by NCC. Based on the time schedules for the reference object, it is stated that the total time spent on a CLT-element frame is 125 working days and 169 working days for a concrete frame. This is a difference of 44 working days, which gives a percentage difference of about 35 %.

<p>Predmet istraživanja ovog rada je istraživanje i eksploatacija mogućnosti da parametri Gausovih komponenti korišćenih Gaussian mixture modela&nbsp; (GMM) aproksimativno leže na niže dimenzionalnoj površi umetnutoj u konusu pozitivno definitnih matrica. U tu svrhu uvodimo novu, mnogo efikasniju meru sličnosti između GMM-ova projektovanjem LPP-tipa parametara komponenti iz više dimenzionalnog parametarskog originalno konfiguracijskog prostora u prostor značajno niže dimenzionalnosti. Prema tome, nalaženje distance između dva GMM-a iz originalnog prostora se redukuje na nalaženje distance između dva skupa niže dimenzionalnih euklidskih vektora, ponderisanih odgovarajućim težinama. Predložena mera je pogodna za primene koje zahtevaju visoko dimenzionalni prostor obeležja i/ili veliki ukupan broj Gausovih komponenti. Razrađena metodologija je primenjena kako na sintetičkim tako i na realnim eksperimentalnim podacima.</p> / <p>This thesis studies the possibility that the parameters of Gaussian components of a<br />particular Gaussian Mixture Model (GMM) lie approximately on a lower-dimensional<br />surface embedded in the cone of positive definite matrices. For that case, we deliver<br />novel, more efficient similarity measure between GMMs, by LPP-like projecting the<br />components of a particular GMM, from the high dimensional original parameter space,<br />to a much lower dimensional space. Thus, finding the distance between two GMMs in<br />the original space is reduced to finding the distance between sets of lower<br />dimensional euclidian vectors, pondered by corresponding weights. The proposed<br />measure is suitable for applications that utilize high dimensional feature spaces and/or<br />large overall number of Gaussian components. We confirm our results on artificial, as<br />well as real experimental data.</p>

This diploma thesis introduces three different methods of creating a neural network binary classifier for the purpose of automated weather prediction with attribute pre-selection using association rules and correlation patters mining by the LISp-Miner system. First part of the thesis consists of collection of theoretical knowledge enabling the creation of such predictive model, whereas the second part describes the creation of the model itself using the CRISP-DM methodology. Final part of the thesis analyses the performance of created classifiers and concludes the proposed methods and their possible benefits over training the network without attribute pre-selection.

The risk of moistening wood material should be considered and avoided as varying moisture content in the air causes the wood to swell and shrink. Moisture content in the material has influence on its mechanical properties, which may lead to damage due to reduced strength of the material. Wood material should be stored in places where the external climate does not contribute to an increase in the moisture content of the element. An optimal storage location is where the moisture content of wood material decreases. The purpose of this study is to increase knowledge about how the moisture content in cross-laminated timber (CLT) changes under different storage conditions. To achieve the purpose and get a broad picture of how moisture varies within CLT element, a measurement method that could handle measurements at different depths in the element was chosen. Furthermore, the CLT elements were placed in several storage places that simulated storage conditions at the workplace, where the amount of moisture in the environment varied based on the diurnal variation. The study has shown that, over the entire measurement period, the CLT elements that were placed indoors had a reduction of moisture content by 4.5%. The elements that were placed in a tent hall showed a decreasing of moisture content by about 1.5%. The elements that were placed outdoors with weather protection had an increasing moisture ratio of about 3%. The elements that were placed outdoors without weather protection had been affected greatly by the external climate and during precipitation the moisture content increased even greater. According to the study, CLT should be stored indoors or in a tent hall where the external climate does not contribute to an increase in the element's moisture content compared with outdoor storage. Furthermore, the study showed that cracks in the CLT elements led to higher moisture contents in the material. For the elements that were placed outdoors without weather protection that the cracked element reached its saturation point during precipitation, while the element without cracks showed increase of moisture content by a certain percentage and this increase is much slower.

No description available.

Denna fuktstudie handlar om användandet av diffusionsspärr i väggkonstruktioner i anslutning till badrumsmiljöer med KL-trä som stomme. Metoden som används är datorprogrammet WUFI, som beräknar värme- och fukttransport genom Fourier’s lag, Fick’s lag och Darcy’s andra lag. Indata till programmet är skapad genom mätningar av verkliga badrumsklimats variationer. Resultat från simulering visar att KL-trästommen riskerar mögelangrepp om inget ytterligare ytskikt finns applicerad på insidan av KL-trästommen. En ytskiktssimulering genomförs som visar att ett ytskiktsmaterial med en viss diffusionstäthet måste appliceras på KL-träytan för att undvika fuktrisknivåer inuti konstruktionen. / Torparängen