The diploma thesis is focused on creating a business plan for establishment a new restaurant that will dedicate itself to serve slovakian, hungarian and czech cuisine. Part of the thesis is to put together sales and costs prediction model in a pertinent subbranch. It also contains internal and external environment analysis, financial and business continuity plan.

No description available.

This thesis presents performance evaluation and a field validation study of a time and temperature indexed autoregressive with exogenous (4-3-5 ARX) building thermal load prediction model with an aim to integrate the model with actual predictive control systems. The 4-3-5 ARX model is very simple and computationally efficient with relatively high prediction accuracy compared to the existing sophisticated prediction models, such as artificial neural network prediction models. However, performance evaluation and field validation of the model are essential steps before implementing the model in actual practice. The performance of the model was evaluated under different climate conditions as well as under modeling uncertainty. A field validation study was carried out for three buildings at Mississippi State University. The results demonstrate that the 4-3-5 ARX model can predict building thermal loads in an accurate manner most of the times, indicating that the model can be readily implemented in predictive control systems.

Cooling towers form an important part of chilled water systems and perform the function of rejecting the heat to the atmosphere. These systems are often not operated optimally, and cooling towers being an integral part of the system present a significant area to study and determine possible energy saving measures. Operation of cooling towers in economizer mode in winter and variable frequency drives (VFDs) on cooling tower fans are measures that can provide considerable energy savings. The chilled water system analysis tool (CWSAT) software is developed as a primary screening tool for energy evaluation for chilled water systems and quantifies the energy usage of the various components and typical measures that can be applied to these systems to conserve energy, all while requiring minimum number of inputs to analyze component-wise energy consumption and incurred overall cost. A careful investigation of the current model in CWSAT indicates that the prediction capability of the model at lower wet bulb temperatures and at low fan power is not very accurate. A new model for accurate tower performance prediction is imperative, since economizer operation occurs at low temperatures and most cooling towers come equipped with VFDs. In this thesis, a new model to predict cooling tower performance is created to give a more accurate prediction of energy savings for a tower. Further the economic feasibility of having additional cooling tower capacity to allow for economizer cooling, in light of reduced tower capacity at lower temperatures is investigated.

The main purpose of the AOD-converter is to lower the carbon content in stainless steel production. The carbon content can be estimated by static theoretical models. It can also be estimated through dynamic models based on analysis of the exhaust gases from the converter. This master thesis is a study on an extended use of exhaust gas analysis data on the AOD-converter at Outokumpu’s stainless steel plant in Avesta, Sweden. There are two main methods of predicting the carbon content based on exhaust gas analysis, mass balance and a linear regression between decarburization rate and carbon content. This master thesis mainly focuses on the development of the linear regression model for steel grades ASTM 304L, 316L, S32101 and S32205 for the last step of the decarburization, as well as ASTM S32205 and S30815 for the second last step of the decarburization. The results showed that the linear regression model can predict the carbon content at the last step of decarburization with a standard deviation between 0,00626 %C and 0,0109 %C for the different steel grades. An equation for carbon prediction dependent on the steel composition was also developed in the master thesis, making it theoretically possible to use for all steel grades, it has however not yet been tested on other steel grades. The CRE measured from the exhaust gases was also studied to find out if it is possibleto use as basis for step changes during the decarburization, but the resultswere inconclusive. / Huvudsyftet med AOD-konvertern är att sänka kolhalten i produktionen av rostfritt stål. Kolhalten kan uppskattas av statiska teoretiska modeller. Den kan också uppskattas av dynamiska modeller baserade på analys av avgaserna från konvertern. Det här examensarbetet handlar om utvidgning av användandet av avgasanalysdata på AOD-konvertern på Outokumpus stålverk i Avesta, Sverige. Det finns i huvudsak två metoder för att bestämma kolhalten med hjälp av avgasanalys, massbalans och en linjär regression mellan kolfärskningshastigheten och kolhalten. Det här examensarbetet fokuserar i huvudsak på utvecklingen av den linjära modellen för stålsorterna ASTM 304L, 316L, S32101 och S32205 för sista steget i kolfärskningen. Samt stålsorterna ASTM S32205 och S30815 för näst sista steget i kolfärskningen. Resultaten visade att den linjära modellen kunde uppskatta kolhalten i sista steget av kolfärskningen med en standardavvikelse mellan 0,00626 %C och 0,0109 %C för de fyra olika stålsorterna. En ekvation som anger sambandet mellan sammansättningen på stålet under kolfärskningen och ekvationen för den linjära regressionen togs också fram i examensarbetet. Teoretiskt kan ekvationen användas för alla stålsorter men den har inte än blivit testad på andra stålsorter. CRE uppmätt med hjälp av avgasanalys undersöktes också för att ta reda på om CRE kan användas för att bestämma när stegbytena ska ske, det gick dock inte att utgöra från resultaten.

京都大学 / 新制・課程博士 / 博士(薬科学) / 甲第24559号 / 薬科博第176号 / 新制||薬科||19(附属図書館) / 京都大学大学院薬学研究科医薬創成情報科学専攻 / (主査)教授 馬見塚 拓, 教授 山下 富義, 教授 金子 周司 / 学位規則第4条第1項該当 / Doctor of Pharmaceutical Sciences / Kyoto University / DFAM

<p dir="ltr">This study aims to develop Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), and Transformer-based Informer models and evaluate the performance of these models using data from one, two, three, and four weeks in advance to predict the progression of rice leaf blast disease; and assess the generalizability of these models across various climatic regions in Taiwan. This research utilized multi-location rice leaf blast diseased leaf percentage data collected between 2015 and 2021 in Taiwan, along with weather data from the Taiwanese meteorological observation network to predict rice blast disease one week in advance, serving as a benchmark for comparing with predictions made two, three, and four weeks in advance.</p>

Doctor of Philosophy / Department of Civil Engineering / Mustaque Hossain / A Superpave asphalt mixture with 4.75-mm nominal maximum aggregate size (NMAS) is a promising, low-cost pavement preservation treatment for agencies such as the Kansas Department of Transportation (KDOT). The objective of this research study is to develop an optimized 4.75-mm NMAS Superpave mixture in Kansas. In addition, the study evaluated the residual tack coat application rate for the 4.75-mm NMAS mix overlay. Two, hot-in-place recycling (HIPR) projects in Kansas, on US-160 and K-25, were overlaid with a 15- to 19-mm thick layer of 4.75-mm NMAS Superpave mixture in 2007. The field tack coat application rate was measured during construction. Cores were collected from each test section for Hamburg wheel tracking device (HWTD) and laboratory bond tests performed after construction and after one year in service. Test results showed no significant effect of the tack coat application rate on the rutting performance of rehabilitated pavements. The number of wheel passes to rutting failure observed during the HWTD test was dependent on the aggregate source as well as on in-place density of the cores. Laboratory pull-off tests showed that most cores were fully bonded at the interface of the 4.75-mm NMAS overlay and the HIPR layer, regardless of the tack application rate. The failure mode during pull-off tests at the HMA interface was highly dependent on the aggregate source and mix design of the existing layer material. This study also confirmed that overlay construction with a high tack coat application rate may result in bond failure at the HMA interface. Twelve different 4.75-mm NMAS mix designs were developed using materials from the aforementioned but two binder grades and three different percentages of natural (river) sand. Laboratory performance tests were conducted to assess mixture performance. Results show that rutting and moisture damage potential in the laboratory depend on aggregate type irrespective of binder grade. Anti-stripping agent affects moisture sensitivity test results. Fatigue performance is significantly influenced by river sand content and binder grade. Finally, an optimized 4.75-mm NMAS mixture design was developed and verified based on statistical analysis of performance data.

There is an intense, worldwide effort to develop durable lithium ion batteries with high energy and power densities for a wide range of applications, including electric and hybrid electric vehicles. For improvement of battery technology understanding the capacity fading mechanism in batteries is of utmost importance. Novel electrode material and improved electrode designs are needed for high energy- high power batteries with less capacity fading. Furthermore, for applications such as automotive applications, precise cycle-life prediction of batteries is necessary. One of the critical challenges in advancing lithium ion battery technologies is fracture and decrepitation of the electrodes as a result of lithium diffusion during charging and discharging operations. When lithium is inserted in either the positive or negative electrode, there is a volume change associated with insertion or de-insertion. Diffusion-induced stresses (DISs) can therefore cause the nucleation and growth of cracks, leading to mechanical degradation of the batteries. With different mathematical models we studied the behavior of diffusion induces stresses and effects of electrode shape, size, concentration dependent material properties, pre-existing cracks, phase transformations, operating conditions etc. on the diffusion induced stresses. Thus we develop tools to guide the design of the electrode material with better mechanical stability for durable batteries. Along with mechanical degradation, chemical degradation of batteries also plays an important role in deciding battery cycle life. The instability of commonly employed electrolytes results in solid electrolyte interphase (SEI) formation. Although SEI formation contributes to irreversible capacity loss, the SEI layer is necessary, as it passivates the electrode-electrolyte interface from further solvent decomposition. SEI layer and diffusion induced stresses are inter-dependent and affect each-other. We study coupled chemical-mechanical degradation of electrode materials to understand the capacity fading of the battery with cycling. With the understanding of chemical and mechanical degradation, we develop a simple phenomenological model to predict battery life. On the experimental part we come up with a novel concept of using liquid metal alloy as a self-healing battery electrode. We develop a method to prepare thin film liquid gallium electrode on a conductive substrate. This enabled us to perform a series of electrochemical and characterization experiments which certify that liquid electrode undergo liquid-solid-liquid transition and thus self-heals the cracks formed during de-insertion. Thus the mechanical degradation can be avoided. We also perform ab-initio calculations to understand the equilibrium potential of various lithium-gallium phases.

Avec la mise en place de la nouvelle limite maximale de 400 000 cellules somatiques par millilitres de lait (c/mL) au réservoir, le mois d’août 2012 a marqué une étape importante en termes de qualité du lait pour les producteurs de bovins laitiers du Canada. L’objectif de cette étude consistait en l’établissement d’un modèle de prédiction de la violation de cette limite au réservoir à l’aide des données individuelles et mensuelles de comptages en cellules somatiques (CCS) obtenues au contrôle laitier des mois précédents. Une banque de donnée DSA comprenant 924 troupeaux de laitiers québécois, en 2008, a été utilisée pour construire un modèle de régression logistique, adapté pour les mesures répétées, de la probabilité d’excéder 400 000 c/mL au réservoir. Le modèle final comprend 6 variables : le pointage linéaire moyen au test précédent, la proportion de CCS > 500 000 c/mL au test précédent, la production annuelle moyenne de lait par vache par jour, le nombre de jours en lait moyen (JEL) au test précédent ainsi que les proportions de vaches saines et de vaches infectées de manière chronique au test précédant. Le modèle montre une excellente discrimination entre les troupeaux qui excèdent ou n’excèdent pas la limite lors d’un test et pourrait être aisément utilisé comme outil supplémentaire de gestion de la santé mammaire à la ferme. / August 2012 represents an in important step in terms of milk quality for the Canadian bovine dairy producers because the upper tolerance limit for bulk tank somatic cell count (BTSCC) was lowered to 400 000 somatic cells per millilitre (c/mL). The objective of this study was to develop a predictive model of exceeding the BTSCC limit based on monthly individuals somatic cell count (SCC) measures obtained in the previous months. A database including DHI data from the year 2008 of 924 dairy herds in Québec, Canada was used. A logistic regression model for repeated measures was constructed. The final model included 6 variables related to monthly individual cow somatic cell count: mean individual linear score at the previous test, proportion of cows over 500 000 cells/mL at the previous test, mean herd annual daily milk production per cow, average days in milk at the previous test, proportion of healthy cows at previous test, proportion of chronic cows at previous test. The model has excellent discrimination between herd that exceeded and herd that did not exceed 400 000 cells/mL and can be use to advise dairy producers of impending risks of exceeding the BTSCC limit.