No description available.

After reaching its light-off temperature, the catalytic aftertreatment system plays a major part in maintaining emissions at low levels for vehicles equipped with combustion engines. In this thesis, modelling of the exhaust gas temperature is investigated along with optimal control strategy for variable ignition and exhaust valve opening angles for optimal catalytic converter heating. Models for exhaust gas temperature and mass flow are presented and validated against measurement data. According to the model validation, the proposed models capture variations in ignition and exhaust valve opening angles well. Optimal control strategy for the ignition and exhaust valve opening angles to heat the catalytic converter to a predetermined temperature in the most fuel and time optimal ways are investigated by implementation of the validated models. Optimal control analysis indicates that with open wastegate, the heating time for the catalytic converter can be reduced by up to 16.4 % and the accumulated fuel to reach the desired temperature can be reduced by up to 4.6 %, compared to the case with ignition and exhaust valve opening angles fixed at nominal values. With closed wastegate the corresponding figures are 16.4 % and 4.7 %. By also including control of the variable λ-value, the heating time can be further reduced by up to 19.8 %, and the accumulated fuel consumption by up to 9.5 % with open wastegate. With closed wastegate the corresponding figures are 20.1 % decrease in heating time, and 9.8 % decrease in accumulated fuel consumption.

In this thesis, we plan to use a group-based semantic-expansion approach to design a new personalised system framework. Semantic web and group preference offer solution to the above problem. In this thesis, ontologies and semantic techniques are applied in different components of the framework. Information has been gathered from different resources and each of the resource might be using various types of identifiers for the same concept, therefore semantic web technologies are used to find out if the concept is the same or not. On the other hand, we create group preference in our personalization system. If the system fails to obtain personal preference from new user, group preference supports the system providing recommendation to the new user according to group classification.

Proton exchange membrane fuel cell (PEMFC) has been considered as one of the most promising energy conversion devices for the future in automotive applications. One of the major technical challenges for the commercialization of PEMFC is the effective start-up from subzero temperatures, often referred to as “cold start”. The major problem of PEMFC cold start is that the product water freezes when the temperature inside the PEMFC is lower than the freezing point. If the catalyst layer (CL) is fully occupied by ice before the cell temperature rises above the freezing point, the electrochemical reaction may stop due to the blockage of the reaction sites. However, only a few of the previous PEMFC studies paid attention to cold start. Hence, understanding the ice formation mechanisms and optimizing the design and operational strategies for PEMFC cold start are critically important. In this research, an experimental setup for the cold start testing with simultaneous measurement of current and temperature distributions is designed and built; a one-dimensional (1D) analytical model for quick estimate of purging durations before the cold start processes is formulated; and a comprehensive three-dimensional (3D) PEMFC cold start model is developed. The unique feature of the cold start experiment is the inclusion of the simultaneous measurement of current and temperature distributions. Since most of the previous numerical models are limited to either 1D or two-dimensional (2D) or 3D but only considering a section of the entire cell due to computational requirement, the measured distribution data are critically important to better understand the PEMFC cold start characteristics. With a full set of conservation equations, the 3D model comprehensively accounts for the various transport phenomena during the cold start processes. The unique feature of this model is the inclusion of: (i) the water freezing in the membrane electrolyte and its effects on the membrane conductivity; (ii) the non-equilibrium mass transfer between the water in the ionomer and the water (vapour, liquid and ice) in the pore region of the CL; and (iii) both the water freezing and melting in the CL and gas diffusion layer (GDL). This model therefore provides the fundamental framework for the future top-down multi-dimensional multiphase modelling of PEMFC. The experimental and numerical results elaborate the ice formation mechanisms and other important transport phenomena during the PEMFC cold start processes. The effects of the various cell designs, operating conditions and external heating methods on the cold start performance are studied. Independent tests are carried out to identify and optimize the important design and operational parameters.

Proton exchange membrane fuel cell (PEMFC) has been considered as one of the most promising energy conversion devices for the future in automotive applications. One of the major technical challenges for the commercialization of PEMFC is the effective start-up from subzero temperatures, often referred to as “cold start”. The major problem of PEMFC cold start is that the product water freezes when the temperature inside the PEMFC is lower than the freezing point. If the catalyst layer (CL) is fully occupied by ice before the cell temperature rises above the freezing point, the electrochemical reaction may stop due to the blockage of the reaction sites. However, only a few of the previous PEMFC studies paid attention to cold start. Hence, understanding the ice formation mechanisms and optimizing the design and operational strategies for PEMFC cold start are critically important. In this research, an experimental setup for the cold start testing with simultaneous measurement of current and temperature distributions is designed and built; a one-dimensional (1D) analytical model for quick estimate of purging durations before the cold start processes is formulated; and a comprehensive three-dimensional (3D) PEMFC cold start model is developed. The unique feature of the cold start experiment is the inclusion of the simultaneous measurement of current and temperature distributions. Since most of the previous numerical models are limited to either 1D or two-dimensional (2D) or 3D but only considering a section of the entire cell due to computational requirement, the measured distribution data are critically important to better understand the PEMFC cold start characteristics. With a full set of conservation equations, the 3D model comprehensively accounts for the various transport phenomena during the cold start processes. The unique feature of this model is the inclusion of: (i) the water freezing in the membrane electrolyte and its effects on the membrane conductivity; (ii) the non-equilibrium mass transfer between the water in the ionomer and the water (vapour, liquid and ice) in the pore region of the CL; and (iii) both the water freezing and melting in the CL and gas diffusion layer (GDL). This model therefore provides the fundamental framework for the future top-down multi-dimensional multiphase modelling of PEMFC. The experimental and numerical results elaborate the ice formation mechanisms and other important transport phenomena during the PEMFC cold start processes. The effects of the various cell designs, operating conditions and external heating methods on the cold start performance are studied. Independent tests are carried out to identify and optimize the important design and operational parameters.

In recent years government, academia and industry have been pursuing fuel cell technology as an alternative to current power generating technologies. The automotive industry has targeted fuel cell technology as a potential alternative to internal combustion engines. The goal of this research is to understand and quantify the impact and effects of low temperature operation has on the performance and efficiency of vehicle fuel cell systems through modeling. More specifically, this work addresses issues of the initial thermal transient known to the automotive community as "cold-start" effects. Cold-start effects play a significant role in power limitations in a fuel cell vehicle, and may require hybridization (batteries) to supplement available power. A fuel cell system model developed as part of this work allows users to define the basic thermal fluid relationships in a fuel cell system. The model can be used as a stand-alone version or as part of a complex fuel cell vehicle model. Fuel cells are being considered for transportation primarily because they have the ability to increase vehicle energy efficiency and significantly reduce or eliminate tailpipe emissions. A proton exchange membrane fuel cell is an electrochemical device for which the operational characteristics depend heavily upon temperature. Thus, it is important to know how the thermal design of the system affects the performance of a fuel cell, which governs the efficiency and performance of the system. This work revealed that the impact on efficiency of a cold-start yielded a 5 % increase in fuel use over a regulated drive cycle for the converted sport utility vehicle. The performance of the fuel cell vehicle also suffered due to operation at low temperatures. Operation of the fuel cell at 20 C yielded only 50% of the available power to the vehicle system. / Master of Science

Recommender systems apply machine learning methods to solve the task of providing appropriate suggestions to users in both static and dynamic environments. An example of this is a movie service like Netflix that recommends movies to its users. Although many algorithms have been proposed, making predictions for users with few ratings remains a challenge in recommender systems. In this study the performance of the algorithm k-NN, both user- and item-based, was empirically evaluated. This was done using the MovieLens 1M and 100K datasets in scenarios where the users have between 1 and 9 ratings, simulating cold-start scenarios of various degree. The results were then compared with the accuracy of the algorithm in a simulated normal case, to see how the cold-start affected the two algorithms, and which one of them that handled it best. In summary, this report shows that user-based k-NN performs better in relation to item-based k-NN for new users having few rated items. Overall the accuracy improved as the number of ratings increased for the new users for both user- and item-based k-NN.

Sistemas de Recomendação (SR) vêm se apresentando como poderosas ferramentas para portais web tais como sítios de comércio eletrônico. Para fazer suas recomendações, os SR se utilizam de fontes de dados variadas, as quais capturam as características dos usuários, dos itens e suas transações, bem como de modelos de predição. Dada a grande quantidade de dados envolvidos, é improvável que todas as recomendações possam ser bem representadas por um único modelo global de predição. Um outro importante aspecto a ser observado é o problema conhecido por cold-start, que apesar dos avanços na área de SR, é ainda uma questão relevante que merece uma maior atenção. O problema está relacionado com a falta de informação prévia sobre novos usuários ou novos itens do sistema. Esta tese apresenta uma abordagem híbrida de recomendação capaz de lidar com situações extremas de cold-start. A abordagem foi desenvolvida com base no algoritmo SCOAL (Simultaneous Co-Clustering and Learning). Na sua versão original, baseada em múltiplos modelos lineares de predição, o algoritmo SCOAL mostrou-se eficiente e versátil, podendo ser utilizado numa ampla gama de problemas de classificação e/ou regressão. Para melhorar o algoritmo SCOAL no sentido de deixá-lo mais versátil por meio do uso de modelos não lineares, esta tese apresenta uma variante do algoritmo SCOAL que utiliza modelos de predição baseados em Máquinas de Aprendizado Extremo. Além da capacidade de predição, um outro fator que deve ser levado em consideração no desenvolvimento de SR é a escalabilidade do sistema. Neste sentido, foi desenvolvida uma versão paralela do algoritmo SCOAL baseada em OpenMP, que minimiza o tempo envolvido no cálculo dos modelos de predição. Experimentos computacionais controlados, por meio de bases de dados amplamente usadas na prática, comprovam que todos os desenvolvimentos propostos tornam o SCOAL ainda mais atraente para aplicações práticas variadas. / Recommender Systems (RS) are powerful and popular tools for e-commerce. To build its recommendations, RS make use of multiple data sources, capture the characteristics of items, users and their transactions, and take advantage of prediction models. Given the large amount of data involved in the predictions made by RS, is unlikely that all predictions can be well represented by a single global model. Another important aspect to note is the problem known as cold-start that, despite that recent advances in the RS area, it is still a relevant issue that deserves further attention. The problem arises due to the lack of prior information about new users and new items. This thesis presents a hybrid recommendation approach that addresses the (pure) cold start problem, where no collaborative information (ratings) is available for new users. The approach is based on an existing algorithm, named SCOAL (Simultaneous Co-Clustering and Learning). In its original version, based on multiple linear prediction models, the SCOAL algorithm has shown to be efficient and versatile. In addition, it can be used in a wide range of problems of classification and / or regression. The SCOAL algorithm showed impressive results with the use of linear prediction models, but there is still room for improvements with nonlinear models. From this perspective, this thesis presents a variant of the SCOAL based on Extreme Learning Machines. Besides improving the accuracy, another important issue related to the development of RS is system scalability. In this sense, a parallel version of the SCOAL, based on OpenMP, was developed, aimed at minimizing the computational cost involved as prediction models are learned. Experiments using real-world datasets has shown that all proposed developments make SCOAL algorithm even more attractive for a variety of practical applications.

Sistemas de Recomendação (SR) vêm se apresentando como poderosas ferramentas para portais web tais como sítios de comércio eletrônico. Para fazer suas recomendações, os SR se utilizam de fontes de dados variadas, as quais capturam as características dos usuários, dos itens e suas transações, bem como de modelos de predição. Dada a grande quantidade de dados envolvidos, é improvável que todas as recomendações possam ser bem representadas por um único modelo global de predição. Um outro importante aspecto a ser observado é o problema conhecido por cold-start, que apesar dos avanços na área de SR, é ainda uma questão relevante que merece uma maior atenção. O problema está relacionado com a falta de informação prévia sobre novos usuários ou novos itens do sistema. Esta tese apresenta uma abordagem híbrida de recomendação capaz de lidar com situações extremas de cold-start. A abordagem foi desenvolvida com base no algoritmo SCOAL (Simultaneous Co-Clustering and Learning). Na sua versão original, baseada em múltiplos modelos lineares de predição, o algoritmo SCOAL mostrou-se eficiente e versátil, podendo ser utilizado numa ampla gama de problemas de classificação e/ou regressão. Para melhorar o algoritmo SCOAL no sentido de deixá-lo mais versátil por meio do uso de modelos não lineares, esta tese apresenta uma variante do algoritmo SCOAL que utiliza modelos de predição baseados em Máquinas de Aprendizado Extremo. Além da capacidade de predição, um outro fator que deve ser levado em consideração no desenvolvimento de SR é a escalabilidade do sistema. Neste sentido, foi desenvolvida uma versão paralela do algoritmo SCOAL baseada em OpenMP, que minimiza o tempo envolvido no cálculo dos modelos de predição. Experimentos computacionais controlados, por meio de bases de dados amplamente usadas na prática, comprovam que todos os desenvolvimentos propostos tornam o SCOAL ainda mais atraente para aplicações práticas variadas. / Recommender Systems (RS) are powerful and popular tools for e-commerce. To build its recommendations, RS make use of multiple data sources, capture the characteristics of items, users and their transactions, and take advantage of prediction models. Given the large amount of data involved in the predictions made by RS, is unlikely that all predictions can be well represented by a single global model. Another important aspect to note is the problem known as cold-start that, despite that recent advances in the RS area, it is still a relevant issue that deserves further attention. The problem arises due to the lack of prior information about new users and new items. This thesis presents a hybrid recommendation approach that addresses the (pure) cold start problem, where no collaborative information (ratings) is available for new users. The approach is based on an existing algorithm, named SCOAL (Simultaneous Co-Clustering and Learning). In its original version, based on multiple linear prediction models, the SCOAL algorithm has shown to be efficient and versatile. In addition, it can be used in a wide range of problems of classification and / or regression. The SCOAL algorithm showed impressive results with the use of linear prediction models, but there is still room for improvements with nonlinear models. From this perspective, this thesis presents a variant of the SCOAL based on Extreme Learning Machines. Besides improving the accuracy, another important issue related to the development of RS is system scalability. In this sense, a parallel version of the SCOAL, based on OpenMP, was developed, aimed at minimizing the computational cost involved as prediction models are learned. Experiments using real-world datasets has shown that all proposed developments make SCOAL algorithm even more attractive for a variety of practical applications.

Published Article / The simulation of an exhaust manifold's thermal behaviour is an important concern for various reasons. Amongst them is the need to minimise catalyst light-offtime as significant exhaust emissions are generated within this period. Modelling such behaviour is not simplistic as it is governed by complex interactions between exhaust gas flow and the manifold itself. Computational fluid dynamics (CFD) is a powerful tool for such simulations. However its applicability for transient simulations is limited by high central processing unit (CPU) demands. The present study proposes an alternative computational method to assess and rank the relative impact of the manifold's thermal properties on its exterior temperature. The results show that stainless steel manifolds potentially minimise heat loss from the exhaust gas when compared with their cast iron counterparts. This may result in an increase in thermal energy being available to heat the catalyst.