<p> Software Architecture is a concept that arose during the last two decades as a consequence of the need for a structured design at an early stage. Software Architecture is defined as a pattern of interconnected components satisfying some structural rule. Software architectures are widely used in many types of systems; Multi-Agent Systems should not be an exception. Multi-Agent Systems have emerged as a design paradigm for large and distributed systems. They are composed of autonomous elements that work together in order to pursue a common goal. They are mostly used in Electronic Commerce, Human-Computer Interfaces, and so on.</p> <p> In this research, we investigate the state of the art of Software Architectures in the Multi-Agent Systems field, showing that, generally Multi-Agent Systems do no use the software architecture concept properly and, when they do, they do not show specific architectures for Multi-Agent Systems. The approach followed is based on the analysis of six case studies, which are implemented applications that have been published in some of the most important conferences in the area. Additionally we show that, based on the initial design of each case and existing architectural patterns, it is possible to impose a software architecture on each case.</p> <p> Furthermore, we analyze the way that the term "software architecture" is used in the Multi-Agent Systems literature, showing that, usually, it refers to abstract architectures proposed in standards and frameworks or to an initial design in a system. In addition we clarify related concepts, such as reference architecture, reference models, architectural patterns and design patterns. Finally, we do an exhaustive comparison of the case studies, which aims to highlight commonalities and differences. The objective of this comparison is to analyze if they share a similar architecture that can be reused in more cases and to show how specific properties of Multi-Agent Systems affect in the design of an architecture.</p> / Thesis / Master of Science (MSc)

Modular Multilevel Converters (MMC) are being widely considered for medium to high voltage DC generation systems. Integrated system design optimization of the generator-MMC system through multi-objective optimization is of interest, because such an approach allows the trade-off between competing objectives (for example, mass and loss) to be explicitly and quantitatively identified. In this work, such an optimization based design paradigm for MMC based generator rectifier systems is developed. To formulate the design problem as a multi-objective optimization problem, it is required that the system waveforms can be obtained to facilitate the imposition of constraints and the estimation of power losses. Similarly, it is also desired to include detailed electric machine magnetic and electrical analysis in design optimization, as well as aspects such as the inductor and heat sink design. Such development typically requires detailed component design and simulation models for the electric machine and converter which are computationally expensive. As an alternative, the proposed work utilizes an electric machine metamodel, heat sink metamodel, and high-speed steady-state simulation model for the MMC to facilitate multi-objective optimization minimizing system metrics of interest while satisfying system constraints. Using the developed component simulation and design models, a multi-objective optimization based design of an MMC based generator-rectifier system is conducted.

In recent years the service-oriented design paradigm has enabled applications to be built by incorporating third party services. With the increasing popularity of this new paradigm, many companies and organizations have started to adopt this technology, which has resulted in an increase of the number and variety of third party providers. With the vast improvement of global networking infrastructure, a large number of providers offer their services for worldwide clients. As a result, clients are often presented with a number of providers that offer services with the same or similar functionalities, but differ in terms of non-functional attributes (or Quality of Service – QoS), such as performance. In this environment, the role of provider recommendation has become more important - in assisting clients in choosing the provider that meets their QoS requirement. / In this thesis we focus on provider recommendation based on one of the most important QoS attributes – performance. Specifically, we investigate client-perceived performance, which is the application-level performance measured at the client-side every time the client invokes the service. This performance metric has the advantage of accurately representing client experience, compared to the widely used server-side metrics in the current frameworks (e.g. Service Level Agreement or SLA in Web Services context). As a result, provider recommendation based on this metric will be favourable from the client’s point of view. / In this thesis we address two key research challenges related to provider recommendation based on client-perceived performance - performance assessment and performance prediction. We begin by identifying heterogeneity factors that affect client-perceived performance among clients in a global Internet environment. We then perform extensive real-world experiments to evaluate the significance of each factor to the client-perceived performance. / From our finding on heterogeneity factors, we then develop a performance estimation technique to address performance assessment for cases where direct measurements are unavailable. This technique is based on the generalization concept, i.e. estimating performance based on the measurement gathered by similar clients. A two-stage grouping scheme based on the heterogeneity factors we identified earlier is proposed to address the problem of determining client similarity. We then develop an estimation algorithm and validate it using synthetic data, as well as real world datasets. / With regard to performance prediction, we focus on the medium-term prediction aspect to address the needs of the emerging technology requirements: distinguishing providers based on medium-term (e.g. one to seven days) performance. Such applications are found when the providers require subscription from their clients to access the service. Another situation where the medium-term prediction is important is in temporal-aware selection: the providers need to be differentiated, based on the expected performance of a particular time interval (e.g. during business hours). We investigate the applicability of classical time series prediction methods: ARIMA and exponential smoothing, as well as their seasonal counterparts – seasonal ARIMA and Holt-Winters. Our results show that these existing models lack the ability to capture the important characteristics of client-perceived performance, thus producing poor medium-term prediction. We then develop a medium-term prediction method that is specifically designed to account for the key characteristics of a client-perceived performance series, and to show that our prediction methods produce higher accuracy for medium-term prediction compared to the existing methods. / In order to demonstrate the applicability of our solution in practice, we developed a provider recommendation framework based on client-perceived performance (named PROPPER), which utilizes our findings on performance assessment and prediction. We formulated the recommendation algorithm and evaluated it through a mirror selection case study. It is shown that our framework produces better outcomes in most cases, compared to country-based or geographic distance-based selection schemes, which are the current approach of mirror selection nowadays.

A framework is proposed that combines separately developed multidisciplinary optimization, multi-objective optimization, and joint probability assessment methods together but in a decoupled way, to solve joint probabilistic constraint, multi-objective, multidisciplinary optimization problems that are representative of realistic conceptual design problems of design alternative generation and selection. The intent here is to find the Weak Pareto Frontier (WPF) solutions that include additional compromised solutions besides the ones identified by a conventional Pareto frontier. This framework starts with constructing fast and accurate surrogate models of different disciplinary analyses. A new hybrid method is formed that consists of the second order Response Surface Methodology (RSM) and the Support Vector Regression (SVR) method. The three parameters needed by SVR to be pre-specified are automatically selected using a modified information criterion based on model fitting error, predicting error, and model complexity information. The model predicting error is estimated inexpensively with a new method called Random Cross Validation. This modified information criterion is also used to select the best surrogate model for a given problem out of the RSM, SVR, and the hybrid methods. A new neighborhood search method based on Monte Carlo simulation is proposed to find valid designs that satisfy the deterministic constraints and are consistent for the coupling variables featured in a multidisciplinary design problem, and at the same time decouple the three loops required by the multidisciplinary, multi-objective, and probabilistic features. Two schemes have been developed. One scheme finds the WPF by finding a large enough number of valid design solutions such that some WPF solutions are included in those valid solutions. Another scheme finds the WPF by directly finding the WPF of each consistent design zone. Then the probabilities of the PCs are estimated, and the WPF and corresponding design solutions are found. Various examples demonstrate the feasibility of this framework.

A pragmatic method for selecting components and devices from a database or parameterized models is developed in this thesis. The quantitative framework presented here is sufficiently general to accommodate an entire device assembly, a component, or a sub-assembly. The details pertaining to a device or a component are classified into three sets of variables: (i) user-specifications, s (ii) device parameters, p , and (iii) device characteristics, c . Functional, practical, and performance-related attributes that a user can provide comprise user-specifications. Since, most often, a specification cannot be specified as a single number, we allow the user to enter a range with lower and upper bounds. Device parameters comprise the geometry and material properties, and device characteristics include functional requirements and performance criteria. Thus, for a device, all its functional and utility attributes are contained in the union of sets s and c , whereas the geometry and the material properties are in set p . The equations governing the physical behavior of the device are written in terms of s , p , and c . These equations may sometimes be readily available; when they are not, it may be necessary to formulate them as required. By solving the governing equations along with the inequalities that arise from the lower and upper bounds on s , we obtain feasible ranges on p and c . Then, for any pair of device characteristics, a 2D feasible map is drawn, to visually portray the consequences of user-specifications. If the feasible map is null, small, or large, it indicates that the user-specifications are infeasible, stringent, or there is much scope for design, respectively. This can be inferred even before the designs are considered. Juxtaposed on the feasible map are points or lozenges corresponding to the quantitative attributes of the entries in the database. The ones that lie inside the feasible map can be reckoned as meeting the user-specifications and thus, enabling selection. On the other hand, if there is no database or none of the devices in the database lie inside the feasible map, we can identify the feasible ranges of all the design parameters for every point inside the feasible map. This information is useful to the designer to redesign and arrive at feasible designs by using parameterized models of the device. A Graphical User Interface (GUI) is developed to facilitate the user-interaction. The utility of the selection framework is demonstrated with a variety of case-studies including miniature pumps, heat pulse-based soil-moisture sensors, springs, flywheels, compliant mechanisms, micromechanical suspensions, etc. The latter two use kineto-elastic characteristics of deformable components. The framework, when used for materials selection, can be seen as an extension of Ashby’s materials selection method. This is also illustrated with two examples.