No description available.

<p>Advanced development environments are essential for efficient realization of complex industrial products. Powerful equation-based object-oriented (EOO) languages such as Modelica are successfully used for modeling and virtual prototyping complex physical systems and components. The Modelica language enables engineers to build large, sophisticated and complex models. Modelica environments should scale up and be able to handle these large models. This thesis addresses the scalability of Modelica tools by employing incremental compilation and dynamic loading. The design, implementation and evaluation of this approach is presented. OpenModelica is an open-source Modelica environment developed at PELAB in which we have implemented our strategy for incremental compilation and dynamic loading of functions. We have tested the performance of these strategies in a number of different scenarios in order to see how much of an impact they have on the compilation and execution time.</p><p>Our solution contains an overhead of one or two hash calls during runtime as it uses dynamic hashes instead of static arrays.</p>

The need is explained for a new type of computer for solving partial differential equations, the Digital Field Computer. The operation of such a machine for solving Laplace's and Poisson's equations is explained and circuits for its realisation, using incremental switching of magnetic ferrite cores, are given. Its operation is predicted by simulation on a Pegasus digital computer, which shows that it solves Laplace's equation correctly.

This work studied the effects of step size, angle, spindle speed, and feed rate on the external surface roughening, orange peel effect, observed in single point incremental forming (SPIF). Experimental results were used to estimate models to categorize the extent of orange peel roughening based on visual inspection and on surface roughness measurements. Tests were performed at very high rotational speeds and feed rates and showed various influences on surface roughness, thickness distribution, and grain size. Friction at the tool-sheet interface was also studied with a completely instrumented tool that measured and recorded torsion and forming forces through deformation strains. Coefficients of friction for each part were determined and through statistical analysis, the influence of each of the following forming parameters was established: material thickness, formed shape, tool size, step size, forming speeds (feed rate and rotational speed), and forming angle. Multidimensional response surfaces were generated to show when and under what condition friction was minimized. A new contact zone representation for SPIF was also established. This formulation used common forming parameters and geometric considerations to determine the contacting zone between the sheet and the tool. Area models were proposed for both the tangential and torsional component of friction in SPIF. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2010-02-01 16:47:17.249

Model driven development (MDD) is on the rise in software engineering and no more so than in the realm of real-time and embedded systems. Being able to leverage the code generation and validation techniques made available through MDD is worth exploring, and is the focus of much academic and industrial research. However given the iterative nature of MDD, the natural evolution of models causes test case generation to occur multiple times throughout a software modeling project. Currently, the existing process of regenerating test cases for a modified model of a system can be costly, inefficient, and even redundant. The focus of this research was to achieve an improved understanding of the impact of typical model evolution steps on both the execution of the model and its test cases, and how this impact can be mitigated by reusing previously generated test cases. In this thesis we use existing techniques for symbolic execution and test case generation to perform an analysis on example models and determine how evolution affects model artifacts; these findings were then used to classify evolution steps based on their impact. From these classifications, we were able to determine exactly how to perform updates to existing symbolic execution trees and test suites in order to obtain the resulting test suites using minimal computational resources whenever possible. The approach was implemented in a software plugin, IncreTesCaGen, that is capable of incrementally generating test cases for a subset of UML-RT models by leveraging the existing testing artifacts (symbolic execution trees and test suites), as well as presenting additional analysis results to the user. Finally, we present the results of an initial evaluation of our tool, which provides insight into the tool’s performance, the effects of model evolution on execution and test case generation, as well as design tips to produce optimal models for evolution. / Thesis (Master, Computing) -- Queen's University, 2012-09-26 14:18:50.838

In timber-concrete composite systems, timber and concrete are inherently brittle materials that behave linearly elastic in both tension and bending. However, the shear connection between the members can exhibit significant ductility. It is therefore possible to develop timber-concrete composite systems with ductile connection that behave in a ductile fashion. This study illustrates the use of an elastic-perfectly plastic analytical approach to this problem. In addition, the study proposes an incremental method for predicting the nonlinear load-deflection response of the composite system. The accuracy of the analytical model is confirmed with a computer model, and numerical solutions of the analytical model are compared to experimental results from the bending tests conducted by previous researchers. Reasonable agreement is found from the comparisons, which validates the capacity of the analytical model in predicting the structural behaviour of the timber-concrete composite systems in both elastic and post-elastic stages.

In timber-concrete composite systems, timber and concrete are inherently brittle materials that behave linearly elastic in both tension and bending. However, the shear connection between the members can exhibit significant ductility. It is therefore possible to develop timber-concrete composite systems with ductile connection that behave in a ductile fashion. This study illustrates the use of an elastic-perfectly plastic analytical approach to this problem. In addition, the study proposes an incremental method for predicting the nonlinear load-deflection response of the composite system. The accuracy of the analytical model is confirmed with a computer model, and numerical solutions of the analytical model are compared to experimental results from the bending tests conducted by previous researchers. Reasonable agreement is found from the comparisons, which validates the capacity of the analytical model in predicting the structural behaviour of the timber-concrete composite systems in both elastic and post-elastic stages.

This dissertation explores methods of reducing the oversampling ratio (OSR) of both delta-sigma modulators and incremental data converters. The first reduced-OSR architecture is the high-order cascaded delta-sigma modulator. These delta-sigma modulators are shown to reduce the in-band noise sufficiently at OSRs as low as 3 while providing power savings. The second low OSR architecture is the high-order cascaded incremental data converter which possesses signal-to-quantization noise ratio (SQNR) advantages over equivalent delta-sigma modulators at low OSRs. The final architecture is the time-interleaved incremental data converter where two designs are identified as potential methods of increasing the throughput of low OSR incremental data converters. A prototype chip is designed in 0.18um CMOS technology which can operate in three modes by simply changing the resetting clock phases. It can operate as an 8-stage pipeline analog-to-digital (A/D) converter, an 8th-order cascaded delta-sigma modulator, and an 8th-order cascaded incremental data converter with an OSR of 3.

Incremental sheet forming process is a relatively new method of forming which is increasingly being used in the industry. Complex shapes can be manufactured using this method and the forming operation doesn't require any dies. High strains of over 300 % can also be achieved. Incremental sheet forming method is used to manufacture many different components presently. Prototype examples include car headlights, tubs, train body panels and medical products. The work done in the thesis deals with the prediction of the steady state forces acting on the tool during forming. Prediction of forces generated would help to design the machine against excessive vibrations. It would help the user to protect the tool and the material blank from failure. An efficient design ensures that the tool would not get deflected out of its path while forming, improving the accuracy of the finished part. To study the forces, experiments were conducted by forming pyramid and cone shapes. An experimental arrangement was set up and experimental data was collected using a data acquisition system. The effect that the various process parameters, like the thickness of the sheet, wall angle of the part and tool diameter had on the steady state force were studied. A three dimensional model was developed using commercial finite element software ABAQUS using a new modeling technique to simulate the deformation of the sheet metal blank during incremental sheet forming. The steady state forces generated for any shape, with any set of parameters used, could be predicted using the numerical model. The advantage of having a numerical model is that the forces can be predicted without doing experiments. The model was used to predict the steady state forces developed during forming of pyramid and cone shapes. The results were compared and were seen to be reasonably close to the experimental results. Later, the numerical model was validated by forming arbitrary shapes and comparing the value obtained from simulations to the value of the measured steady state forces. The results obtained from the numerical model were seen to match very well with the experimental forces for the new shapes. The numerical model developed using the new technique was seen to predict forces to a reasonable extent with less computational time as compared to the models currently available.

Learning is one such innate general cognitive ability which has empowered the living animate entities and especially humans with intelligence. It is obtained by acquiring new knowledge and skills that enable them to adapt and survive. With the advancement of technology, a large amount of information gets amassed. Due to the sheer volume of increasing information, its analysis is humanly unfeasible and impractical. Therefore, for the analysis of massive data we need machines (such as computers) with the ability to learn and evolve in order to discover new knowledge from the analysed data. The majority of the traditional machine learning algorithms function optimally on a parametric (static) data. However, the datasets acquired in real practices are often vast, inaccurate, inconsistent, non-parametric and highly volatile. Therefore, the learning algorithms’ optimized performance can only be transitory, thus requiring a learning algorithm that can constantly evolve and adapt according to the data it processes. In light of a need for such machine learning algorithm, we look for the inspiration in humans’ innate cognitive learning ability. Active learning is one such biologically inspired model, designed to mimic humans’ dynamic, evolving, adaptive and intelligent cognitive learning ability. Active learning is a class of learning algorithms that aim to create an accurate classifier by iteratively selecting essentially important unlabeled data points by the means of adaptive querying and training the classifier on those data points which are potentially useful for the targeted learning task (Tong & Koller, 2002). The traditional active learning techniques are implemented under supervised or semi-supervised learning settings (Pang et al., 2009). Our proposed model performs the active learning in an unsupervised setting by introducing a discriminative selective sampling criterion, which reduces the computational cost by substantially decreasing the number of irrelevant instances to be learned by the classifier. The methods based on passive learning (which assumes the entire dataset for training is truly informative and is presented in advance) prove to be inadequate in a real world application (Pang et al., 2009). To overcome this limitation, we have developed Active Mode Incremental Nonparametric Discriminant Analysis (aIncNDA) which undertakes adaptive discriminant selection of the instances for an incremental NDA learning. NDA is a discriminant analysis method that has been incorporated in our selective sampling technique in order to reduce the effects of the outliers (which are anomalous observations/data points in a dataset). It works with significant efficiency on the anomalous datasets, thereby minimizing the computational cost (Raducanu & Vitri´a, 2008). NDA is one of the methods used in the proposed active learning model. This thesis presents the research on a discrimination-based active learning where NDA is extended for fast discrimination analysis and data sampling. In addition to NDA, a base classifier (such as Support Vector Machine (SVM) and k-Nearest Neighbor (k-NN)) is applied to discover and merge the knowledge from the newly acquired data. The performance of our proposed method is evaluated against benchmark University of California, Irvine (UCI) datasets, face image, and object image category datasets. The assessment that was carried out on the UCI datasets showed that Active Mode Incremental NDA (aIncNDA) performs at par and in many cases better than the incremental NDA with a lower number of instances. Additionally, aIncNDA also performs efficiently under the different levels of redundancy, but has an improved discrimination performance more often than a passive incremental NDA. In an application that undertakes the face image and object image recognition and retrieval task, it can be seen that the proposed multi-example active learning system dynamically and incrementally learns from the newly obtained images, thereby gradually reducing its retrieval (classification) error rate by the means of iterative refinement. The results of the empirical investigation show that our proposed active learning model can be used for classification with increased efficiency. Furthermore, given the nature of network data which is large, streaming, and constantly changing, we believe that our method can find practical application in the field of Internet security.