Em sentido geral, a quimiotaxia é o movimento dirigido que desenvolvem alguns seres vivos em resposta aos gradientes químicos presentes no seu ambiente. Uma bactéria é um organismo unicelular que usa a quimiotaxia como mecanismo de mobilização para encontrar os nutrientes de que precisa para sobreviver e para escapar de ambientes nocivos. Evoluída durante milhões de anos pela natureza, a quimiotaxia de bactérias é um processo altamente otimizado de busca e exploração em espaços desconhecidos. Graças aos avanços no campo da computação, as estratégias quimiotácticas das bactérias e sua excelente capacidade de busca podem ser modeladas, simuladas e emuladas para desenvolver métodos de otimização inspirados na natureza que sejam uma alternativa aos métodos já existentes. Neste trabalho, desenvolvem-se dois algoritmos baseados em estratégias quimiotácticas de bactérias: o BCBTOA (Bacterial Chemotaxis Based Topology Optimization Algorithm) e o BCMOA (Bacterial Chemotaxis Multiobjective Optimization Algorithm) os quais são um algoritmo de otimização topológica e um algoritmo de otimização multi-objetivo, respectivamente. O desempenho dos algoritmos é avaliado mediante a sua aplicação à solução de diversos problemas de prova e os resultados são comparados com os de outros algoritmos atualmente relevantes. O algoritmo de otimização multi-objetivo desenvolvido, também foi aplicado na solução de três problemas de otimização de projeto mecânico de eixos. Os resultados obtidos e os analise comparativos feitos, permitem concluir que os algoritmos desenvolvidos são altamente competitivos e demonstram o potencial do processo de quimiotaxia de bactérias como fonte de inspiração de algoritmos de otimização distribuída, contribuindo assim, a dar resposta à constante demanda por técnicas de otimização mais eficazes e robustas. / In general, chemotaxis is the biased movement developed by certain living organisms as a response to chemical gradients present in their environment. A bacterium is a unicellular organism that uses chemotaxis as a mechanism for mobilization that allows it to find nutrients needed to survive and to escape from harmful environments. Millions of years of natural evolution became bacterial chemotaxis a highly optimized process in searching and exploration of unknown spaces. Thanks to advances in the computing field, bacterial chemotactical strategies and its excellent ability in searching can be modeled, simulated and emulated developing bio-inspired optimization methods as alternatives to classical methods. Two algorithms based on bacterial chemotactical strategies were designed, developed and implemented in this work: i) the topology optimization algorithm, BCBTOA (Bacterial Chemotaxis Based Topology Optimization Algorithm) and ii) the multi-objective optimization algorithm, BCMOA (Bacterial Chemotaxis Multiobjective Optimization Algorithm). Algorithms performances were evaluated by their applications in the solution of benchmark problems and the results obtained were compared with other algorithms also relevant today. The BCMOA developed here was also applied in the solution of three mechanical design problems. The results obtained as well as the comparative analysis conducted lead to conclude that the algorithms developed were competitive. This also demonstrates the potential of bacterial chemotaxis as a process in which distributed optimization techniques can be inspired.

Vibrations of a wind turbine have a negative impact on its performance and therefore approaches to effectively control turbine vibrations are sought by wind industry. The body of previous research on wind turbine vibrations has focused on physics-based models. Such models come with limitations as some ideal assumptions do not reflect reality. In this Thesis a data-driven approach to analyze the wind turbine vibrations is introduced. Improvements in the data collection of information system allow collection of large volumes of industrial process data. Although the sufficient information is contained in collected data, they cannot be fully utilized to solve the challenging industrial modeling issues. Data-mining is a novel science offers platform to identify models or recognize patterns from large data set. Various successful applications of data mining proved its capability in extracting models accurately describing the processes of interest. The vibrations of a wind turbine originate at various sources. This Thesis focuses on mitigating vibrations with wind turbine control. Data mining algorithms are utilized to construct vibration models of a wind turbine that are represented by two parameters, drive train acceleration and tower acceleration. An evolutionary strategy algorithm is employed to optimize the wind turbine performance expressed with three objectives, power generation, vibration of wind turbine drive train, and vibration of wind turbine tower. The methodology presented in this Thesis is applicable to industrial processes other than wind industry.

Heating, ventilating and air-conditioning (HVAC) system is complex non-linear system with multi-variables simultaneously contributing to the system process. It poses challenges for both system modeling and performance optimization. Traditional modeling methods based on statistical or mathematical functions limit the characteristics of system operation and management. Data-driven models have shown powerful strength in non-linear system modeling and complex pattern recognition. Sufficient successful applications of data mining have proved its capability in extracting models accurately describing the relation of inner system. The heuristic techniques such as neural networks, support vector machine, and boosting tree have largely expanded to the modeling process of HVAC system. Evolutionary computation has rapidly merged to the center stage of solving the multi-objective optimization problem. Inspired from the biology behavior, it has shown the tremendous power in finding the optimal solution of complex problem. Different applications of evolutionary computation can be found in business, marketing, medical and manufacturing domains. The focus of this thesis is to apply the evolutionary computation approach in optimizing the performance of HVAC system. The energy saving can be achieved by implementing the optimal control setpoints with IAQ maintained at an acceptable level. A trade-off between energy saving and indoor air quality maintenance is also investigated by assigning different weights to the corresponding objective function. The major contribution of this research is to provide the optimal settings for the existing system to improve its efficiency and different preference-based operation methods to optimally utilize the resources.

In traditional optimization approaches the interaction of parameters associated with a problem is not a significant issue, but in the domain of Evolutionary Multi-Objective Optimization (EMOO) traditional genetic algorithm approaches have difficulties in optimizing problems with parameter interactions. Parameter interactions can be introduced when the search space is rotated. Genetic algorithms are referred to as being not rotationally invariant because their behavior changes depending on the orientation of the search space. Many empirical studies in single and multi-objective evolutionary optimization are done with respect to test problems which do not have parameter interactions. Such studies provide a favorably biased indication of genetic algorithm performance. This motivates the first aspect of our work; the improvement of the testing of EMOO algorithms with respect to the aforementioned difficulties that genetic algorithms experience in the presence of para meter interactions. To this end, we examine how EMOO algorithms can be assessed when problems are subject to an arbitrarily uniform degree of parameter interactions. We establish a theoretical basis for parameter interactions and how they can be measured. Furthermore, we ask the question of what difficulties a multi-objective genetic algorithm experiences on optimization problems exhibiting parameter interactions. We also ask how these difficulties can be overcome in order to efficiently find the Pareto-optimal front on such problems. Existing multi-objective test problems in the literature typically introduce parameter interactions by altering the fitness landscape, which is undesirable. We propose a new suite of test problems that exhibit parameter interactions through a rotation of the decision space, without altering the fitness landscape. In addition, we compare the performance of a number of recombination operators on these test problems. The second aspect of this work is concerned with developing an efficient multi-objective optimization algorithm which works well on problems with parameter interactions. We investigate how an evolutionary algorithm can be made more efficient on multi-objective problems with parameter interactions by developing four novel rotationally invariant differential evolution approaches. We also ask whether the proposed approaches are competitive in comparison with a state-of-the-art EMOO algorithm. We propose several differential evolution approaches incorporating directional information from the multi-objective search space in order to accelerate and direct the search. Experimental results indicate that dramatic improvements in efficiency can be achieved by directing the search towards points which are more dominant and more diverse. We also address the important issue of diversity loss in rotationally invariant vector-wise differential evolution. Being able to generate diverse solutions is critically important in order to avoid stagnation. In order to address this issue, one of the directed approaches that we examine incorporates a novel sampling scheme around better individuals in the search space. This variant is able to perform exceptionally well on the test problems with much less computational cost and scales to very high decision space dimensions even in the presence of parameter interactions.

One of the main advantages of the Gantry-Tau machine is a large accessible workspace\footprint ratio compared to many other parallel machines. The Gantry-Tau improves this ratio further by allowing a change of assembly mode without internal link collisions or collisions between the links and the moving TCP platform. In this Thesis some of the features of the Gantry-Tau structure are described and results are presented from the analysis of the kinematic, elastostatic and elastodynamic properties of the PKM. However, the optimal kinematic, elastostatic and elastodynamic design parameters of the machine are still difficult to calculate and this thesis introduces a multi-objective optimisation scheme based on the geometric approach for the workspace area, unreachable area, joint angle limitations and link collisions as well as the functional dependencies of the elements of the static matrix and the Laplace transform to define the first resonance frequency and Cartesian and torsional stiffness. The method to calculate the first resonance frequency assumes that each link and universal joint can be described by a mass-springdamper model and calculates the transfer function from a Cartesian (TCP) force or torque to Cartesian position or orientation. The geometric methods involve the simple geometric shapes (spheres, circles, segments, etc) and vectors. The functional dependencies are based on the properties between the kinematic parameters. These approaches are significantly faster than analytical methods based on the inverse kinematics or the general Finite Elements Method (FEM). The reconfigurable Gantry-Tau kinematic design obtained by multi-objective optimisation gives the following features: • Workspace/footprint ratio more than 3.19. • First resonance frequency greater than 48 Hz. • Lowest Cartesian stiffness in the workspace 5N/μm. • The unreachable space in the middle of the workspace is not detected. • No link collisions. The results show that by careful design of the PKM, a collision free workspace without the unreachable area in the middle can be achieved. High stiffness and high first resonance frequency are important parameters for the the Gantry-Tau when used in industrial applications, such as cutting, milling and drilling of steel or aluminium and pick-and-place operations. These applications require high static and dynamic accuracy in combination with high speed and acceleration. The optimisation parameters are the support frame lengths, actuator positions,endeffector kinematics and the robot’s arm lengths. Because of the fast computational speed of the geometric approaches and computational time saving of the methods based on the functional dependency, they are ideal for inclusion in a design optimisation framework, normally a nonlinear optimisation routine. In this Thesis the evolutionary algorithm based on the complex search method is used to optimise the 3-DOF Gantry-Tau. The existing lab prototype of this machine was assembled and completed at the University of Agder

The present work develops several methodologies for solving engineering analysis and design problems involving uncertainties and evidences from multiple sources. The influence of uncertainties on the safety/failure of the system and on the warranty costs (to the manufacturer) are also investigated. Both single and multiple objective optimization problems are considered. A methodology is developed to combine the evidences available from single or multiple sources in the presence (or absence) of credibility information of the sources using modified Dempster Shafer Theory (DST) and Fuzzy Theory in the design of uncertain engineering systems. To optimally design a system, multiple objectives, such as to maximize the belief for the overall safety of the system, minimize the deflection, maximize the natural frequency and minimize the weight of an engineering structure under both deterministic and uncertain parameters, and subjected to multiple constraints are considered. We also study the various combination rules like Dempster's rule, Yager's rule, Inagaki's extreme rule, Zhang's center combination rule and Murphy's average combination rule for combining evidences from multiple sources. These rules are compared and a selection procedure was developed to assist the analyst in selecting the most suitable combination rule to combine various evidences obtained from multiple sources based on the nature of evidence sets. A weighted Dempster Shafer theory for interval-valued data (WDSTI) and weighted fuzzy theory for intervals (WFTI) were proposed for combining evidence when different credibilities are associated with the various sources of evidence. For solving optimization problems which cannot be solved using traditional gradient-based methods (such as those involving nonconvex functions and discontinuities), a modified Particle Swarm Optimization (PSO) algorithm is developed to include dynamic maximum velocity function and bounce method to solve both deterministic multi-objective problems and uncertain multi-objective problems (vertex method is used in addition to the modified PSO algorithm for uncertain parameters). A modified game theory approach (MGT) is coupled with the modified PSO algorithm to solve multi-objective optimization problems. In case of problems with multiple evidences, belief is calculated for a safe design (satisfying all constraints) using the vertex method and the modified PSO algorithm is used to solve the multi-objective optimization problems. The multiobjective problem related to the design of a composite laminate simply supported beam with center load is also considered to minimize the weight and maximize buckling load using modified game theory. A comparison of different warranty policies for both repairable and non repairable products and an automobile warranty optimization problem is considered to minimize the total warranty cost of the automobile with a constraint on the total failure probability of the system. To illustrate the methodologies presented in this work, several numerical design examples are solved. We finally present the conclusions along with a brief discussion of the future scope of the research.

Sustainability concerns are increasingly shaping customers' behavior as well as companies' strategy. In this context, optimizing the supply chain with sustainability considerations is becoming a critical issue. However, work with quantitative models is still scarce. Our research contributes by revisiting classical inventory models taking sustainability concerns into account. We believe that reducing all aspects of sustainable development to a single objective is not desirable. We thus reformulate single and multi-echelon economic order quantity models as multi-objective problems. These models are then used to study several options such as buyer-supplier coordination or green technology investment. We also consider that firms are becoming increasingly proactive with respect to sustainability. We thus propose to apply multiple criteria decision aid techniques instead of considering sustainability as a constraint. In this sense, the firm may provide preference information about economic, environmental and social tradeoffs and quickly identify a satisfactory solution.

Due to the significant effects on the performance and competitiveness of aircraft, high lift devices are of extreme importance in aircraft design. The flow physics of high lift devices is so complex, that traditional one pass and multi-pass design approaches can’t reach the most optimised concept and multi-objective design optimisation (MDO) methods are increasingly explored in relation to this design task. The accuracy of the optimisation, however, depends on the accuracy of the underlying Computational Fluid Dynamics (CFD) solver. The complexity of the flow around high-lift configuration, namely transition and separation effects leads to a substantial uncertainty associated with CFD results. Particularly, the uncertainty related to the turbulence modelling aspect of the CFD becomes important. Furthermore, employing full viscous flow solvers within MDO puts severe limitations on the density of computational meshes in order to achieve a computationally feasible solution, thereby adding to the uncertainty of the outcome. This thesis explores the effect of uncertainties in CFD modelling when detailed aerodynamic analysis is required in computational design of aircraft configurations. For the purposes of this work, we select the benchmark NLR7301 multi-element airfoil (main wing and flap). This flow around this airfoil features all challenges typical for the high-lift configurations, while at the same time there is a wealth of experimental and computational data available in the literature for this case. A benchmark shape bi-objective optimization problem is formed, by trying to reveal the trade-off between lift and drag coefficients at near stall conditions. Following a detailed validation and grid convergence study, three widely used turbulence models are applied within Reynolds-Averaged Navier-Stokes (RANS) approach. K- Realizable, K- SST and Spalart-Allmaras. The results show that different turbulent models behave differently in the optimisation environment, and yield substantially different optimised shapes, while maintaining the overall optimisation trends (e.g. tendency to maximise camber for the increased lift). The differences between the models however exhibit systemic trends irrespective of the criteria for the selection of the target configuration in the Pareto front. A-posteriori error analysis is also conducted for a wide range of configurations of interest resulting from the optimisation process. Whereas Spalart-Allmaras exhibits best accuracy for the datum airfoil, the overall arrangement of the results obtained with different models in the (Lift, Drag) plane is consistent for all optimisation scenarios leading to increased confidence in the MDO/RANS CFD coupling.

No description available.

No description available.