An exploratory analysis in low-dimensional objective space of the vector evaluated particle swarm optimization (VEPSO) algorithm is presented. A novel visualization technique is presented and applied to perform the exploratory analysis. The exploratory analysis together with a quantitative analysis revealed that the VEPSO algorithm continues to explore without exploiting the well-performing areas of the search space. A detailed investigation into the influence that the choice of archive implementation has on the performance of the VEPSO algorithm is presented. Both the Pareto-optimal front (POF) solution diversity and convergence towards the true POF is considered during the investigation. Attainment surfaces are investigated for their suitability in efficiently comparing two multi-objective optimization (MOO) algorithms. A new measure to objectively compare algorithms in multi-dimensional objective space, based on attainment surfaces, is presented. This measure, referred to as the porcupine measure, adapts the attainment surface measure by using a statistical test along with weighted intersection lines. Loosely based on the VEPSO algorithm, the multi-guided particle swarm optimization (MGPSO) algorithm is presented and evaluated. The results indicate that the MGPSO algorithm overcomes the weaknesses of the VEPSO algorithm and also outperforms a number of state of the art MOO algorithms on at least two benchmark test sets. / Thesis (PhD)--University of Pretoria, 2017. / Computer Science / PhD / Unrestricted

Beim Designprozess elektromagnetischer Aktoren ist die zuverlässige Bestimmung der zu erwartenden Verluste von großer Bedeutung. Während ohmsche Verluste sehr einfach bestimmt werden können, stellen Eisen-/Hystereseverluste häufig einen Unsicherheitsfaktor dar. Hier sind Herstellerangaben meist nur für einige wenige Arbeitspunkte bei harmonischem Betrieb vorhanden. Für den Einsatz in numerischen Berechnungen bei der Auslegung und Simulation solcher Aktoren ist eine detaillierte Beschreibung der ferromagnetischen Hysterese notwendig. Zu diesem Zweck werden häufig das Jiles-Atherton-Hysteresemodell und dessen Weiterentwicklungen eingesetzt. Aufgrund der Vielzahl an verfügbaren modifizierten Varianten wurde im Rahmen dieses Beitrages zunächst untersucht, welche Modellversionen zueinander kompatibel sind. So wird die Verwendung statischer und dynamischer Hysteresemodelle sowie die jeweilig dazu passende inverse Modellform bei konsistenter Parametrierung ermöglicht. Weiterhin wird die Parameteridentifikation anhand experimentell ermittelter Hysteresekurven für verschiedene Werkstoffe mit Hilfe der Particle-Swarm-Optimization vorgestellt. / The reliable determination of the expected losses is important for the design process of electromagnetic actuators. While resistive losses can be determined very easily, iron/hysteresis losses often represent an uncertainty factor. Manufacturer’s specifications are usually only available for a few operating points with harmonic excitation. A detailed description of the ferromagnetic hysteresis is necessary for the use in numerical calculations in the design and simulation of such actuators. For this purpose, the Jiles-Atherton hysteresis model and its further developments are often used. Due to the large number of available modified variants, at first an examination on which model versions are compatible with each other has been performed. This allows the use of static and dynamic hysteresis models as well as the corresponding inverse model form with consistent parameterization. Furthermore, the parameter identification based on experimentally determined hysteresis curves for different materials is presented using particle swarm optimization.

Although traditionally Wireless Sensor Network (WSNs) have been regarded as static sensor arrays used mainly for environmental monitoring, recently, its applications have undergone a paradigm shift from static to more dynamic environments, where nodes are attached to moving objects, people or animals. Applications that use WSNs in motion are broad, ranging from transport and logistics to animal monitoring, health care and military. These application domains have a number of characteristics that challenge the algorithmic design of WSNs. Firstly, mobility has a negative effect on the quality of the wireless communication and the performance of networking protocols. Nevertheless, it has been shown that mobility can enhance the functionality of the network by exploiting the movement patterns of mobile objects. Secondly, the heterogeneity of devices in a WSN has to be taken into account for increasing the network performance and lifetime. Thirdly, the WSN services should ideally assist the user in an unobtrusive and transparent way. Fourthly, energy-efficiency and scalability are of primary importance to prevent the network performance degradation. This thesis contributes toward the design of a new hybrid optimization algorithm; ENAMS (Energy optimizatioN Algorithm for Mobile Sensor networks) which is based on the Evolutionary Computation and Swarm Intelligence to increase the life time of mobile wireless sensor networks. The presented algorithm is suitable for large scale mobile sensor networks and provides a robust and energy- efficient communication mechanism by dividing the sensor-nodes into clusters, where the number of clusters is not predefined and the sensors within each cluster are not necessary to be distributed in the same density. The presented algorithm enables the sensor nodes to move as swarms within the search space while keeping optimum distances between the sensors. To verify the objectives of the proposed algorithm, the LEGO-NXT MIND-STORMS robots are used to act as particles in a moving swarm keeping the optimum distances while tracking each other within the permitted distance range in the search space.

This thesis studies the effectiveness of bio-inspired optimization algorithms in controlling adaptive antenna arrays. Smart antennas are able to automatically extract the desired signal from interferer signals and external noise. The angular pattern depends on the number of antenna elements, their geometrical arrangement, and their relative amplitude and phases. In the present work different antenna geometries are tested and compared when their array weights are optimized by different techniques. First, the Genetic Algorithm and Particle Swarm Optimization algorithms are used to find the best set of phases between antenna elements to obtain a desired antenna pattern. This pattern must meet several restraints, for example: Maximizing the power of the main lobe at a desired direction while keeping nulls towards interferers. A series of experiments show that the PSO achieves better and more consistent radiation patterns than the GA in terms of the total area of the antenna pattern. A second set of experiments use the Signal-to-Interference-plus-Noise-Ratio as the fitness function of optimization algorithms to find the array weights that configure a rectangular array. The results suggest an advantage in performance by reducing the number of iterations taken by the PSO, thus lowering the computational cost. During the development of this thesis, it was found that the initial states and particular parameters of the optimization algorithms affected their overall outcome. The third part of this work deals with the meta-optimization of these parameters to achieve the best results independently from particular initial parameters. Four algorithms were studied: Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing and Hill Climb. It was found that the meta-optimization algorithms Local Unimodal Sampling and Pattern Search performed better to set the initial parameters and obtain the best performance of the bio-inspired methods studied.

This thesis discusses clustering related works with emphasis on Particle Swarm Optimization (PSO) principles. Specifically, we review in detail the PSO clustering algorithm proposed by Van Der Merwe & Engelbrecht, the particle swarm clustering (PSC) algorithm proposed by Cohen & de Castro, Szabo’s modified PSC (mPSC), and Georgieva & Engelbrecht’s Cooperative-Multi-Population PSO (CMPSO). In this thesis, an improvement over Van Der Merwe & Engelbrecht’s PSO clustering has been proposed and tested for standard datasets. The improvements observed in those experiments vary from slight to moderate, both in terms of minimizing the cost function, and in terms of run time.

The optimal design of geothermal power plants across the entire spectrum of meaningful geothermal brine temperatures and climates is investigated, while accounting for vital real-world constraints that are typically ignored in the existing literature. The constrained design space of both double-flash and binary geothermal power plants is visualized, and it is seen that inclusion of real-world constraints is vital to determining the optimal feasible design of a geothermal power plant. The effect of varying condenser temperature on optimum plant performance and optimal design specifications is analyzed. It is shown that condenser temperature has a significant effect on optimal plant design as well. The optimum specific work output and corresponding optimal design of geothermal power plants across the entire range of brine temperatures and condenser temperatures is illustrated and tabulated, allowing a scientifically sound assessment of both feasibility and appropriate plant design under any set of conditions. The performance of genetic algorithms and particle swarm optimization are compared with respect to the constrained, non-linear, simulation-based optimization of a prototypical geothermal power plant, and particle swarm optimization is shown to perform significantly better than genetic algorithms. The Pareto-optimal front of specific work output and specific heat exchanger area is visualized and tabulated for binary and double-flash plants across the full range of potential geothermal brine inlet conditions and climates, allowing investigation of the specific trade-offs required between specific work output and specific heat exchanger area. In addition to the novel data, this dissertation research illustrates the development and use of a sophisticated analysis tool, based on multi-objective particle swarm optimization, for the optimal design of geothermal power plants.

Optimisation Swarm Particle (PSO) est basé sur une métaphore de l'interaction sociale […] en ajustant les trajectoires des vecteurs individuels, appelés «particules» conceptualisées comme des points se déplaçant dans un espace multidimensionnel. Le poids aléatoire des paramètres de contrôle est utilisé pour provoquer les particules à aller stochastiquement vers une région ayant plus de succès dans un espace tridimensionnel. Les particules itératives ajustent leur vitesse et leur direction en fonction de leurs personnels et des meilleures positions dans l'essaim. PSO a été appliquée avec succès pour optimiser une large gamme de problèmes. Cependant, les algorithmes standard PSO sont facilement piégés dans les points locaux suboptimaux lorsqu'il est appliqué à des problèmes avec de nombreux extrema locaux ou avec des contraintes. Cette thèse présente plusieurs algorithmes / techniques pour améliorer la capacité de l'OPS recherche mondiale: 1) Deux nouveaux algorithmes chaotiques de particules essaim d'optimisation, d'avoir une chaotiques Hopfield Neural Network (HNN) la structure, sont proposées. L'utilisation d'un système chaotique pour déterminer les poids des particules aide des algorithmes OSP pour échapper à des extrema locaux et de trouver l'optimum global. 2) Pour les algorithmes existants OSP, la relation et l'influence compter que sur les composants correspondants dimensions de l'essaim de particules. Pour montrer la relation intérieure entre les différentes composantes d'une particule, les réseaux de neurones peuvent être utilisés pour modéliser les projections d'ordre du problème d'optimisation, et une optimisation des intérieurs entièrement connecté essaim de particules est proposé à cet effet. 3) En raison de la complexité des contraintes, une solution déterministe générale est souvent difficile à trouver. Par conséquent, une particule détendue contrainte optimisation par essaim algorithme est proposé. Cette méthode améliore la capacité de recherche de l'OSP. 4) Pour améliorer les performances de l'optimisation par essaim de particules, une méthode adaptative de particules essaim d'optimisation basée sur les tests d'hypothèses sont proposées. Cette méthode applique un test d'hypothèse pour déterminer si le piège des particules dans un minimum local ou non. 5) Afin de renforcer la capacité du MPSO de recherche globale, une approche adaptative multi-objectif l'optimisation par essaim de particules (MOPSO) est proposé. Les résultats de simulation et d'analyse confirment l'efficacité des algorithmes proposés / techniques par rapport à l'autre état d'algorithmes / Particle Swarm Optimisation (PSO) is based on a metaphor of social interaction such as birds flocking or fish schooling to search a space by adjusting the trajectories of individual vectors, called “particles” conceptualized as moving points in a multidimensional space. The random weights of the control parameters are used to cause the particles to stochastically move towards a successful region in a higher dimensional space. Particles iteratively adjust their speed and direction based on their personal best positions and the best position in the swarm. PSO has been successfully applied to optimise a wide range of problems. However, the standard PSO algorithms are easily trapped in local suboptimal points when applied to problems with many local extrema or with constraints. This thesis presents several algorithms/techniques to improve the PSO's global search ability: 1) Two new chaotic particle swarm optimisation algorithms, having a chaotic Hopfield Neural Network (HNN) structure, are proposed. Using a chaotic system to determine particle weights helps the PSO algoritms to escape from local extrema and to find the global optimum. 2) For the existing PSO algorithms, the relationship and influence only rely on the corresponding dimensional components of the particle swarm. To show the inner relationship among the different components of one particle, neural networks can be used to model the characteristcs of the optimisation problem, and an inner fully connected particle swarm optimisation is proposed for this purpose. 3) Due to the complexity of constraints, a general deterministic solution is often hard to find. Therefore, a relaxed constraint particle swarm optimisation algorithm is proposed. This method improves the PSO's search ability. 4) To improve the performance of particle swarm optimisation, an adaptive particle swarm optimisation method based on hypothesis testing is proposed. This method applies a hypothesis test to determine whether the particles trap into a local minimum or not. 5) To enhance the MPSO's global search ability, an adaptive multi-objective particle swarm optimisation (MOPSO) is proposed. Simulation and analytical results confirm the efficiency of the proposed algorithms/techniques when compared to the other state of the art algorithms

This dissertation has three contributions in the area of path planning for Unmanned Aerial Vehicle (UAV) Search And Track (SAT) missions. These contributions are: (a) the study of a novel metric, G, used to quantify the value of the target information gained during a search and track mission, (b) an optimal planning horizon that minimizes time-error of a planning horizon when interrupted by Poisson random events, and (c) a modified Particle Swarm Optimization (PSO) algorithm for search missions that uses the prior target distribution in the generation of paths rather than just in the evaluation of them. UAV route planning is an important topic with many applications. Of these, military applications are the best known. This dissertation focuses on route planning for SAT missions that jointly optimize the conflicting objectives of detecting new targets and monitoring previously detected targets. The information theoretic approach proposed here is different from and is superior to existing approaches. One of the main differences is that G quantifies the value of the target information rather than the information itself. Several examples are provided to highlight G’s desirable properties. Another important component of path planning is the selection of a planning horizon, which specifies the amount of time to include in a plan. Unfortunately, little research is available to aid in the selection of a planning horizon. The proposed planning horizon is derived in the context of plan updates triggered by Poisson random events. To our knowledge, it is the only theoretically derived horizon available making it an important contribution. While the proposed horizon is optimal in minimizing planning time errors, simulation results show that it is also near optimal in minimizing the average time needed to capture an evasive target. The final contribution is the modified PSO. Our modification is based on the idea that PSO should be provided with the target distribution for path generation. This allows the algorithm to create candidate path plans in target rich regions. The modified PSO is studied using a search mission and is used in the study of G.

<p>In the field of systems biology the task of finding optimal model parameters is a common procedure. The optimization problems encountered are often multi-modal, i.e., with several local optima. In this thesis, a class of algorithms for multi-modal problems called metaheuristics are studied. A downside of metaheuristic algorithms is that they are dependent on algorithm settings in order to yield ideal performance.This thesis studies an approach to tune these algorithm settings using user constructed test functions which are faster to evaluate than an actual biological model. A statistical procedure is constructed in order to distinguish differences in performance between different configurations. Three optimization algorithms are examined closer, namely, scatter search, particle swarm optimization, and simulated annealing. However, the statistical procedure used can be applied to any algorithm that has configurable options.The results are inconclusive in the sense that performance advantages between configurations in the test functions are not necessarily transferred onto real biological models. However, of the algorithms studied a scatter search implementation was the clear top performer in general. The set of test functions used must be studied if any further work is to be made following this thesis.In the field of systems biology the task of finding optimal model parameters is a common procedure. The optimization problems encountered are often multi-modal, i.e., with several local optima. In this thesis, a class of algorithms for multi-modal problems called metaheuristics are studied. A downside of metaheuristic algorithms is that they are dependent on algorithm settings in order to yield ideal performance.</p><p>This thesis studies an approach to tune these algorithm settings using user constructed test functions which are faster to evaluate than an actual biological model. A statistical procedure is constructed in order to distinguish differences in performance between different configurations. Three optimization algorithms are examined closer, namely, scatter search, particle swarm optimization, and simulated annealing. However, the statistical procedure used can be applied to any algorithm that has configurable options.</p><p>The results are inconclusive in the sense that performance advantages between configurations in the test functions are not necessarily transferred onto real biological models. However, of the algorithms studied a scatter search implementation was the clear top performer in general. The set of test functions used must be studied if any further work is to be made following this thesis.</p>

This thesis describes a systematical study, including multidisciplinary modeling, simulation, control, and optimization, of a fuel cell - gas turbine hybrid power system that aims to increase the system efficiency and decrease the energy costs by combining two power sources. The fuel cell-gas turbine hybrid power systems can utilize exhaust fuel and waste heat from fuel cells in the gas turbines to increase system efficiency. This research considers a hybrid power system consisting of an internally reforming solid oxide fuel cell and a gas turbine. In the hybrid power system, the anode exhaust, which contains the remainder of the fuel, is mixed with the cathode exhaust as well as an additional supply of fuel and compressed air and then burned in a catalytic oxidizer. The hot oxidizer exhaust is expanded through the turbine section, driving an electric generator. After leaving the gas turbine, the oxidizer exhaust passes through a heat recovery unit in which it preheats the compressed air that is to be supplied to the fuel cell and the oxidizer. This research concentrates on multidisciplinary modeling and simulation of the fuel cell-gas turbine hybrid power system. Different control strategies for the power sharing between the subsystems are investigated. Also, the power electronics interfaces and controls for the hybrid power system are discussed. Two different power sharing strategies are studied and compared. Simulation results are presented and analyzed. Transient response of the hybrid energy system is studied through time-domain simulation. In addition, in this effort, Particle Swarm Optimization (PSO) is used to optimize the power sharing for the hybrid power system to increase the efficiency and decrease the fuel consumption.