The boost power factor correction (PFC) circuit is a common circuit in power electronics. Through years of experience, many designers have optimized the design of these circuits for particular applications. In this study, a new design procedure is presented that guarantees optimal results for any application. The algorithm used incorporates the principles of evolution in order to find the best design. This new design technique requires a rethinking of the traditional design process. Electrical models have been developed specifically for use with the optimization tool. One of the main focuses of this work is the implementation and verification of computationally efficient thermal and electro-magnetic interference (EMI) models for the boost PFC circuit. The EMI model presented can accurately predict noise levels into the 100's of kilohertz range. The thermal models presented provide very fast predictions and they have been adjusted to account for different thermal flows within the layout. This tuning procedure results in thermal predictions within 10% of actual measurement data. In order to further reduce the amount of analysis that the optimization tool must perform, some of the converter design has been performed using traditional methods. This part of the design is discussed in detail. Additionally, a per unit analysis of EMI and thermal levels is introduced. This new analysis method allows EMI and thermal levels to be compared on the same scale thus highlighting the tradeoffs between the both behaviors.

Instead of having fixed px and pm, this paper presents the use of fuzzy logic to adaptively tune px and pm for optimization of power electronic circuits throughout the process. By applying the Kmeans algorithm, distribution of the population in the search space is clustered in each training generation. Inferences of px and pm are performed by a fuzzy-based system that fuzzifies the relative sizes of the clusters containing the best and worst chromosomes. The proposed adaptation method is applied to optimize a buck regulator that requires satisfying some static and dynamic requirements. The optimized circuit component values, the regulator’s performance, and the convergence rate in the training are favorably compared with the GA’s using fixed px and pm Categories and Subject Descriptors D.2.2 [Evolutionary prototyping]

This thesis analyzes the procedural approach and benefits of applying optimization techniques to the design of a boost power factor correction (PFC) converter with an input electromagnetic interference (EMI) filter at the component level. The analysis is performed based on the particular minimum cost design study of a 1.15 kW unit satisfying a set of specifications. A traditional design methodology is initially analyzed and employed to obtain a first design. A continuous design optimization is then formulated and solved to gain insight into the converter design tradeoffs and particularities. Finally, a discrete optimization approach using a genetic algorithm is defined to develop a completely automated user-friendly software design tool able to provide in a short period of time globally optimum designs of the system for different sets of specifications. The software design tool is then employed to optimize the system design, and the savings with respect to the traditional design methodology are highlighted. The optimization problem formulation in both the continuous and discrete cases is presented in detail. The system design variables, objective function (system component cost) and constraints are identified. The objective function is expressed as a function of the design variables. A computationally efficient and experimentally validated model of the system, including second-order effects, allows the constraint values (also as a function of the design variables) to be obtained. iii Acknowledgments This thesis is the result of a joint effort of a number of colleagues and friends. First of all, I would like to specially thank my advisor Dr. Dushan Boroyevich, a brilliant engineer and excellent human being, who lighted my way through the course of my graduate work and...

Abstract--This paper presents an enhanced Latin Square Genetic Algorithm (LSGA). It makes the chromosomes to be more sensible to their surrounding regions. The algorithm applies orthogonal design method to every chromosome in the population to detect chromosomes with high fitness values in the surrounding regions. Orthogonal design method makes it more concise and direct to find the delegate to represent the situation of the surrounding regions. We execute the proposed algorithm to solve 15 test functions and compare it with traditional algorithm without using orthogonal design method. The results show that the proposed algorithm can find optimal or close-to-optimal solutions with higher speed and more accuracy.

Abstract- Research on adjusting the probabilities of crossover pI and mutation p. in genetic algorithms (GA’s) is one of the most significant and promising areas of investigation in evolutionary computation, since pr and p. greatly determine whether the algorithm will find a near-optimum solution or whether it will find a solution emciently. Instead ef having fixed pr and p-, this paper presents the use of iurzy logic to adaptively tunep, andp. for optimization of power electronic circuits throughout the process. By applying the K-means algorithm, distribution of the population in the search space is clustered in each training generation. Inferences ofp. sndp. are performed by a furzy-based system that fuuifies the relative sizes of the clusters containing the best and worst chromosomes. The proposed adaptation method is applied to optimize a buck regulator that requires satisfying some static and dynamic requirements. The optimized circuit component valuer, the regulator’s performance, and the convergence rate in the training are favorably compared with the GA’i using rued pr andp.. 1.

Abstract—Ant colony optimization (ACO) is typically used to search paths through graphs. The concept is based on simulating the behavior of ants in finding paths from the colony to food. Its searching mechanism is applicable for optimizing electric circuits with components, like resistors and capacitors, available in discrete values. However, power electronic circuits (PECs) generally consist of components, like inductors, manufactured in continuous values. Therefore, the traditional ACO algorithm cannot be applied directly. In this paper, an extended ACO (eACO) that can search the optimal values of components manufactured in discrete and continuous values is presented. The idea is based on using the orthogonal design method (ODM) to dynamically update the database of the components available with continuous values, so that these components will have pseudo-discrete values in the search space. To speed up the optimization process, the ODM performs local search of the best combination around the best ant. The eACO also takes the component tolerances into account in evaluating the fitness value of each ant. The proposed algorithm has been successfully used to optimize the design of a buck regulator. The predicted results have been compared with the published results available in the literature and verified with experimental measurements. Index Terms—Ant colony optimization (ACO), circuit optimization, orthogonal design method (ODM), power electronics circuits

Abstract—This correspondence presents pseudocoevolutionary genetic algorithms (GAs) for power electronic circuit (PEC) optimization. Circuit parameters are optimized through two parallel coadapted GA-based optimization processes for the power conversion stage (PCS) and feedback network (FN), respectively. Each process has tunable and untunable parametric vectors. The best candidate of the tunable vector in one process is migrated into the other process as an untunable vector through a migration controller, in which the migration strategy is adaptively controlled by a first-order projection of the maximum and minimum bounds of the fitness value in each generation. Implementation of this method is suitable for systems with parallel computation capacity, resulting in considerable improvement of the training speed. Optimization of a buck regulator for meeting requirements under large-signal changes and at steady state is illustrated. Simulation predictions are verified with experimental results. Index Terms—Evolutionary computation, genetic algorithms (GAs), power electronics. I.

Abstract. The development of power electronics results in a growing need for automatic design and optimization for power electronic circuits (PECs). This paper presents a particle swarm optimization (PSO) approach for the PECs design. The optimization problem is divided into two processes using a decoupled technique and PSO is employed to optimize the values of the circuit components in the power conversion stage (PCS) and the feedback network (FN), respectively. A simple mutation operator is also incorporated into PSO to enhance the population diversity. The algorithm is applied to the optimization of a buck regulator for meeting requirements under large-signal changes and at steady state. Compared with genetic algorithm (GA), PSO can yield more optimized values of circuit components with lower computational effort.