O. Tolga Altinoz

Calculation of optimized parameters of rectangular patch antenna using gravitational search algorithm

Gravitational Search Algorithm (GSA) is a novel optimization algorithm developed recently. Hence, it has not yet been applied for determination of the optimized parameters of microstrip patch antennas. Therefore, in this study, GSA has been applied for calculation of the length and width of the rectangular patch antenna. These parameters of rectangular patch antenna have been obtained under various resonant frequencies, substrate permittivity and thickness of the antenna.

Reference point based distributed computing for multiobjective optimization

As the computational complexity of the problem and/or the number of objectives increases, a large population has to be evaluated at each generation of algorithm, and this process needs more computational resources, or requires more time for the same computational resource. However, distributing the tasks into different processors (or cores) is a good solution in speeding up the process overall. In this study, a novel and pragmatic distributed computing approach for multiobjective evolutionary optimization algorithm is proposed. Instead of dividing the objective space into pre-defined cone-domination principles, as proposed in an earlier study, a distribution of reference points initialized on a hyper-plane spanning the entire objective space is assigned to different processors and the R-NSGA-II procedure is invoked to find respective partial efficient fronts. Our results show that the proposed distributed computing approach reduces the overall computational effort compared to that needed with a single-processor method.

Implementation of PSO-based fixed frequency sliding mode controller for buck converter

This study presents the application of the optimized Fixed Frequency Sliding Mode (FFSM) Controller for a Synchronous Buck Converter. The sliding coefficient of the controller, which affects the controller performance, is optimized by using Particle Swarm Optimization (PSO). In order to analysis the performance of the optimized controller, a DC-DC Buck converter which has a variable structure and nonlinear model is used as a test-bed. The controller design procedure and optimization approach are presented in details. Simulation and hardware implementation results are provided to demonstrate the performance of the optimized controller under load and line variations.

Evaluation of the migrated solutions for distributing reference point-based multi-objective optimization algorithms

Abstract As the number of objectives and/or dimension of a given problem increases, or a real-world optimization problem is modeled in more detail, the optimization algorithm requires more computation time if the computational resources are fixed. Therefore, some more tools are needed to be developed for deployment of these resources. The parallelization is one of these tools based on distribution of the overall problem to different computational units. In this study, a distributed computing approach for multi-objective evolutionary optimization algorithms is proposed by application of a migration policy which is based on sharing the information for inter-processor collaboration. This idea is also intensified with the crossover operator at the evolutionary algorithms where the migrated solutions are applied to the crossover operator so that the performance of the overall approach increases. Besides, a new metric is defined for evaluation of the performance of the proposed distribution methodology. The performance of the proposed approaches is evaluated via well-known two- and three-objective well-known test problems.

A multiobjective optimization approach via systematical modification of the desirability function shapes

In this study, a method for solution of the multi-objective optimization problems via the desirability function aided particle swarm optimization has been proposed. The desirability function has been applied for normalization of each objective and then aggregation to a single objective. The geometric mean of the desirability values regarding each objective has been calculated as a part of the method. On the other hand, using a single-objective optimization algorithm yields a single solution rather than a complete solution set. Hence, the idea of changing the shapes of the desirability functions is implemented in order to achieve a complete solution set; in fact, this constitutes the main theme of this study. Therefore, the multi-objective problem has been degraded to a single-objective one, and it has been solved numerous times for each alternative desirability function shape. As a result, a set of biased solutions has been obtained in a very practical manner.

Conjunction of heuristic algorithms with multidimensional scaling for localization at wireless ad-hoc networks

In wireless networks, the nodes are generally distributed in a field randomly. The topological information about the nodes is only supported from the distances between each node. This distance information may not be available due to the imperfect conditions among nodes. The positions of the distributed nodes on the field by using available distance data is called node localization problem. One of the methods which can be solved this problem is classical multidimensional algorithm (cMDS). In this study, cMDS is improved with the aid of heuristic algorithms. Three heuristic algorithms (simulated annealing, particle swarm optimization and genetic algorithm) are applied and results are compared with each other.

Particle Swarm Optimization with social exclusion and its application in electromagnetics

The behavior of Particle Swarm Optimization (PSO), a population based optimization algorithm, depends on the movements of the particles and the attractions among them. This behavior was extracted from the observations of the swarms in nature. Every swarm desires to remain powerful in order to survive in nature and to protect its descendants. Therefore, the weakest members in the swarm are isolated, and generally abandoned to live on their own resources. This act is known as social exclusion. In this research, this phenomenon is incorporated to PSO. At the early phase of time-line, the swarm is divided into two groups based on their cost/fitness values. Each group proceeds their own journey without the knowledge of other group. This new algorithm is named as Social Exclusion-PSO (SEPSO). First, the performance of this new algorithm was evaluated/compared with an inertia weight PSO via unimodal, multimodal, expended benchmark functions, and then, it is applied to the circular antenna array design problem. For each implementation, the performance of two sub-populations and the undivided population are presented to demonstrate and compare the behaviour of the socially excluded swarm. The results show that excluding the members with the worst cost values from the population increases the performance of the algorithm in terms of global best solution with approximately 20% smaller number of function evaluations.

Impact of problem dimension on the execution time of parallel particle swarm optimization implementation

In this study, parallel particle swarm optimization algorithm has been investigated as regards the impact of the problem properties on the execution time. Two major factors affect the performance of parallel evolutionary algorithms: the population size and the problem dimension. In this study, five well-know benchmark functions have been applied with different dimensions. Then, these functions have been compared as regards the execution time. Finally, uniformly distributed population has been compared with the chaotic distributed population based on the dimension and population size from previous discussion.

Prediction of knee angle from accelerometer data for microcontroller implementation of semi-active knee prosthesis

In this study, the gait phase determination from accelerometer data is discussed for semi-active leg prosthesis for microcontroller implementation. The gait phase prediction is aimed by using knee angle obtained from the image of walking subject and accelerometer data recorded synchronously in the laboratory. For the phase determination of a gait, an artificial neural network is used because of its adaptive features for variable path and user. The accelerometer and knee angle data are prepared for the training and the testing set of the artificial neural network. The applicable network structure to be used in microcontroller based artificial knee is investigated and their performances are tested in terms of the the number of neurons and data window size.

Multiobjective Hooke–Jeeves algorithm with a stochastic Newton–Raphson-like step-size method

Abstract Computational optimization algorithms are focused on the improvement of meta-heuristic algorithms in a way that they can able to handle problems with more than one objective; such improved algorithms are called multiobjective optimization algorithms. As the number of objectives is increased, the complexity of the algorithm is increased with respect to the computational cost. Because classical optimization algorithms follow the direction of descending values by calculating derivations of the function, it is possible to evaluate a classical optimization algorithm as the core of a novel multiobjective optimization algorithm. Among the classical optimization algorithms, in this study, the Hooke–Jeeves (HJ) algorithm is selected as the basis of the proposed multiobjective optimization algorithm, in which members of the proposed population-based HJ algorithm move to the Pareto front by checking two neighborhood solutions at each dimension, with a dynamic distance that is calculated by using the Newton–Raphson-like stochastic step-size method. Unlike various multiobjective optimization algorithms, the performance of the proposed algorithm is greatly dependent on the decision space dimension instead of the number of objectives. As the number of objectives increases without changing the decision dimension, the computational cost almost remains the same. In addition, the proposed algorithm can be applied to single, multiple and many objective optimization problems. In this study, initially, the behaviors of the HJ and proposed multiobjective HJ algorithms are evaluated by theoretical and graphical demonstrations. Next, the performance of the proposed method is evaluated on well-known benchmark problems, and the performance of this algorithm is compared with the Nondominated Sorting Genetic Algorithm-II (NSGA-II) algorithm by using three different metric calculations. Finally, the algorithm is applied to many-objective optimization problems, and the performance of the proposed algorithm is evaluated based on the obtained results.