M. R. Mosavi, M. Khishe, Y. Hatam Khani, M. Shabani,
Volume 13, Issue 1 (3-2017)
Abstract
Radial Basis Function Neural Networks (RBF NNs) are one of the most applicable NNs in the classification of real targets. Despite the use of recursive methods and gradient descent for training RBF NNs, classification improper accuracy, failing to local minimum and low-convergence speed are defects of this type of network. In order to overcome these defects, heuristic and meta-heuristic algorithms have been conventional to training RBF network in the recent years. This study uses Stochastic Fractal Search Algorithm (SFSA) for training RBF NNs. The particles in the new algorithm explore the search space more efficiently by using the diffusion property, which is observed regularly in arbitrary fractals. To assess the performance of the proposed classifier, this network will be evaluated with the two benchmark datasets and a high-dimensional practical dataset (i.e., sonar). Results indicate that new classifier classifies sonar dataset six percent better than the best algorithm and its convergence speed is better than the other algorithms. Also has better performance than classic benchmark algorithms about all datasets.
Nguyen Cong Chinh,
Volume 20, Issue 3 (9-2024)
Abstract
This paper presents an intelligent meta-heuristic algorithm, named improved equilibrium optimizer (IEO), for addressing the optimization problem of multi-objective simultaneous integration of distributed generators at unity and optimal power factor in a distribution system. The main objective of this research is to consider the multi-objective function for minimizing total power loss, improving voltage deviation, and reducing integrated system operating costs with strict technical constraints. An improved equilibrium optimizer is an enhanced version of the equilibrium optimizer that can provide better performance, stability, and convergence characteristics than the original algorithm. For evaluating the effectiveness of the suggested method, the IEEE 69-bus radial distribution system is chosen as a test system, and simulation results from this method are also compared fairly with many previously existing methods for the same targets and constraints. Thanks to its ability to intelligently expand the search space and avoid local traps, the suggested method has become a robust stochastic optimization method in tackling complex optimization tasks.