A Modified Particle Swarm Optimization Algorithm for Optimizing Artificial Neural Network in Classification Tasks

Processes, Journal Year: 2022, Volume and Issue: 10(12), P. 2579 - 2579

Published: Dec. 3, 2022

Artificial neural networks (ANNs) have achieved great success in performing machine learning tasks, including classification, regression, prediction, image processing, recognition, etc., due to their outstanding training, learning, and organizing of data. Conventionally, a gradient-based algorithm known as backpropagation (BP) is frequently used train the parameters’ value ANN. However, this method has inherent drawbacks slow convergence speed, sensitivity initial solutions, high tendency be trapped into local optima. This paper proposes modified particle swarm optimization (PSO) variant with two-level phases ANN for classification. A multi-swarm approach social scheme are designed primary phase enhance population diversity solution quality, respectively. Two search operators different characteristics incorporated secondary improve algorithm’s robustness handling various problems. Finally, proposed formulated training optimize its neuron weights, biases, selection activation function based on given classification dataset. The model trained by reported outperform those existing PSO variants terms accuracy when solving majority selected datasets, suggesting potential applications challenging real-world problems, such intelligent condition monitoring complex industrial systems.

Language: Английский

---

Parameter Estimation of Solar Cells via Weighted Mean of Vectors Algorithm

Published: Oct. 26, 2022

The renewable energy sources has seen an intense attraction from the research community due to their undeniable potential in terms of sustainability. Amongst them, solar holds a unique place because global availability. However, certain challenges must be addressed appropriately integrate photovoltaic cells with power networks. In this regard, paper investigates promise new intelligent optimization technique named weighted mean vectors (also known as INFO) algorithm successfully estimating parameters single and double diode models. obtained results are compared that available approaches reported literature. proposed is demonstrated have good performance for models cells.

Language: Английский

---

Hybrid Four Vector Intelligent Metaheuristic with Differential Evolution for Structural Single-Objective Engineering Optimization

Algorithms, Journal Year: 2024, Volume and Issue: 17(9), P. 417 - 417

Published: Sept. 20, 2024

Complex and nonlinear optimization challenges pose significant difficulties for traditional optimizers, which often struggle to consistently locate the global optimum within intricate problem spaces. To address these challenges, development of hybrid methodologies is essential solving complex, real-world, engineering design problems. This paper introduces FVIMDE, a novel algorithm that synergizes Four Vector Intelligent Metaheuristic (FVIM) with Differential Evolution (DE). The FVIMDE rigorously tested evaluated across two well-known benchmark suites (i.e., CEC2017, CEC2022) an additional set 50 challenging functions. Comprehensive statistical analyses, including mean, standard deviation, Wilcoxon rank-sum test, are conducted assess its performance. Moreover, benchmarked against state-of-the-art revealing superior adaptability robustness. also applied solve five structural challenges. results highlight FVIMDE’s ability outperform existing techniques diverse range problems, confirming potential as powerful tool complex tasks.

Language: Английский

---

Antenna Array Pattern Synthesis Using Nature-Inspired Computational Techniques: A Review

Archives of Computational Methods in Engineering, Journal Year: 2023, Volume and Issue: 30(5), P. 3235 - 3269

Published: March 10, 2023

Language: Английский

---

Sooty Tern Optimization Algorithm for Economic Emission Dispatch Problem Integration with Wind Energy

Lecture notes in electrical engineering, Journal Year: 2023, Volume and Issue: unknown, P. 175 - 187

Published: Jan. 1, 2023

Language: Английский

---