Design Optimization Using Modified Differential Evolution Algorithm DOI Open Access
Pooja Tiwari, Vishnu Narayan Mishra, Raghav Prasad Parouha

et al.

Indian Journal of Science and Technology, Journal Year: 2023, Volume and Issue: 16(43), P. 3917 - 3926

Published: Nov. 18, 2023

Objectives: The key objective of this article is to suggest a modified differential evolution (MDE) algorithm for design problem optimization particularly reactor network (RND) problem. Methods: During the few decades (DE) achieved noticeable progress and solved wide variety issues. However, DE suffers from low diversification, poor exploration ability stagnation. Hence, using concept particle swarm mechanism (PSO), suggested MDE employed new mutation operator, balance exploitation activities. Also, on basis time-varying scheme operator integrates control parameter, avoid A group 6 unconstrained benchmark functions are solved, investigate presentation algorithm. Moreover, its practical superiority further verified by solving RND Findings: experiential results show that performs well in each case with highest rate success. optimize very effectively lowest time (2.98s) fewer number function evaluations (12729). Furthermore, outcomes proposed exhibits better or at least competitive performance compared evolutionary algorithms. Novelty: balanced efficiently due use memory facts (i.e. novel operator) adapted time-varying) parameters. Keywords: Evolutionary algorithms, evolution, mutation,

Language: Английский

A systematic review of hyperparameter optimization techniques in Convolutional Neural Networks DOI Creative Commons
Mohaimenul Azam Khan Raiaan, Sadman Sakib, Nur Mohammad Fahad

et al.

Decision Analytics Journal, Journal Year: 2024, Volume and Issue: 11, P. 100470 - 100470

Published: April 24, 2024

Convolutional Neural Network (CNN) is a prevalent topic in deep learning (DL) research for their architectural advantages. CNN relies heavily on hyperparameter configurations, and manually tuning these hyperparameters can be time-consuming researchers, therefore we need efficient optimization techniques. In this systematic review, explore range of well used algorithms, including metaheuristic, statistical, sequential, numerical approaches, to fine-tune hyperparameters. Our offers an exhaustive categorization (HPO) algorithms investigates the fundamental concepts CNN, explaining role variants. Furthermore, literature review HPO employing above mentioned undertaken. A comparative analysis conducted based strategies, error evaluation accuracy results across various datasets assess efficacy methods. addition addressing current challenges HPO, our illuminates unresolved issues field. By providing insightful evaluations merits demerits objective assist researchers determining suitable method particular problem dataset. highlighting future directions synthesizing diversified knowledge, survey contributes significantly ongoing development optimization.

Language: Английский

Citations

47
Metaheuristic optimization algorithms for multi-area economic dispatch of power systems: part II—a comparative study DOI Creative Commons
Yan Wang, Guojiang Xiong

Artificial Intelligence Review, Journal Year: 2025, Volume and Issue: 58(5)

Published: Feb. 14, 2025

Language: Английский

Citations

2
Developments and Design of Differential Evolution Algorithm for Non-linear/Non-convex Engineering Optimization DOI
Pooja Tiwari, Vishnu Narayan Mishra, Raghav Prasad Parouha

et al.

Archives of Computational Methods in Engineering, Journal Year: 2024, Volume and Issue: 31(4), P. 2227 - 2263

Published: Jan. 10, 2024

Language: Английский

Citations

9
Optimizing AVR system performance via a novel cascaded RPIDD2-FOPI controller and QWGBO approach DOI Creative Commons
Serdar Ekinci, Václav Snåšel, Rizk M. Rizk‐Allah

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0299009 - e0299009

Published: May 28, 2024

Maintaining stable voltage levels is essential for power systems’ efficiency and reliability. Voltage fluctuations during load changes can lead to equipment damage costly disruptions. Automatic regulators (AVRs) are traditionally used address this issue, regulating generator terminal voltage. Despite progress in control methodologies, challenges persist, including robustness response time limitations. Therefore, study introduces a novel approach AVR control, aiming enhance efficiency. A custom optimizer, the quadratic wavelet-enhanced gradient-based optimization (QWGBO) algorithm, developed. QWGBO refines (GBO) by introducing exploration exploitation improvements. The algorithm integrates interpolation mutation wavelet strategy search Extensive tests using benchmark functions demonstrate QWGBO’s effectiveness optimization. Comparative assessments against existing algorithms recent techniques confirm superior performance. In coupled with cascaded real proportional-integral-derivative second order derivative (RPIDD 2 ) fractional-order proportional-integral (FOPI) controller, precision, stability, quick response. algorithm’s performance verified through rigorous simulations, emphasizing its optimizing complex engineering problems. analyses highlight superiority over algorithms, positioning it as promising solution system contributing advancement of robust efficient systems.

Language: Английский

Citations

6
A new metaphor-less simple algorithm based on Rao algorithms: a Fully Informed Search Algorithm (FISA) DOI Creative Commons
Mojtaba Ghasemi, Abolfazl Rahimnejad, Ebrahim Akbari

et al.

PeerJ Computer Science, Journal Year: 2023, Volume and Issue: 9, P. e1431 - e1431

Published: Aug. 4, 2023

Many important engineering optimization problems require a strong and simple algorithm to achieve the best solutions. In 2020, Rao introduced three non-parametric algorithms, known as which have garnered significant attention from researchers worldwide due their simplicity effectiveness in solving problems. our simulation studies, we developed new version of called Fully Informed Search Algorithm (FISA), demonstrates acceptable performance optimizing real-world while maintaining nature original algorithms. We evaluate suggested FISA approach by applying it optimize shifted benchmark functions, such those provided CEC 2005 2014, using design mechanical system components. compare results obtained RAO method. The outcomes indicate efficacy proposed algorithm, FISA, achieving optimized solutions for aforementioned MATLAB Codes are publicly available at https://github.com/ebrahimakbary/FISA.

Language: Английский

Citations

14
Stability of metaheuristic PID controllers in photovoltaic dc microgrids DOI Creative Commons
Elvin Yusubov,

Lala Bekirova

Informatyka Automatyka Pomiary w Gospodarce i Ochronie Środowiska, Journal Year: 2025, Volume and Issue: 15(1), P. 15 - 21

Published: March 31, 2025

This article presents the stability assessment of metaheuristic PID controllers in hierarchical control system photovoltaic DC microgrids. Stability is a critical aspect microgrid systems. are utilized at primary, secondary and tertiary levels microgrid’s system. Tuning multiple using traditional methods such as Ziegler-Nichols Cohen-Coon tuning techniques becomes challenging under dynamic conditions Metaheuristic optimization algorithms used to construct self-tuning with improved stability. Experiments performed assess controlled Results exhibit that metaheuristic-tuned microgrids achieved superior performance comparison methods.

Language: Английский

Citations

0
A Marine-Predator-Algorithm-Based Optimum FOPID Controller for Enhancing the Stability and Transient Response of Automatic Voltage Regulators DOI Creative Commons
Abdullah M. Noman,

Sulaiman Z. Almutairi,

Mokhtar Aly

et al.

Fractal and Fractional, Journal Year: 2023, Volume and Issue: 7(9), P. 690 - 690

Published: Sept. 16, 2023

An improved design optimization method for fractional-order-based proportional integral derivative (FOPID) controllers is proposed in this paper to enhance the stability and transient response of automatic voltage-regulator systems. The FOPID represents a higher degree-of-freedom controller through having five tunable parameters, compared with only three parameters integer-order PID controller. In literature, performance highly determined its parameter-determination process. Recently, optimum has found wide employment several engineering applications using metaheuristic algorithms. paper, an FOPID’s parameter AVR marine predator algorithm (MPA). MPA–FOPID verified comparing other featured newly developed comparative analysis been proven have better stability, frequency response, robustness, faster disturbance-rejection ability over studied methods paper.

Language: Английский

Citations

10
Augmenting the Stability of Automatic Voltage Regulators through Sophisticated Fractional-Order Controllers DOI Creative Commons
Emad A. Mohamed, Mokhtar Aly, Waleed Alhosaini

et al.

Fractal and Fractional, Journal Year: 2024, Volume and Issue: 8(5), P. 300 - 300

Published: May 20, 2024

The transition from traditional to renewable energy sources is a critical issue in current energy-generation systems, which aims address climate change and the increased demand for energy. This shift, however, imposes additional burdens on control systems maintain power system stability quality within predefined limits. Addressing these challenges, this paper proposes an innovative Modified Hybrid Fractional-Order (MHFO) automatic voltage regulator (AVR) equipped with fractional-order tilt integral proportional derivative filter plus second-order FOTI-PDND2N2 controller. advanced controller combines benefits of (FOTI) controller, known enhancing dynamic performance steady-state response, (PDND2N2) improve robustness adaptability. proposed MHFO stands out its nine tunable parameters, providing more extensive options than conventional three-parameter PID five-parameter FOPID Furthermore, recent optimization approach using growth optimizer (GO) has been formulated applied optimally adjust controller’s parameters simultaneously. AVR based MHFO-GO scrutinized by contrasting it various established developed algorithms. comparative study shows that surpasses other controllers stability, robustness, response speed points view.

Language: Английский

Citations

2
Solution of optimal reactive power dispatch by Lévy-flight phasor particle swarm optimization DOI Creative Commons
Milad Gil, Ebrahim Akbari, Abolfazl Rahimnejad

et al.

Intelligent Systems with Applications, Journal Year: 2024, Volume and Issue: 23, P. 200398 - 200398

Published: June 15, 2024

Optimal reactive power dispatch (ORPD) problems are important tools for the sake of security and economics systems. The ORPD nonlinear optimization to minimize real losses voltage profile enhancement by optimizing several discrete continuous control variables. This paper proposes a Lévy-flight phasor particle swarm (LPPSO) solving while considering in two standard simulation results demonstrate that LPPSO algorithm proves itself as an acceptable method reaching more optimal solution problems.

Language: Английский

Citations

2
Optimization of numerical and engineering problems using altered differential evolution algorithm DOI Creative Commons
Pooja Tiwari, Vishnu Narayan Mishra, Raghav Prasad Parouha

et al.

Results in Control and Optimization, Journal Year: 2024, Volume and Issue: 14, P. 100377 - 100377

Published: Jan. 11, 2024

In this study, an altered differential evolution (ADE) is presented for numerical and engineering problem optimization. It incorporates innovative mutation strategy with new control parameters using the perception of particle swarm optimization (PSO) process, to enhance exploration exploitation activities extra profusely increase global search capacity. Also, a crossover rate employed in ADE, attain higher convergence accuracy quality optimal solutions. Finally, novel selection introduced facilitate information sharing as well escaping local minima keeps progressing. To investigate suggested ADE performance, collection 13 classical benchmark functions, CEC2014 CEC2017 suite are solved. Furthermore, superiority applicability algorithm further demonstrated through experimentation on six famous real-life problems. The experimental statistical test outcomes, collectively indicate that compared other modern algorithms, overall exhibits superior performance. comparison results show has powerful capabilities, excellent strong ability gaining high solution.

Language: Английский

Citations

1