引文信息：
Mohsen Zare, Mojtaba Ghasemi, Amir Zahedi, Keyvan Golalipour, Soleiman Kadkhoda. A Global Best-guided Firefly Algorithm for Engineering Problems. Journal of Bionic Engineering，2023，20（5），2359-2388.
A Global Best-guided Firefly Algorithm for Engineering Problems
Mohsen Zare, Mojtaba Ghasemi, Amir Zahedi, Keyvan Golalipour, Soleiman Kadkhoda
1 Department of Electrical Engineering, Faculty of Engineering, Jahrom University, Jahrom, 7413188941, Fras, Iran
2 Department of Electronics and Electrical Engineering, Shiraz University of Technology, Shiraz, 1387671557, Iran
3 Department of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, 1411713116, Iran
4 Department of Computer Engineering, Sari Branch, Islamic Azad University, Sari, 4816119318, Iran
5 Department of Electrical Engineering, Urmia Branch, Islamic Azad University, Urmia, 571696896, Iran
6 Centre for Artificial Intelligence Research and Optimisation, Torrens University Australia, Brisbane, QLD, 4006, Australia
7 Yonsei Frontier Lab, Yonsei University, Seoul, 03722, South Korea
8 Computer Science Department, Prince Hussein Bin Abdullah Faculty for Information Technology, Al Al-Bayt University, Mafraq, 25113, Jordan
9 Hourani Center for Applied Scientific Research, Al-Ahliyya Amman University, Amman, 19328, Jordan
10 Faculty of Information Technology, Middle East University, Amman, 11831, Jordan
11 School of Computer Sciences, Universiti Sains Malaysia, 11800, George Town, Pulau Pinang, Malaysia
12 University Research and Innovation Center, Obuda University, 1034, Budapest, Hungary
13 School of Engineering and Technology, Sunway University Malaysia, Petaling Jaya, 27500, Malaysia
14 Applied science research center, Applied science private university, Amman, 11931, Jordan
Abstract
The Firefly Algorithm (FA) is a highly efficient population-based optimization technique developed by mimicking the flashing behavior of fireflies when mating. This article proposes a method based on Differential Evolution (DE)/current-to-best/1 for enhancing the FA's movement process. The proposed modification increases the global search ability and the convergence rates while maintaining a balance between exploration and exploitation by deploying the global best solution. However, employing the best solution can lead to premature algorithm convergence, but this study handles this issue using a loop adjacent to the algorithm's main loop. Additionally, the suggested algorithm’s sensitivity to the alpha parameter is reduced compared to the original FA. The GbFA surpasses both the original and five-version of enhanced FAs in finding the optimal solution to 30 CEC2014 real parameter benchmark problems with all selected alpha values. Additionally, the CEC 2017 benchmark functions and the eight engineering optimization challenges are also utilized to evaluate GbFA’s efficacy and robustness on real-world problems against several enhanced algorithms. In all cases, GbFA provides the optimal result compared to other methods. Note that the source code of the GbFA algorithm is publicly available at https://www.optim-app.com/projects/gbfa..
Fig. W1 The process of GbFA for optimization.
Fig. W2 The convergence characteristics of the FA algorithms for some test functions.
Fig. W3 Properties of 6 first studied engineering problems.
Fig. W4 The convergence behavior of GbFA for the best design for all the studied engineering optimization problems.
Information Publisher: Mohsen Zare, Mojtaba Ghasemi, Amir Zahedi
Information Release Unit: Department of Electrical Engineering, Faculty of Engineering, Jahrom University
Information Source: https://rdcu.be/dklJd
