International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 05 | May 2023
p-ISSN: 2395-0072
www.irjet.net
Modified Mountain Gazelle Optimizer Based on Logistic Chaotic Mapping and Truncation Selection Abdul-Fatawu Seini Yussif1, Elvis Twumasi2, Emmanuel Asuming Frimpong3 1,2,3 Department of Electrical and Electronic Engineering, Kwame Nkrumah University of Science and Technology,
Kumasi, Ghana ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In this study, a Modified Mountain Gazelle
Algorithm (SSA) with a mutation strategy for global optimization [11].
Optimizer (MMGO) algorithm is presented. The proposed algorithm was designed to improve the ability of MGO in solving high-dimensional problems, increase convergence speed, and enhance stability. The modification is based on the application of a logistics chaotic mapping at the initialization stage, a modified Migration pattern in the Search of Food (MSF) phase for diversity maintenance, and a controlling factor at the Territorial and Solitary Males (TSM) phase using the truncation selection technique. The proposed algorithm was implemented in MATLAB software and its performance was tested on 23 benchmark functions, and a real-life engineering problem to prove its efficiency and adaptability. The results of the MMGO were compared with those of the basic Mountain Gazelle Optimizer (MGO), Particle Swarm Optimization (PSO), and Gravitational Search Algorithm (GSA). The findings of the work indicated that the MMGO outperformed the other state-of-the-art algorithms in terms of both optimization accuracy and computational efficiency. The results demonstrated the effectiveness and robustness of the proposed MMGO algorithm, in solving high-dimensional optimization problems in engineering and other fields.
Though these improved algorithms offer a promising avenue for solving complex problems, researchers are still in search of novel algorithms that exhibit robustness, flexibility, and the ability to handle diverse problem domains. Also, the dynamic nature of many real-world problems necessitates newly developed algorithms that are adaptive and can quickly respond to changing or evolving problem conditions [9,11]. The Mountain Gazelle Optimizer (MGO) developed by Benyamin et al [12] in 2022 is one of such algorithms. The MGO has been proven to handle problems characterized by high nonlinearity, and combinatorial complexity and has exhibited good performance when tested on some standard benchmark test functions and real-life engineering problems. However, similar to some other metaheuristic algorithms, it still suffers from optimization accuracy, slow convergence, and entrapment in suboptimal solutions when applied to complex high-dimensional optimization problems [12]. These high-dimensional and large-scale problems often exhibit a large number of variables, constraints, and interactions, making it difficult for the MGO to get an exact solution due to the exponential growth of the search space.
Key Words: Mountain Gazelle Optimizer (MGO), logistics chaotic mapping, truncation selection technique, highdimensional problem, optimization.
This is mainly associated with the poor quality of the initial population, lack of proper diversity maintenance mechanism, and lack of effective convergence control in the algorithm derivation [12]. However, in order to ensure the holistic application of this essential algorithm which is based on the social intelligence of mountain gazelles in the wildlife, there needs to be an improvement of its parameters to ensure its exploration and exploitation ability to deal with high dimensional problems.
1. INTRODUCTION The use of metaheuristic algorithms in finding optimal solutions to complex problems has seen wide applications in various fields such as engineering, finance, and computer science [1,2]. However, as the complexity and dynamism of the problem increase, metaheuristic algorithms often struggle to find the global optimum solution within a reasonable time frame and iterations [3-5]. To address this limitation, researchers have proposed modifications to metaheuristic algorithms to enhance their performance in solving complex optimization problems [6-8]. These modifications can improve the search efficiency of the metaheuristic algorithms, thereby enabling them to find better solutions. One such modification is the incorporation of problem-specific operators [9]. For instance, Ogun et al proposed a modified bull optimization algorithm for continuous optimization problems based on genetic operators [10], and Bing et al improved the Sparrow Search
© 2023, IRJET
|
Impact Factor value: 8.226
Therefore, this paper proposes a Modified Mountain Gazelle Optimizer (MMGO), an improvement of the MGO algorithm to enhance its performance in solving high-dimensional engineering problems. Three modifications are incorporated. Firstly, a Logistic Chaotic mapping [13][14] is utilized to replace the random initialization in MGO to improve the quality of the initial population. Secondly, an operator is modified to maintain diversity in the population during the execution process to avoid suboptimal solutions [15]. Thirdly, a Truncation Selection Technique [16] is adopted to determine the value of the newly introduce parameter to control the convergence speed. The effectiveness of the
|
ISO 9001:2008 Certified Journal
|
Page 1769