Article
Mathematics
Mengnan Tian, Yanghan Gao, Xingshi He, Qingqing Zhang, Yanhui Meng
Summary: This paper proposes a new variant of the differential evolution (DE) algorithm to mitigate its drawbacks such as premature convergence and stagnation. It introduces a novel mutation operator and a group-based competitive control parameter setting. The new mutation operator determines the scope of guidance based on the individual's fitness value. The competitive control parameter setting divides the population into equivalent groups and updates the worst location information with the current successful parameters. The proposed algorithm also incorporates a piecewise population size reduction mechanism to enhance exploration and exploitation at different stages. Experimental results demonstrate the superiority of the proposed method compared to other DE variants and non-DE algorithms.
Article
Computer Science, Information Systems
Qiang Yang, Jia-Qi Yan, Xu-Dong Gao, Dong-Dong Xu, Zhen-Yu Lu, Jun Zhang
Summary: This paper proposes a random neighbor elite guided differential evolution (RNEGDE) algorithm to effectively solve optimization problems. It introduces a novel mutation strategy named DE/current-to-rnbest/1, which randomly selects neighbors and uses elite guidance to direct individuals to promising areas. The algorithm also utilizes Gaussian and Cauchy distributions to generate adaptive parameter values for each individual. Extensive experiments show that the proposed algorithm achieves highly competitive or even better performance compared to state-of-the-art methods.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Zhenyu Meng, Yuxin Zhong, Cheng Yang
Summary: This paper introduces a Cooperative Strategy based Differential Evolution (CS-DE) algorithm with enhanced population diversity, utilizing two similar mutation strategies to tackle complex black-box optimization problems. Experimental results demonstrate the competitiveness of the CS-DE algorithm with several state-of-the-art DE variants on the CEC2013 and CEC2014 test suites.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Libao Deng, Chunlei Li, Yanfei Lan, Gaoji Sun, Changjing Shang
Summary: The paper introduces a novel differential evolution algorithm DCDE based on a dynamic combination mutation operator and a two-level parameter regulation strategy. By dynamically combining mutation operator and control parameters, DCDE achieves a balance between global exploration ability and local exploitation ability, demonstrating superior performance compared to state-of-the-art DE variants and non-DE algorithms on IEEE CEC 2017 benchmark suite.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Dong Liu, Hao He, Qiang Yang, Yiqiao Wang, Sang-Woon Jeon, Jun Zhang
Summary: This paper proposes a simple and effective mutation scheme named DE/current-to-rwrand/1 to enhance the optimization ability of differential evolution (DE) in solving complex optimization problems. The proposed mutation strategy, called function value ranking aware differential evolution (FVRADE), balances high diversity and fast convergence of the population. Experimental results demonstrate that FVRADE outperforms several state-of-the-art methods and shows promise in solving real-world optimization problems.
SWARM AND EVOLUTIONARY COMPUTATION
(2023)
Article
Computer Science, Information Systems
Zhenyu Meng, Cheng Yang
Summary: DE algorithm is a powerful evolutionary algorithm for global optimization with great success in engineering applications. However, existing DE variants have two main weaknesses in mutation strategy and parameter control. To address these weaknesses, a novel Hip-DE algorithm is proposed with historical population-based mutation strategy and parameter adaptive mechanisms for performance enhancement.
INFORMATION SCIENCES
(2021)
Article
Mathematics
Tian-Tian Wang, Qiang Yang, Xu-Dong Gao
Summary: This paper proposes a dual elite groups-guided mutation strategy called DE/current-to-duelite/1 to solve complex optimization problems in continuous optimization. By guiding the mutation of all individuals using both the elites in the current population and the obsolete parent individuals stored in an archive, DEGGDE achieves a good balance between exploring the complex search space and exploiting the found promising regions, resulting in good optimization performance.
Article
Computer Science, Artificial Intelligence
Lianzheng Cheng, Jia-Xi Zhou, Xing Hu, Ali Wagdy Mohamed, Yun Liu
Summary: Differential evolution (DE) is an efficient algorithm for global optimization problems. This paper introduces a novel crossover rate (CR) generation scheme called fcr, adjusts control parameters using unused bimodal settings based on individual evolution status, and updates the mean value of crossover rate and scale factor using L1 norm distance. These modifications are integrated with JADE mutation strategy to propose JADEfcr and LJADEfcr. Experimental results show that JADEfcr outperforms twelve state-of-the-art algorithms in terms of robustness, stability and solution quality, while LJADEfcr is statistically competitive with nine powerful algorithms in the competition.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Multidisciplinary Sciences
Tassawar Ali, Hikmat Ullah Khan, Tasswar Iqbal, Fawaz Khaled Alarfaj, Abdullah Mohammad Alomair, Naif Almusallam
Summary: Differential evolution is an evolutionary algorithm that balances exploration and exploitation to find the optimal genes for an objective function. To address the challenge of finding this balance, a clustering-based mutation strategy called Agglomerative Best Cluster Differential Evolution (ABCDE) is proposed. ABCDE efficiently converges without getting trapped in local optima by clustering the population and avoiding poor-quality genes through adaptive crossover rate. ABCDE outperforms classical mutation strategies and random neighborhood mutation strategy in generating a population where the difference between the values of the trial vector and objective vector is less than 1% for some benchmark functions. The optimal and fast convergence of differential evolution has potential applications in weight optimization of artificial neural networks and stochastic/time-constrained environments like cloud computing.
Article
Computer Science, Hardware & Architecture
Zhiping Tan, Kangshun Li, Yuan Tian, Najla Al-Nabhan
Summary: The paper introduces a novel DE algorithm called LFLDE based on local fitness landscape for guiding mutation strategy selection. Experimental results show that the proposed algorithm outperforms five representative DE algorithms in terms of performance.
JOURNAL OF SUPERCOMPUTING
(2021)
Article
Computer Science, Information Systems
Mengnan Tian, Yanhui Meng, Xingshi He, Qingqing Zhang, Yanghan Gao
Summary: This paper introduces an enhanced adaptive differential evolution algorithm (MWADE), aiming to improve its search capability and overcome the problems of premature convergence and stagnation in differential evolution algorithm. By introducing multi-schemes mutation, weighted parameter setting, random opposition mechanism, and adaptive population size reduction, the algorithm achieves better search performance.
Article
Computer Science, Information Systems
Heba Abdel-Nabi, Mostafa Z. Ali, Arafat Awajan, Rami Alazrai, Mohammad I. Daoud, Ponnuthurai N. Suganthan
Summary: This paper proposes a novel evolutionary algorithm, Ic3-aDSF-EA, which combines the exploitative and explorative merits of two main evolutionary algorithms, Stochastic Fractal Search (SFS) and a Differential Evolution (DE) variant. The algorithm gradually emphasizes the work of the best-performing algorithm during the search process without ignoring the effects of other inferior algorithms.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Lei Peng, Zhuoming Yuan, Guangming Dai, Maocai Wang, Zhe Tang
Summary: This paper proposes an improved reinforcement learning-based hybrid differential evolution algorithm RL_HDE for global exploration and design of interplanetary trajectories. Experimental results demonstrate that RL_HDE outperforms other algorithms in terms of convergence efficiency and accuracy. RL_HDE has better performance for solving complex interplanetary trajectory design problems.
SWARM AND EVOLUTIONARY COMPUTATION
(2023)
Article
Computer Science, Artificial Intelligence
Yuzhen Li, Shihao Wang, Bo Yang, Hu Chen, Zhiqiang Wu, Hongyu Yang
Summary: This paper proposes an individual similarity population reduction strategy and improves the DE algorithm using an elite-oriented strategy. Experimental results show that this method effectively enhances the performance of the DE algorithm.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Information Systems
Zhiping Tan, Kangshun Li, Yi Wang
Summary: This paper proposes a differential evolution algorithm with an adaptive mutation operator based on fitness landscape (FLDE), which uses machine learning to choose the optimal mutation strategy. Experimental results show that the FLDE algorithm is highly competitive with five other DE algorithms.
INFORMATION SCIENCES
(2021)
Article
Automation & Control Systems
Chang Liu, Lixin Tang, Jiyin Liu, Zhenhao Tang
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2019)
Article
Metallurgy & Metallurgical Engineering
Yan-he Jia, Li-xin Tang, Zhe George Zhang, Xiao-feng Chen
JOURNAL OF IRON AND STEEL RESEARCH INTERNATIONAL
(2019)
Article
Engineering, Industrial
Peixin Ge, Ying Meng, Jiyin Liu, Lixin Tang, Ren Zhao
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2020)
Article
Engineering, Industrial
Yanyan Zhang, Gary G. Yen, Lixin Tang
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2020)
Article
Engineering, Multidisciplinary
L. J. Tang, X. P. Wang, L. X. Tang, C. Cheng, Y. Yang
ENGINEERING OPTIMIZATION
(2020)
Article
Computer Science, Information Systems
Fei Zou, Gary G. Yen, Lixin Tang
INFORMATION SCIENCES
(2020)
Article
Economics
Defeng Sun, Ying Meng, Lixin Tang, Jinyin Liu, Baobin Huang, Jiefu Yang
TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW
(2020)
Article
Automation & Control Systems
Chang Liu, Lixin Tang, Jiyin Liu
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2020)
Article
Automation & Control Systems
Guodong Zhao, Jiyin Liu, Lixin Tang, Ren Zhao, Yun Dong
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2020)
Article
Automation & Control Systems
Lixin Tang, Xiangman Song, Jiyin Liu, Chang Liu
Summary: The Estimation of Distribution Algorithm (EDA) proposed in this article utilizes Kalman filtering and a learning strategy to address issues related to nonlinearity, variable coupling, and large-scale optimization problems. Computational experiments demonstrate the effectiveness of the algorithm. In practical applications, it has the potential to optimize process control parameters for continuous production processes like blast furnaces.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2021)
Article
Automation & Control Systems
Zuocheng Li, Lixin Tang, Jiyin Liu
Summary: This article proposes a memetic algorithm based on probability learning to solve the multidimensional knapsack problem (MKP), highlighting the problem-dependent heuristics and a novel framework. Experimental results demonstrate the effectiveness and practical values of the proposed method for MKP.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Interdisciplinary Applications
Pol Arias-Melia, Jiyin Liu, Rupal Mandania
Summary: This paper examines the problem of vehicle sharing and task allocation, proposing an integer programming model and a heuristic algorithm. Results show that sharing vehicles can save on vehicle usage and reduce carbon emissions.
COMPUTERS & OPERATIONS RESEARCH
(2022)
Article
Computer Science, Information Systems
Linlin Li, Zan Wang, Xianpeng Wang, Lixin Tang
