Abstract
Due to the increasing complexity of optimization problems, distributed differential evolution (DDE) has become a promising approach for global optimization. However, similar to the centralized algorithms, DDE also faces the difficulty of strategies' selection and parameters' setting. To deal with such problems effectively, this article proposes an adaptive DDE (ADDE) to relieve the sensitivity of strategies and parameters. In ADDE, three populations called exploration population, exploitation population, and balance population are co-evolved concurrently by using the master-slave multipopulation distributed framework. Different populations will adaptively choose their suitable mutation strategies based on the evolutionary state estimation to make full use of the feedback information from both individuals and the whole corresponding population. Besides, the historical successful experience and best solution improvement are collected and used to adaptively update the individual parameters (amplification factor F and crossover rate CR) and population parameter (population size N), respectively. The performance of ADDE is evaluated on all 30 widely used benchmark functions from the CEC 2014 test suite and all 22 widely used real-world application problems from the CEC 2011 test suite. The experimental results show that ADDE has great superiority compared with the other state-of-the-art DDE and adaptive differential evolution variants.
Highlights
D IFFERENTIAL evolution (DE), proposed by Storn and Price [1], is a kind of evolutionary algorithm (EA) that solves the optimization problems by the iterations of evolutionary operators, including mutation, crossover, and selection [2]–[7]
As we focus on adaptive mutation strategy selection and parameters control in distributed DE (DDE) in this article, we make a brief review on some adaptive DE (ADE) variants in the following two categories
In adaptive DDE (ADDE), three populations called exploration population, exploitation population, and balance population are co-evolved concurrently to maximize their strengths by cooperation and enhance the global optimality of DDE
Summary
D IFFERENTIAL evolution (DE), proposed by Storn and Price [1], is a kind of evolutionary algorithm (EA) that solves the optimization problems by the iterations of evolutionary operators, including mutation, crossover, and selection [2]–[7]. The mutation strategies and the parameters need to be adaptively controlled to meet the searching requirement of each individual/population. The ADDE algorithm uses a master– slave multipopulation distributed framework, cooperatively with both adaptive mutation strategy selection and adaptive parameters’ settings based on the evolutionary state estimation (ESE), historical successful experience (HSE), and best solution improvement (BSI). 2) Different populations will adaptively choose their suitable mutation strategies based on the ESE, which can make full use of the feedback information from both individuals and the whole corresponding population.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
