http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 주제어 : Multipopulation (검색결과 3 건)
- 무료
- 기관 내 무료
- 유료
-
Liu Cong,Liu Yunqing,Wu Tong,Yan Fei,Zhang Qiong 대한전기학회 2022 Journal of Electrical Engineering & Technology Vol.17 No.4
The particle swarm optimization (PSO) algorithm can easily become trapped in a local optimum when the objective function to be optimized is high-dimensional and complex. A multipopulation, cooperative, bare-bones particle swarm optimization (MCBBPSO) algorithm is proposed to solve this problem. In MCBBPSO, three types of parallel cooperative learning populations are used to optimize the objective function. By adding disturbance factors and adopting an elite reverse learning strategy, the optimization ability of the algorithm is increased. Through the “mirror wall” method, the cross-boundary particles are processed to reduce the loss of particle resources. Experiments using the CEC2013 standard test functions demonstrate the convergence accuracy and speed of the algorithm on high-dimensional and complex objective functions. Finally, MCBBPSO is applied to perform parameter optimization in support vector regression (SVR), and short-term wind speeds are then predicted using the optimized SVR approach to reduce resource consumption in a power grid.
-
Huang Chen,Zhu Junyi,Xu Mingyao 한국CDE학회 2025 Journal of computational design and engineering Vol.12 No.3
Differential evolution (DE) is widely recognized in global optimization. However, it is not immune to the issue of premature convergence. To address this challenge, a dynamic subpopulation-based DE algorithm (i.e., DyS-MPADE) is presented in this paper. In DyS-MPADE, spectral hashing clustering is introduced to adaptively adjust the population structure. The innovative approach enhances SHADE’s capacity by balancing exploration and exploitation, thereby improving its convergence properties. The population of DyS-MPADE is decomposed into subpopulations, facilitating independent evolution and emulating geographical separation. The number of subpopulations is dynamically adjusted by clustering analysis of optimal and pessimal individuals. We conducted a thorough evaluation of DyS-MPADE using the CEC 2017 benchmark functions to assess its performance of different dimensions. Compared to DE, adaptive differential evolution with optional external archive (JADE), success-history based parameter adaptation for differential evolution (SHADE), Bernstein-Levy differential evolution algorithm (BDE), Bezier Search Differential Evolution Algorithm (BeSD), Egret swarm optimization algorithm (ESOA), Artificial rabbits optimization (ARO), Sea-horse optimizer (SHO), GRO, and DO, the average ranks of DyS-MPADE are 1.76 and 1.2 in 50 and 100 dimensions, respectively. The accuracy of DyS-MPADE is superior to SHADE and other state-of-the-art DE variants. Additionally, DyS-MPADE is applied to airport gate allocation problem, where it demonstrated better performance compared to other algorithms, indicating its significance for engineering applications.
-
다중 레이블 인문 데이터의 효과적인 특징 선별을 위한 이주 개체 정제연산 기반 다중 개체군 유전알고리즘
박민우 ( Park Minwoo ),이재성 ( Lee Jaesung ) 중앙대학교 인공지능인문학연구소 2021 인공지능인문학연구 Vol.7 No.-
Multi-label feature selection is a preprocessing method that can be used to analyze, for example multi-label humanity data. In particular, a multi-population genetic algorithm is verified to exhibit a better performance for identifying an appropriate subset compared with existing genetic algorithms in that a variety of populations was preserved, and premature convergence was prevented. However, with this method, the inflow of closely related features to multi-labels is unlikely to search for the solution. This study proposes an effective multi-population genetic algorithm for multi-label feature selection. In the proposed method, a multi-population genetic algorithm with a refinement process in migrated individuals maintains a variety of populations, promotes the inflow of features closely related to multi-labels, and ultimately enhances the search performance. Experimental results indicate that the proposed method exhibit better performance than the compared multi-population algorithms.
KCI
KISS연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
