Article
Materials Science, Multidisciplinary
Yousef Nikravesh, Yinwei Zhang, Jian Liu, George N. Frantziskonis
Summary: This study presents a partition based topology optimization framework for the design of large-scale problems, specifically focusing on mechanical stiffness optimization. The method utilizes physical partitioning and assigns density design variables to each spatial partition, resulting in improved computational efficiency.
MECHANICS OF MATERIALS
(2022)
Article
Computer Science, Artificial Intelligence
Ye Tian, Ruchen Liu, Xingyi Zhang, Haiping Ma, Kay Chen Tan, Yaochu Jin
Summary: The article proposes an evolutionary algorithm for solving large-scale multimodal multiobjective optimization problems, which can effectively handle problems with a large number of decision variables and outperform state-of-the-art methods in neural architecture search.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Information Systems
Pradipta Ghosh, Jonathan Bunton, Dimitrios Pylorof, Marcos A. M. Vieira, Kevin Chan, Ramesh Govindan, Gaurav S. S. Sukhatme, Paulo Tabuada, Gunjan Verma
Summary: While most networks have long lifetimes, temporary network infrastructure is often useful for special events, pop-up retail, or disaster response. An instant IoT network is quickly constructed and dismantled, and we study the synthesis of such networks in urban settings. The synthesis problem involves complex constraints such as sensor coverage, line-of-sight visibility, and network connectivity, and the challenge lies in quickly scaling to large regions while producing cost-effective solutions. We explore different representations of the synthesis problem using SMC and MILP, and develop a hierarchical synthesis technique to tackle larger problem sizes.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Review
Computer Science, Interdisciplinary Applications
Sougata Mukherjee, Dongcheng Lu, Balaji Raghavan, Piotr Breitkopf, Subhrajit Dutta, Manyu Xiao, Weihong Zhang
Summary: Large-scale structural topology optimization has been constrained by high computational costs, primarily due to solving Finite Element equations and the need for fine 3D models for accurate simulation. In recent years, there have been promising techniques to accelerate the optimization process, such as re-analysis, multi-grid solvers, model reduction, machine learning, and high-performance computing.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Cheng He, Ran Cheng, Ye Tian, Xingyi Zhang, Kay Chen Tan, Yaochu Jin
Summary: This paper proposes a multiobjective EA algorithm based on paired offspring generation for constrained large-scale optimization problems, which highlights the role of offspring generation in producing promising feasible or useful infeasible offspring solutions. The algorithm first constructs subpopulations with fixed number of neighborhood solutions using a small set of reference vectors and then adopts a pairing strategy to determine parent solutions for offspring generation.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Artificial Intelligence
Xu Yang, Juan Zou, Shengxiang Yang, Jinhua Zheng, Yuan Liu
Summary: This article proposes a fuzzy decision variables framework for large-scale multiobjective optimization. The framework divides the evolutionary process into fuzzy evolution and precise evolution stages. Fuzzy evolution blurs the decision variables to reduce the search range in the decision space for quick convergence, while precise evolution directly optimizes the actual decision variables to increase population diversity. Experimental results demonstrate that this framework significantly improves the performance and computational efficiency of multiobjective optimization algorithms in large-scale problems.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Article
Computer Science, Artificial Intelligence
Zhuanlian Ding, Lve Cao, Lei Chen, Dengdi Sun, Xingyi Zhang, Zhifu Tao
Summary: In this study, a multipopulation multimodal evolutionary algorithm based on hybrid hierarchical clustering is proposed to solve large-scale multimodal multiobjective optimization problems. The algorithm uses hybrid hierarchical clustering on subpopulations to distinguish the resources of different equivalent Pareto optimal solution sets and achieve efficient cooperative coevolution. An adaptive variation method incorporating both local and global guiding information is designed, and an improved environmental selection method based on local guiding information is conducted to improve convergence and introduce diversity. Experimental results confirm that the proposed algorithm outperforms state-of-the-art MOEAs, especially for a large number of equivalent Pareto optimal solution sets.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Amarjeet Prajapati
Summary: In this study, the performance of nine large-scale multi-objective optimization optimizers was evaluated and compared over five large-scale many-objective software clustering problems. The results showed that S3-CMA-ES and LMOSCO performed better in most cases, while H-RVEA was the worst performer.
COMPLEX & INTELLIGENT SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Xiangyu Wang, Kai Zhang, Jian Wang, Yaochu Jin
Summary: This article proposes an enhanced competitive swarm optimization algorithm assisted by a strongly convex sparse operator to address sparse multiobjective optimization problems, achieving superior performance compared to state-of-the-art methods in both test problems and application examples.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2022)
Article
Automation & Control Systems
Cheng He, Ran Cheng, Danial Yazdani
Summary: In large-scale multiobjective optimization, the proposed adaptive offspring generation method effectively generates promising candidate solutions, enhancing convergence and maintaining diversity.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Engineering, Aerospace
Zhichao Wang, Ali Y. Tamijani
Summary: This paper presents a methodology for designing material distribution and orientation of three-dimensional non-uniform lattice structures, demonstrating its effectiveness through the design of different types of lattice structures. The research focuses on the construction and optimization of lattices with varying degrees of anisotropy, and the parallelization of analysis to handle large-scale meshes for synthesizing complex lightweight lattice structures.
AEROSPACE SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Jin Ren, Feiyue Qiu, Huizhen Hu
Summary: Sparse multiobjective optimization problems are common in practical applications and are characterized by large-scale decision variables and sparse optimal solutions. This paper proposes an algorithm based on multiple sparse detection to address these problems, which can generate sparse solutions by detecting the sparsity of individuals. An enhanced sparse detection strategy using binary coefficient vectors is also proposed to improve the deficiency of local detection. Additionally, the algorithm adopts an improved weighted optimization strategy to balance exploration and optimization. The proposed algorithm, named MOEA-ESD, is compared to the current state-of-the-art algorithm to verify its effectiveness.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Engineering, Mechanical
Peter Dorffler Ladegaard Jensen, Fengwen Wang, Ignazio Dimino, Ole Sigmund
Summary: This work presents a systematic topology optimization approach for designing the morphing functionality and actuation in three-dimensional wing structures. The optimization problem aims to minimize structural compliance while enforcing morphing functionality by constraining the morphing error. A feature-mapping approach is utilized to simplify actuator geometries and enhance structural performance.
Article
Computer Science, Artificial Intelligence
Jing Jiang, Fei Han, Jie Wang, Qinghua Ling, Henry Han, Yue Wang
Summary: The paper introduces a two-stage MOEA named TS-SparseEA tailored to large-scale sparse multiobjective problems. The method integrates prior information into evolution and enables population spreading over the Pareto front through two stages. It uses a binary weight optimization framework in the first stage and an improved evolutionary algorithm with hybrid encoding and specialized matching strategy in the second stage to address LSMOPs effectively.
SWARM AND EVOLUTIONARY COMPUTATION
(2022)
Article
Computer Science, Software Engineering
Di Zhang, Xiaoya Zhai, Xiao-Ming Fu, Heming Wang, Ligang Liu
Summary: We propose a novel topology optimization method to efficiently minimize the maximum compliance for a high-resolution model bearing uncertain external loads. Our method utilizes a modified power method that quickly computes the maximum eigenvalue to evaluate the worst-case compliance. By using the adjoint variable method, we perform sensitivity analysis and update the density variables to achieve optimized models with high efficiency.
COMPUTER GRAPHICS FORUM
(2022)