An automated design framework for composite mechanical metamaterials and its application to 2D pentamode materials

International Journal of Mechanical Sciences, Год журнала: 2024, Номер 276, С. 109393 - 109393

Опубликована: Май 16, 2024

Until relatively recently most mechanical metamaterial classes being studied have been composed of a single solid constituent phase and design has focused almost exclusively on structural geometry. Additional dimensions can be introduced by accepting heterogeneity varying materiality, i.e., allowing properties to vary across the metamaterial's unit cells or even from cell in domain, creating composite metamaterials. This higher dimensionality significantly expands effective property envelope, but additional complexity also presents significant hurdle. To overcome challenge, an automated framework is proposed that leverages modern evolutionary computation techniques, combined with finite element analysis for fitness evaluation, discretized voxelated domain. However, this approach introduces stochastic statistical aspects process, which requires processing successfully extract useful solutions. A case study presented used generate 2D structures exhibit pentamode-like behavior. Pentamode metamaterials, are best known extreme bulk-to-shear modulus ratios (B/G), offer unique control over elastic make particularly interesting test case. The objective was defined as maximizing B/G square It found process converges solution rapidly, generally less hundred generations. ratio values 10,000 more were obtained, largely exceeding those commonly literature experimental pentamode These generated designs feature reduced stress concentrations due elimination point-like connections between lattice struts, addresses key practical limitation diamond pentamodes. observed whatever initial variety moduli voxels evolution progressed collapsed much smaller number, often binary very stiff limited number softer at locations acted hinges.

Язык: Английский

---

A multi-stage competitive swarm optimization algorithm for solving large-scale multi-objective optimization problems

Expert Systems with Applications, Год журнала: 2024, Номер unknown, С. 125411 - 125411

Опубликована: Сен. 1, 2024

Язык: Английский

---

Exploring high-dimensional optimization by sparse and low-rank evolution strategy

Swarm and Evolutionary Computation, Год журнала: 2025, Номер 92, С. 101828 - 101828

Опубликована: Янв. 1, 2025

Язык: Английский

---

Window Method: A Plug-in-Style Large-Scale Handling Technique for Evolutionary Algorithm

Опубликована: Янв. 1, 2025

Large-scale optimization constitutes a pivotal characteristic of numerous real-world problems, where large-scale evolutionary algorithms emerge as potent instrument for addressing such intricacies. However, existing methods are typically tailored to address only particular class problems and lack the versatility be readily adapted other or generalized across diverse problem domains. To issue above, this paper proposes window method, simple yet effective enhancement that can seamlessly integrated into low-dimensional bolster their performance in optimization. Specifically, method involves grouping subset randomly selected dimensions during each iteration, restricting population's evolution within window. Furthermore, effectiveness is analyzed, improved based on insights gained, including isometric segmentation individual-level length neural network-guided element. Extensive experiments single-objective, multi-objective, constrained discrete test with attributes demonstrate proposed significantly mitigates curse dimensionality enhances EAs settings.

Язык: Английский

---

A local minima escape procedure to improve the convergence of differential evolution

Applied Soft Computing, Год журнала: 2025, Номер unknown, С. 112753 - 112753

Опубликована: Янв. 1, 2025

Язык: Английский

---

Research on large-scaled wire-bonding machine scheduling in SAT: EAHA with knowledge learning and progressive fusion decomposition

Enterprise Information Systems, Год журнала: 2025, Номер unknown

Опубликована: Янв. 30, 2025

Язык: Английский

---

Electrical Storm Optimization (ESO) Algorithm: Theoretical Foundations, Analysis, and Application to Engineering Problems

Machine Learning and Knowledge Extraction, Год журнала: 2025, Номер 7(1), С. 24 - 24

Опубликована: Март 6, 2025

The electrical storm optimization (ESO) algorithm, inspired by the dynamic nature of storms, is a novel population-based metaheuristic that employs three dynamically adjusted parameters: field resistance, intensity, and conductivity. Field resistance assesses spread solutions within search space, reflecting strategy diversity. intensity balances exploration new territories exploitation promising areas. conductivity adjusts adaptability process, enhancing algorithm’s ability to escape local optima converge on global solutions. These adjustments enable ESO adapt in real-time various scenarios, steering toward potential optima. ESO’s performance was rigorously tested against 60 benchmark problems from IEEE CEC SOBC 2022 suite 20 well-known metaheuristics. results demonstrate superior ESOs, particularly tasks requiring nuanced balance between exploitation. Its efficacy further validated through successful applications four engineering domains, highlighting its precision, stability, flexibility, efficiency. Additionally, computational costs were evaluated terms number function evaluations overhead, reinforcing status as standout choice field.

Язык: Английский

---

Hybrid Optimization Method for Social Internet of Things Service Provision Based on Community Detection

Engineering Reports, Год журнала: 2025, Номер 7(4)

Опубликована: Март 29, 2025

ABSTRACT The Internet of things (IoT) and social networks integrate into a new area called the (SIoT). SIoT is characterized as network that has enhanced intelligence awareness. Essential criteria for both IoT involve effective service provisioning determination device methods. discovery services selecting optimal solution to composite them are challenges environment. Addressing these requires efficient optimization Traditional algorithms have strengths weaknesses. For example, genetic algorithm (GA) can find global optima but suffer from diversity disappearing prematurely, whereas backtracking search (BSA) offers better exploration converges more slowly. This article proposes hybrid improved based on community detection (IGBSA‐CD) overview limitations. approach improves GA's ability integrates with advantages BSA identify suitable devices fulfill user requirements by applying optimized provision (discovery, selection, composition) in detected communities. It reduce space discovery. experimental results show suggested surpasses current clustering techniques execution time cluster quality. IGBSA‐CD rapidly produces solutions near‐optimal average success rates over 96.3% different sample sizes. fitness values each size task also exhibit similar convergence, which stabilizes at 0.2–0.3 after multiple generations. response presents it all three tasks 0.04 s. consistently lower time, even when complex. Furthermore, outperforms other approaches superior quality adaptability within

Язык: Английский

---

CLDE: a competitive learning-driven differential evolution optimization for the influence maximization problem in social networks

The Journal of Supercomputing, Год журнала: 2025, Номер 81(5)

Опубликована: Апрель 5, 2025

Язык: Английский

---

Hierarchical Design of Mechanical Metamaterials: an Application on Pentamode-like Structures

International Journal of Mechanical Sciences, Год журнала: 2025, Номер unknown, С. 110232 - 110232

Опубликована: Апрель 1, 2025

Язык: Английский

---