Scholarlink

<p>Nature-inspired metaheuristics are powerful for complex optimization yet often suffer from exploration–exploitation imbalance in nonlinear and high-dimensional settings. We propose the Stratospheric Vortex Optimization Algorithm (SVOA), which introduces a dynamics-informed tri-mechanism by coupling PV-gradient–driven radial convergence (intensified local search), Rossby-like spiral motion (enhanced global exploration), and a heavy-tailed SSW perturbation (adaptive diversification), coordinated via adaptive control. Comprehensive evaluations on CEC2017, CEC2020, and CEC2022 show consistent state-of-the-art performance: on CEC2017 (29 functions, 10/30/50/100D) SVOA achieves an average Friedman rank of 2.81 (overall 1st), outperforming GWO (7.06) and AVOA (7.61), with particularly strong 30D performance (1.79); on CEC2020 (10/20D) it attains 3.10 (overall 1st) ahead of DMO (7.05); on CEC2022 (10/20D) it reaches 3.08 (overall 1st) surpassing DMO (6.65). These correspond to ≈ 54–63% lower average ranks than leading competitors. Wilcoxon tests (α = 0.05) confirm significance, with SVOA outperforming GWO on 17–18 high-dimensional CEC2017 problems. Engineering validations further demonstrate practicality: SVOA achieves best or near-best solutions in four classic designs—spring (1.27 × 10⁻²), pressure vessel (5.885 × 10³), tubular column (2.65 × 10¹), and rolling bearing (− 2.43 × 10⁵)—while outperforming 36 state-of-the-art algorithms. These results evidence SVOA’s superior stability, scalability (up to 100D), and accuracy, and establish the PV–Rossby–SSW framework as a rigorous, interpretable paradigm for atmospheric-inspired metaheuristic design and complex engineering optimization.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Stratospheric vortex optimization algorithm: a new meta-heuristic algorithm, applied to continuous optimization and engineering applications

  • 期刊论文

  • 发表日期：

    2026年2月19日

查看全文
  • Haiquan He,
  • Haisong Huang,
  • Sen Yuan,
  • Yongpeng Zhao

摘要

Nature-inspired metaheuristics are powerful for complex optimization yet often suffer from exploration–exploitation imbalance in nonlinear and high-dimensional settings. We propose the Stratospheric Vortex Optimization Algorithm (SVOA), which introduces a dynamics-informed tri-mechanism by coupling PV-gradient–driven radial convergence (intensified local search), Rossby-like spiral motion (enhanced global exploration), and a heavy-tailed SSW perturbation (adaptive diversification), coordinated via adaptive control. Comprehensive evaluations on CEC2017, CEC2020, and CEC2022 show consistent state-of-the-art performance: on CEC2017 (29 functions, 10/30/50/100D) SVOA achieves an average Friedman rank of 2.81 (overall 1st), outperforming GWO (7.06) and AVOA (7.61), with particularly strong 30D performance (1.79); on CEC2020 (10/20D) it attains 3.10 (overall 1st) ahead of DMO (7.05); on CEC2022 (10/20D) it reaches 3.08 (overall 1st) surpassing DMO (6.65). These correspond to ≈ 54–63% lower average ranks than leading competitors. Wilcoxon tests (α = 0.05) confirm significance, with SVOA outperforming GWO on 17–18 high-dimensional CEC2017 problems. Engineering validations further demonstrate practicality: SVOA achieves best or near-best solutions in four classic designs—spring (1.27 × 10⁻²), pressure vessel (5.885 × 10³), tubular column (2.65 × 10¹), and rolling bearing (− 2.43 × 10⁵)—while outperforming 36 state-of-the-art algorithms. These results evidence SVOA’s superior stability, scalability (up to 100D), and accuracy, and establish the PV–Rossby–SSW framework as a rigorous, interpretable paradigm for atmospheric-inspired metaheuristic design and complex engineering optimization.