Scholarlink

<p>The golden jackal optimization (GJO) algorithm has demonstrated remarkable capabilities in solving complex optimization problems, but it suffers from imbalances between exploration and exploitation and a tendency to become trapped in local optima. To overcome these limitations, a quantum tunneling and spatiotemporal synergy-enhanced GJO algorithm (QTGJO) is proposed. The algorithm integrates a time-dimension escape energy decay mechanism with a space-dimension population diversity perception, enabling dynamic switching among multiple search modes to maintain an adaptive balance between global exploration and local exploitation. In addition, a quantum tunneling mechanism is employed, allowing adaptive execution of long-distance global jumps or local fine-grained searches based on tunneling probability, thereby enhancing global search capability and the ability to escape local optima. QTGJO is evaluated on 23 classical benchmark functions, the CEC 2017 and CEC 2020 test suites, and several engineering design problems, and it achieves consistently superior performance compared with existing GJO variants, classical algorithms, and state-of-the-art algorithms. Furthermore, QTGJO is applied to optimize the hyperparameters of a CNN-BiLSTM-Attention model for diagnosing coking severity in ethylene cracking furnaces, yielding an 8.28% improvement in diagnostic accuracy over the baseline model. With the support of a high-performance computing platform, the hyperparameter optimization process is accelerated by 64.62%, providing a high-accuracy and low-latency diagnostic solution for petrochemical applications. The source code of QTGJO is publicly available at <a href="https://github.com/FullCourage/QTGJO">https://github.com/FullCourage/QTGJO</a>.</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

A quantum tunneling and spatiotemporal synergy-enhanced golden jackal optimization algorithm with application in coking severity diagnosis of ethylene cracking furnaces

  • 期刊论文

  • 发表日期：

    2026年1月13日

查看全文
  • Yunyi Tan,
  • Jieguang He,
  • Zhiping Peng,
  • Delong Cui,
  • Qirui Li

摘要

The golden jackal optimization (GJO) algorithm has demonstrated remarkable capabilities in solving complex optimization problems, but it suffers from imbalances between exploration and exploitation and a tendency to become trapped in local optima. To overcome these limitations, a quantum tunneling and spatiotemporal synergy-enhanced GJO algorithm (QTGJO) is proposed. The algorithm integrates a time-dimension escape energy decay mechanism with a space-dimension population diversity perception, enabling dynamic switching among multiple search modes to maintain an adaptive balance between global exploration and local exploitation. In addition, a quantum tunneling mechanism is employed, allowing adaptive execution of long-distance global jumps or local fine-grained searches based on tunneling probability, thereby enhancing global search capability and the ability to escape local optima. QTGJO is evaluated on 23 classical benchmark functions, the CEC 2017 and CEC 2020 test suites, and several engineering design problems, and it achieves consistently superior performance compared with existing GJO variants, classical algorithms, and state-of-the-art algorithms. Furthermore, QTGJO is applied to optimize the hyperparameters of a CNN-BiLSTM-Attention model for diagnosing coking severity in ethylene cracking furnaces, yielding an 8.28% improvement in diagnostic accuracy over the baseline model. With the support of a high-performance computing platform, the hyperparameter optimization process is accelerated by 64.62%, providing a high-accuracy and low-latency diagnostic solution for petrochemical applications. The source code of QTGJO is publicly available at https://github.com/FullCourage/QTGJO.