Scholarlink

<p>The steel support system of industrial composite closed-circuit cooling towers is characterized by lightweight, low damping, and high fundamental frequency. Its optimal design should focus on two key aspects: steel consumption and dynamic response. Based on a preliminary optimization scheme, this study first employs experimental methods to measure the vibration acceleration of the lower structure during fan operation, obtaining both time-domain and frequency-domain response curves. Subsequently, an orthogonal experimental design is conducted using the cross-sectional areas of three types of sensitive members in the precooling section as optimization variables. A multivariate regression is then performed to derive an empirical expression for the natural frequencies. The variable ranges are constrained in accordance with relevant design codes, and a complete mathematical optimization model is constructed using the identified critical frequencies. Finally, the NSGA-II algorithm is applied to iteratively obtain a refined optimization solution. Analysis of the optimized scheme reveals that the structural weight of the precooling section is reduced by 29.519%, the frequency separation between the first-order natural frequency and the excitation frequency of the upper fan increases by 16.1%, and the higher-order natural frequencies exhibit growth rates exceeding 200%. These findings provide a feasible multi-objective optimization reference for the structural design of large-scale cooling tower support systems.</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

Optimization design of steel structure support system for cooling towers based on NSGA-II algorithm

  • 期刊论文

  • 发表日期：

    2025年11月15日

查看全文
  • Chenlu Liu,
  • Jianjun Ma,
  • Qiujun Ning,
  • Hang Yang,
  • Zhiyuan Fang,
  • Tianyi Song

摘要

The steel support system of industrial composite closed-circuit cooling towers is characterized by lightweight, low damping, and high fundamental frequency. Its optimal design should focus on two key aspects: steel consumption and dynamic response. Based on a preliminary optimization scheme, this study first employs experimental methods to measure the vibration acceleration of the lower structure during fan operation, obtaining both time-domain and frequency-domain response curves. Subsequently, an orthogonal experimental design is conducted using the cross-sectional areas of three types of sensitive members in the precooling section as optimization variables. A multivariate regression is then performed to derive an empirical expression for the natural frequencies. The variable ranges are constrained in accordance with relevant design codes, and a complete mathematical optimization model is constructed using the identified critical frequencies. Finally, the NSGA-II algorithm is applied to iteratively obtain a refined optimization solution. Analysis of the optimized scheme reveals that the structural weight of the precooling section is reduced by 29.519%, the frequency separation between the first-order natural frequency and the excitation frequency of the upper fan increases by 16.1%, and the higher-order natural frequencies exhibit growth rates exceeding 200%. These findings provide a feasible multi-objective optimization reference for the structural design of large-scale cooling tower support systems.