复合材料科学与工程 ›› 2024, Vol. 0 ›› Issue (12): 69-74.DOI: 10.19936/j.cnki.2096-8000.20241228.010

• 应用研究 • 上一篇    下一篇

基于神经网络的风电叶片极限载荷预测及玻碳混合铺层结构优化

徐权威1, 郭小锋1*, 乔书杰2, 李思卿1, 车江宁1   

  1. 1.中原工学院 机电学院,郑州 450000;
    2.郑州财经学院 智能工程学院,郑州 450000
  • 收稿日期:2023-09-14 出版日期:2024-12-28 发布日期:2025-01-14
  • 通讯作者: 郭小锋(1983—),男,博士,副教授,硕士生导师,主要从事风力机气弹动力学及叶片设计方面的研究,mexf@foxmail.com。
  • 作者简介:徐权威(1995—),男,在读研究生,主要从事风力机叶片设计方面的研究。
  • 基金资助:
    国家自然科学基金(51705545);河南省科技攻关项目(222102220058,222102220091)

Prediction of the ultimate loads and structural optimization design for the wind turbine blades with glass-carbon laminate based on neural network

XU Quanwei1, GUO Xiaofeng1*, QIAO Shujie2, LI Siqing1, CHE Jiangning1   

  1. 1. School of Mechanical Science and Engineering, Zhongyuan University of Technology, Zhengzhou 450000, China;
    2. School of Intelligent Engineering, Zhengzhou College of Finance and Economics, Zhengzhou 450000, China
  • Received:2023-09-14 Online:2024-12-28 Published:2025-01-14

摘要: 为了能以实际的极限载荷对风电叶片进行铺层结构优化设计,以长度为89 m的DTU10MW风电叶片为研究对象,通过拉丁超立方实验建立了以叶根三轴布厚度、尾缘单轴布厚度、梁帽单轴布厚度、叶尖预弯量以及预弯指数作为输入变量,以叶尖变形量、叶根极限载荷为输出变量的神经网络模型,实现了对设计叶片个体极限载荷的准确预测。采用粒子群算法,对风电叶片的铺层结构进行优化设计。针对优化设计的玻碳混合叶片,采用新提出的叶片质量及成本计算方法,对其载荷特性和经济性进行了分析。本文的研究对大型风电叶片玻碳混合结构优化设计和成本评价分析具有实际的参考价值,对风电机组的轻量化设计具有重要意义。

关键词: 碳纤维复合材料, 铺层结构, 神经网络, 极限载荷预测, 优化设计

Abstract: In order to optimize the layup structure of wind turbine blades with the practical ultimate loads, the study was conducted on the DTU10MW wind turbine blade with a length of 89 m. A neural network model was developed through a Latin hypercube experiment, by using the root triaxial ply thickness, trailing edge uniaxial ply thickness, spar cap uniaxial ply thickness, pre-bend value of blade-tip, pre-bend index as input variables, and the blade tip deformation and ultimate loads of blade root as output variables. The layup structure of the wind turbine blades was optimized by using the particle swarm algorithm. For the optimized design of the glass-carbon hybrid blades, a newly proposed method for calculating blade mass and cost was used to analyze their load characteristics and economic feasibility. This research provides practical reference value for the optimization design and cost evaluation analysis of large-scale wind turbine blades with glass-carbon hybrid structures, and holds significant importance for the lightweight design of wind turbine units.

Key words: carbon fiber composites, layer structure, neural network, prediction of ultimate load, optimization design

中图分类号: