Data-driven prediction of load transfer function for high neck flexible flange bolts in lattice wind turbine towers

IF 4.3 2区工程技术 Q1 ENGINEERING, CIVIL

Structures Pub Date : 2025-09-19 DOI:10.1016/j.istruc.2025.110224

Hang Du , Chuannan Xiong , Kaoshan Dai , Junlin Heng , Yuxiao Luo , Ke Fan , Bin Wang , Ji Li

{"title":"Data-driven prediction of load transfer function for high neck flexible flange bolts in lattice wind turbine towers","authors":"Hang Du , Chuannan Xiong , Kaoshan Dai , Junlin Heng , Yuxiao Luo , Ke Fan , Bin Wang , Ji Li","doi":"10.1016/j.istruc.2025.110224","DOIUrl":null,"url":null,"abstract":"<div><div>The load transfer function (LTF) of flange bolt connections is crucial for evaluating the fatigue life of lattice wind turbine towers. Traditional LTF methods face accuracy challenges due to larger flange sizes with increasing turbine hub heights. This study validates the Schmidt/Neuper method and introduces a machine learning (ML) approach to calculate bolt internal forces. After validating the finite element model (FEM) through static testing, the study compares the LTFs of integral flange connections and simplified models with the Schmidt/Neuper method, identifying its limitations. A parameter correlation analysis leads to the creation of a FEM database, from which various machine learning models: Linear Regression (LR), k-Nearest Neighbors (KNN), Support Vector Regression (SVR), Decision Tree (DT), Gaussian Process Regression (GPR), Random Forest (RF), CatBoost (CB), and XGBoost (XG) are developed and tested. A user-friendly graphical interface is provided. The finding reveals the limitations of Schmidt/Neuper method, mainly due to its assumption of a single bolt joint and neglecting load distribution, leading to inaccuracies. The ML approach improves LTF prediction accuracy, providing a more reliable method for fatigue life assessment in flange bolt connections.</div></div>","PeriodicalId":48642,"journal":{"name":"Structures","volume":"81 ","pages":"Article 110224"},"PeriodicalIF":4.3000,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Structures","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2352012425020399","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}

引用次数: 0

Abstract

The load transfer function (LTF) of flange bolt connections is crucial for evaluating the fatigue life of lattice wind turbine towers. Traditional LTF methods face accuracy challenges due to larger flange sizes with increasing turbine hub heights. This study validates the Schmidt/Neuper method and introduces a machine learning (ML) approach to calculate bolt internal forces. After validating the finite element model (FEM) through static testing, the study compares the LTFs of integral flange connections and simplified models with the Schmidt/Neuper method, identifying its limitations. A parameter correlation analysis leads to the creation of a FEM database, from which various machine learning models: Linear Regression (LR), k-Nearest Neighbors (KNN), Support Vector Regression (SVR), Decision Tree (DT), Gaussian Process Regression (GPR), Random Forest (RF), CatBoost (CB), and XGBoost (XG) are developed and tested. A user-friendly graphical interface is provided. The finding reveals the limitations of Schmidt/Neuper method, mainly due to its assumption of a single bolt joint and neglecting load distribution, leading to inaccuracies. The ML approach improves LTF prediction accuracy, providing a more reliable method for fatigue life assessment in flange bolt connections.

查看原文本刊更多论文

栅格式风电塔架高颈柔性法兰螺栓载荷传递函数的数据驱动预测

翼缘螺栓连接的载荷传递函数（LTF）是评估格式风力发电塔疲劳寿命的关键。随着涡轮轮毂高度的增加，翼缘尺寸越来越大，传统的LTF方法面临精度挑战。本研究验证了Schmidt/Neuper方法，并引入了一种机器学习（ML）方法来计算螺栓内力。在通过静力试验验证有限元模型（FEM）后，将整体法兰连接和简化模型的ltf与Schmidt/Neuper方法进行比较，找出其局限性。参数相关性分析导致FEM数据库的创建，从中开发和测试各种机器学习模型：线性回归（LR）， k-近邻（KNN），支持向量回归（SVR），决策树（DT），高斯过程回归（GPR），随机森林（RF）， CatBoost （CB）和XGBoost （XG）。提供了用户友好的图形界面。这一发现揭示了Schmidt/Neuper方法的局限性，主要是由于它假设了单个螺栓连接并忽略了载荷分布，从而导致不准确。ML方法提高了LTF预测精度，为法兰螺栓连接的疲劳寿命评估提供了更可靠的方法。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Structures Engineering-Architecture

CiteScore

5.70

自引率

17.10%

发文量

1187

期刊介绍： Structures aims to publish internationally-leading research across the full breadth of structural engineering. Papers for Structures are particularly welcome in which high-quality research will benefit from wide readership of academics and practitioners such that not only high citation rates but also tangible industrial-related pathways to impact are achieved.