基于CEEMDAN-SVM的叶片故障诊断模型研究
Research on Blade Fault Diagnosis Model Based on CEEMDAN-SVM
赵翊辉1
作者信息
- 1. 沈阳工业大学化工装备学院,辽宁辽阳 111000
- 折叠
摘要
受到恶劣工作环境和高速运转等因素的影响,通风机叶片裂纹的产生成为制约其性能和安全的隐患.为解决这一问题,本研究基于振动信号处理和人工智能技术,提出了一种综合的裂纹检测模型.通过实时监测叶片振动信号,结合小波降噪技术处理原始信号,再利用CEEMDAN和希尔伯特谱分析提取裂纹特征,最终通过支持向量机进行裂纹分类,从而创建了一个裂纹检测模型.这一研究为提高矿用轴流式通风机设备的安全性和可靠性提供了新的思路.
Abstract
The occurrence of cracks in fan blades,caused by harsh working environments and high-speed operation,has become a hidden danger restricting their performance and safety.To address this issue,this study proposes a comprehensive crack detection model based on vibration signal processing and artificial intelligence techniques.By monitoring the vibration signals of fan blades in real-time,the raw signals are processed using wavelet denoising techniques.Subsequently,crack features are extracted using CEEMDAN and Hilbert-Huang Spectrum Analysis(HSA).Finally,a crack classification model is established using Support Vector Machines(SVM).This research provides a new approach to improving the safety and reliability of mine axial flow fan equipment.
关键词
轴流式通风机/CEEMDAN/HSA/支持向量机/小波降噪Key words
axial flow fan/CEEMDAN/HSA/support vector machine/wavelet denoising引用本文复制引用
出版年
2024
NETL
NSTL
万方数据