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

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。

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.