欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为了准确识别转子不平衡、不对中、碰摩和油膜涡动等故障,利用小波分析对转子故障信号进行4层分解,将频率由高到低的5个分支信号作为奇异值分解（Singular Value Decomposition,SVD）矩阵的行向量,经奇异值分解后得到信号的故障特征值。通过支持向量机（Support Vector Machine,SVM）在选择不同的核函数和结构参数下比较其对转子故障诊断结果的影响。结果表明在选择最优SVM模型和参数的基础上,对SVD获得的故障特征值进行诊断,得出了准确的诊断结果。
英文摘要：
In order to accurately identify the rotor faults of imbalance,misalignment,rubbing and oil whirl,this paper adopts wavelet analysis method to decompose the fault signal to obtain the five branches from high to low frequency as the row vectors of SVD（Singular Value Decomposition）matrix,the characteristic value of fault signal can be obtained after singular value decomposition.By using SVM（Support Vector Machine）to select different kernel functions and structural parameters,the paper studies the influence of different kernel functions and structural parameters on rotor fault diagnosis result.The results show that selecting optimal SVM model and parameters to analyze the characteristic value obtained by SVD can get an accurate diagnosis result.
同期刊论文项目
同项目期刊论文
Intelligent Diagnosisof Short Hydraulic Signal Based on Improved EEMD and SVM with Few Low-dimension
期刊信息
位置：