基于小波变换和SVR模型的陀螺仪故障预测

Fault Prediction for Gyroscopes Based on Wavelet Transform and SVR

为了准确可靠地预测陀螺仪的故障,提出一种基于小波变换和支持向量回归机的陀螺仪故障诊断和预测方法。首先,利用小波分析和支持向量机对陀螺仪故障特征进行提取,然后建立SVR的故障预测模型,最后基于实验测得的陀螺仪振动数据对该预测模型进行了仿真验证,结果显示该算法预测效果良好,是一种有效的陀螺仪故障预测算法。

Aiming at the problem that gyroscopes fault diagnosis are carried out based on Wavelet Transform and SVM,a method based on wavelet transform and support vector regression for gyroscopes fault diagnosis and prediction was proposed by analyzing partial ventilator vibration signal. Firstly,fault feature was extracted through wavelet analysis and support vector machine. Then,the SVR forecasting model was established. Finally,according to the experimental data of gyroscopes,the method was verified by simulation. Results showed that the algorithm predicts effective and it＇s a kind of effective gyroscopes fault prediction algorithm.