Algorithm 2 Feature extraction method of bearing fault

Input:  bearing signal sequence $f\left(t\right),t=1,2,\cdots ,m\lambda$, window width $\lambda$, wavelet packet scale l, wavelet packet function type Output:  The feature matrix $T_{m\times r}$ 1:The time-frequency feature matrix $A_{m\times s}$ is obtained by Algorithm 1; 2:The time-frequency matrix $A_{m\times s}$ is normalized. SVD is decomposed according to Equation (2), and weight is calculated according to Equation (3); 3:The matrix $D_{m\times r}$ is obtained according to Equation (16), the matrix $D_{m\times r}^{*}$ is obtained according to Equation (17); 4:The entropy weight of the matrix $D_{m\times r}^{*}$ is obtained according to Equation (18); 5:The characteristic matrix $T_{m\times r}$ of the bearing fault is obtained according to Equation (20).