A method for degradation features extraction of diesel engine valve clearance based on modified complete ensemble empirical mode decomposition with adaptive noise and discriminant correlation analysis feature fusion

The health assessment of the valve clearance is a key link to realize the failure prediction and health management of the valve mechanism. To accurately evaluate the state of valve clearance, this article proposes a diesel engine valve clearance degradation feature extraction method based on modified complete ensemble empirical mode decomposition with adaptive noise and discriminant correlation analysis feature fusion algorithm. First, we use modified complete ensemble empirical mode decomposition with adaptive noise to adaptively filter the cylinder head vibration signal. Then, power spectrum entropy and improved hierarchical dispersion entropy are proposed as degenerate feature entropy. To improve the sensitivity of the degraded feature entropy to the degraded state, the discriminant correlation analysis algorithm is used to fuse the two types of feature entropy to obtain fused degraded feature entropy. Finally, the degenerate fusion features are input into the least squares support vector machine to realize the health status assessment of the valve mechanism. Through the verification of test data, the results show that the proposed method can effectively evaluate the health state of the valve clearance of diesel engines.