Misfire Fault Diagnosis Method for Diesel Engine Based on MEMD and Dispersion Entropy

Algorithm 2

The multivariate EMD algorithm.

(1)	Choose a uniform sampling point set on (n−1)-dimensional by the Hammersly sequence, and then establish an n-dimensional spatial direction vector.
(2)	Calculate the projection of the input signal along all the directions , and then obtain the projection set .
(3)	Find the maxima in each projection direction and its corresponding moment, and then calculate the multivariate envelope curves by spline interpolation function.
(4)	For a set of K direction vectors, calculate the mean of the envelope curves by .
(5)	The remaining value is subtracted as , and check whether satisfies the stoppage criterion. If satisfies, is the first IMF component and apply (2)∼(5) to ; otherwise, apply it to . Finally, the multivariate signal can be decomposed as .