Abstract: In traditional VMD ( Variable Mode Decomposition) combination algorithms，there is no standard for the K value selection，which leads to a certain degree of error in signal extraction． In order to solve the problem，
a method for judging the effective IMFs of VMD based on mutual information is introduced． The algorithm
implements fuzzy optimization of the preset scale K value to extract the main features of the signal by the
difference between the original signal and the sum of the IMFs obtained by the VMD． Simulation results show that
the proposed method has the largest output SNＲ ( Signal-to-Noise Ｒatio) and minimum MSE ( Mean Square
Error) ，and realizes the fuzzy optimization of K． The effectiveness of the proposed method for filtering pipeline
leakage signals is verified by experiments．

Key words: variable mode decomposition ( VMD) , mutual information, correlation coefficient, mean square error ( MES) , pipeline leak detection