Abstract




 
   

IJE TRANSACTIONS A: Basics Vol. 31, No. 10 (October 2018) 1293-1301   

downloaded Downloaded: 0   viewed Viewed: 437

  DOUBLE FUZZY C-MEANS MODEL FOR FAULT DIAGNOSIS OF BEARINGS
 
M. Heidari
 
( Received: August 05, 2017 – Accepted: February 15, 2018 )
 
 

Abstract    In this paper the fuzzy model, based on the objective function (OF) that the difference between the generalized weighted distance squared sum (F1) of all the samples to faults and the generalized weighted distance squared sum (F2) of faults to the sample global center is the smallest, one double fuzzy C-means model (DFCM) is presented . AS a new method F1 and F2 are used as a new feature vector for fault diagnosis of bearings. A new fuzzy cluster validity index and the DFCM are given at the same time. Secondly, the DFCM is applied for the fault diagnosis of the bearings in rotary machinery. Firstly a SpectraQuest machinery fault simulator was used for the machinery fault mechanism study. At last, the validity of the DFCM is tested by using the bearing data of a test rig. The results show that, the rate of the classification of the bearings is 98.83%. The clustering accuracy of the second test rig is 98.30 by the DFCM.

 

Keywords    Double fuzzy C-means model, bearing, fault diagnosis

 




International Journal of Engineering
E-mail: office@ije.ir
Web Site: http://www.ije.ir