An optimized fault diagnosis method for reciprocating air compressors based on SVM

Fault diagnosis in reciprocating air compressors is essential for continuous monitoring of their performance and thereby ensuring quality output. Support Vector Machines (SVMs) are machine learning tools based on structural risk minimization principle and have the advantageous characteristic of good generalization. For this reason, four well-known and widely used SVM based methods, one-against-one (OAO), oneagainst-all (OAA), fuzzy decision function (FDF), and DDAG have been used here and an optimized SVM based technique is proposed for classification based fault diagnosis in reciprocating air compressors. The results obtained through implementation of all five techniques are thus compared as per their accuracy rate in percentages and the performance of the proposed method with 98.03 percent accuracy rate was found to be better than all other classification methods. With the compressor datasets being complex natured, proposed method is found to be of vital importance for classification based fault diagnosis pertaining to reciprocating air compressors.