Sectional normalization and recognization on the PV-Diagram of reciprocating compressor
Lists
Only you see this
Abstract
The shortcomings of familiar normalization method on the PV-Diagram of reciprocating compressor are analyzed in the paper. A sectional normalization method of the PV-Diagram was put forward, and a recognizing technique of fault characteristics based on support vector machines for cylinder and piston system in reciprocating compressor is introduced. Four sections of curve in the PV-Diagram indicate four stages of a gas compression cycle. After the PV-Diagram is normalized with the new method, the curvilinear curvatures are unchanged in comparison with the original diagram. The contour and shape relations between normal and fault state character curves are retained. The pressure signals collected from cylinder are normalized and treated as characteristic vectors, and the vectors are inputted into a multi-class classifier composed of many support vector machines in order to classify fault modes. The experimental results show that the method can identify faults of the cylinder and piston system more correctly.
BibTeX
@inproceedings{wang2015sectional,
author = {Jindong Wang and Yi-qi Gao and Hai-yang Zhao and Rui Cong},
title = {{Sectional normalization and recognization on the PV-Diagram of reciprocating compressor}},
booktitle = {SHILAP},
year = {2015},
}Loading related content...