The methods of data fusion for detecting the phenomenon of the surge in centrifugal superchargers gas pumping units
Keywords:
data fusion, the weighted fusion, Dempster-Shafer’s method, neural networks, Kalman filter, Fuzzy integral, The Karnaugh map, centrifugal superchargers.Abstract
The phenomenon of the surge in centrifugal superchargers (CS) gas pumping units (GPU) is inextricably
linked with processing of a large number of informative parameters. Improving of the diagnostics accuracy of the
surge can be achieved through data fusion that is combining multiple sources of data into one in order to obtain the
most reliable information for the system of surging protection and regulation. Data fusion technology is able to
cope with the problem of incomplete information and uncertainty. This paper highlighted an analysis of the known
methods of data fusion, the advantages and disadvantages of such techniques as the weighted fusion, DempsterShafer’s
method, and neural networks, Kalman filter, Fuzzy integral, The Karnaugh map. It is concluded that fuzzy
integral is best suited for detecting of the surge phenomenon.
Downloads
References
помпажних характеристик газоперекачувальних
агрегатів з газотурбінним приводом для
ефективного регулювання: автореферат дисертації
на здобуття наукового степеня кандидата
технічних наук: 05.13.07 / Юрій Євстахович
Бляут; Івано-Франківський національний технічний
університет нафти і газу. – ІваноФранківськ,
2013. – 20 с.
2 Семенцов Г.Н. Синтез однотактної системи
автоматичного захисту компресора від
помпажу / Г.Н. Семенцов // Технологічні комплекси.
– 2010. – № 2. – С. 137–151.
3 Gorodetski V. Multi-agent data fusion
systems: design and implementation issues /
V. Gorodetski, O. Karsayev, and V. Samoilov //
Proceedings of the 10th International Conference
of Telecommunication Systems – Modeling and
Analysis. – Monterey, CA, 2004. – P. 762-774.
4 Gao Junbin. Some Remarks on Kalman Filters
for the Multisensor Fusion / J. Gao, C.J.Harris
// Information Fusion Journal.– 2002. – V.3. –
P. 191–201.
5 He Y. Using genetic algorithms to detectand
configure shaft crack for rotor-bearing system./
Y.He, D. Guo, F. Chu // Computer Methods in
Applied Mechanics and Engineering. – 2001. –
V.190. – P. 5895-5905.
6 Varshney P. K. Distributed detection and
data fusion / P.K, Varshney. – New York, NY:
Springer-Verlag New York Inc., 1996. – 288 p.
7 Byington C.S. Data fusion for development
predictive diagnostic for electromechanical
systems / C. S. Byington, K. Garga // Handbook of
Sensor Fusion, CRC Press. – 2000. – P. 23-31.
8 Xiaofeng Liu. Machinery Fault Diagnostics
Based on Fuzzy Measure and Fuzzy Integral Data
Fusion Techniques / Liu Xiaofeng. – School of
Engineering Systems, 2005. – 220 p.
9 Sun Z. Z. Theory and technology of
intelligent control / Z. Sun, Z. Zhang, Z. Deng. –
Beijing: Qinghua University Press, 1997. – 373 p.
10 Wu Y. Fusing output information in neural
networks: ensemble performs better / Y. Wu,
J.I. Arribas // Proceedings of the 25th Annual
International Conference of IEEE EMBS. –
Cancun, Mexico, 2003. – P. 2265-2268.
11 Cho S.B. Multiple network fusion using
fuzzy logic / S. B. Cho, J.H. Kim // TFFF Transactions
on Neural Networks. – 1995. – V 6(2). –
P. 497- 501.
12 Grabisch M. The application of fuzzy
integrals in multi criteria decision making /
M. Grabisch // European Journal of Operational
Research. – 1996. – V. 89. – P. 445-456.
Downloads
Published
How to Cite
Issue
Section
License
Авторські права....
1.png)
1.png){kind=logo}
