Title :
Fault Prediction of Power Supply Using SVR Optimized with QPSO
Author :
Yang, Sen ; Meng, Chen
Author_Institution :
Dept. of Missile Eng., Mech. Eng. Coll., Shijiazhuang, China
Abstract :
In this paper, fault prediction approach on power supply mixture in the system of certain guidance radar is analyzed and a fault prediction method to optimize SVR based on QPSO is proposed. The QPSO algorithm is introduced, effect factors of SVR capability are analyzed and the steps to optimize SVR parameters based on QPSO are proposed. Experimental results of the fault prediction on power supply mixture show that this approach achieves low percent of error and has achieved the expected results.
Keywords :
fault diagnosis; power supplies to apparatus; radar; regression analysis; support vector machines; QPSO-based SVR parameters optimization; SVR capability; fault prediction approach; guidance radar; power supply mixture; Accuracy; Kernel; Power supplies; Prediction algorithms; Predictive models; Support vector machines; Training; Fault Prediction; Power Supply Mixture; QPSO; SVR;
Conference_Titel :
Industrial Control and Electronics Engineering (ICICEE), 2012 International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-1-4673-1450-3
DOI :
10.1109/ICICEE.2012.219
