Feature value extraction of coil current signal based on local morphological filter
HUANG Sen nian, ZHU Ge lan, WANG Zhi dong
(School of Electrio Power, South China University of Technology,Guangzhou 510640, China)


Abstract: In order to solve the problem that the extraction efficiency of the coil current signal (CCS) characteristic values (CVs) of the circuit breaker could not meet the requirements of the on line monitoring system, the characteristics of CCSs of four faults, including coil size smaller, spring torque smaller, coil resistance larger, control voltage lower, were summarized, which described the feasibility of regionalization of the CCS. After the analysis of the basic operations and structural elements of mathematical morphology and an ant jamming method of finding extremum by interval searching, a method based on mathematical morphology filtering and interval searching was presented to extract CVs. The normal state and four fault states were tested on the ZF7A 12 experimental platform and experimental data was obtained, then the proposed method was compared with the traditional wavelet derivation method (TWDM) in matlab. The experimental results show that the proposed method can extract CVs accurately and efficiently in all five states, whose error is only 4% and that of TWDM is 5%, while it takes only 25% of the time.
Key words: feature regions; morphological filter; tripping and closing coil current signal; interval detection; feature value extraction