For effective fault status diagnosis of circuit breaker

a kind of fault classification method of circuit breaker based on the improved phase trajectory characteristics of the opening and closing coil current is proposed. First

the phase space reconstruction method is used to process the current data of the opening and closing coils of five fault status

and the mutual information method and pseudo-nearest neighbor algorithm are used to determine the delay time

τ

and the embedding dimension

m

respectively

and the current signal is reconstructed into high-dimensional space to obtain the phase trajectory. Then

the 5D fault feature set is constructed by adopting 2D overall morphological characteristics of the phase trajectory and

at the same time

the 3D local inflection point features is introduced to improve the difference of phase trajectory characteristic of different types of faults. Finally

the fault characteristic set is used as the diagnosis basis input to construct the fault diagnosis model of circuit breaker based on support vector machine. The real sample analysis results show t

hat the method proposed in this paper can accurately and stably extract the phase space characteristics of fault current signal and achieve more efficient fault diagnosis of circuit breaker.