Generally

the alternating current（AC） leakage current of insulators is compose of very steep impulsive component and periodic component.They contain important information about the contamination and flashover condition of the insulators.So

the effective separation of the two components is important.In the paper

two methods for separation of the periodic component and steep impulsive component are presented

namely median filter algorithm and 53H algorithm.After the analysis of the shortcoming of the existing methods

it indicates that the two methods proposed by the paper can effectively identify the catastrophe point of the leakage current signal.And the steep impulsive component is correspongding to the catastrophe point of the leakage current

so which can be located accurately in different condition

then the periodic component and steep impulsive component will be separated well.To verify the effectiveness of the proposed filter

the two methods are used to separate the different leakage current of insulators measured by experiment

the results show the proposed methods can improve the separation performance effectively.In the meantime

the influence of the window length on the separation effect is analyzed by simulation

the result indicates that the two methods all have good effectiveness in large range of the window length.