Currently

ground fault diagnosis can only diagnose with severe corrosion branches or corrosion crackingbranches.Ror this shortage

combined with electrical network theory and extended Kalman filtering algorithm inmodern signal processing

this article proposes a reliable estimation for branch resistance state based on extendedKalman filtering algorithm.Depending on the topology of grounding grids

the first step is establishing state equationand observation equation;the second step is carrying them out respectively at an estimated value of the Taylorexpansion in order to realize the linearization. By means of an iterative algorithm

it can obtain the tendency ofbranch resistance value and resistance change rate within the sampling time. Through practical examples usingPython

results compared with the rate of change in resistance calculated by electrochemical detection verify theaccuracy of the proposed method.