A hybrid intelligent busbar protection strategy using hyperbolic S‐transforms and extreme learning machines

Abstract In power systems, busbars connect important components such as generators, transmission lines, and loads. A typical fault occurrence on the busbar may result in isolation of faulty sections from other normally operating parts system resulting differential protection operation. Although main busbars' scheme is protection, its operation significantly affected by magnetic saturation current transformer (CT), particularly during external or energizing transformers. Saturation CT generate a spurious reason for malfunctioning. Previous research presented different methods to modify improve scheme. However, there has been lack comprehensive study assess efficiency regarding all involved, influencing aspects including various types, transformer, (high) resistance, angle (changing 0° 360°), sources. Thus, this study, hybrid intelligent proposed effects these factors are investigated. The strategy utilizes hyperbolic S‐transform signal processing technique extract an efficient feature that able discriminate internal faults abnormal modes, is, inrush under saturation. To obtain goal, learning‐based classification method known extreme learning machines used classify conditions based selected features. was found have low sensitivity noise accurately detect half cycle one depending resistance.