Study on Zero Sequence Protection of Circuit Breaker Based on PCM Radial Basis Network

At present, the time-limited protection commonly used in the zero-sequence protection of circuit breakers cannot identify the component of harmonic current in the zero-sequence, which makes the circuit breakers easy to operate erroneously under normal conditions. To solve the above problems, A hybrid zero-sequence protection model is proposed, which takes into account the fast breaking of zero-sequence large current and avoids the circuit breaker mis-operation with high zero-sequence harmonic content. In combination with this model, A half-wave harmonic detection algorithm based on fuzzy clustering radial basis neural network is proposed, which fully takes into account the noise generated when the phase current is converted into MCU voltage for A/D conversion. Finally, a hardware simulation platform is designed and tested. The hardware simulation shows that the new zero-sequence model can effectively reduce the zero-sequence error and further improve the power supply stability of the power grid.