A new adaptive active frequency drift method for islanding detection of grid-connected inverters

It is difficult to detect islanding in distributed generation system for power balance between the output of inverter and the load as well as multi grid-connected inverters. Some of frequency or phase shift methods are ineffective when the phase match between the output of inverter and the load. A new adaptive active frequency drift method (AAFD) for islanding detection is proposed in this paper. This method makes the frequency of output current of inverters change by different way according to the different frequency of the voltage in common node. It can adaptively adjust the positive feedback disturbance deviation of frequency to break the balance of the output power or the frequency match, thus a fast islanding detection is implemented. The disturbance is intermittently injected into the current, which make the small disturbance on the grid and prevent false islanding by using two grade algorithms. The algorithms of the islanding detection are presented in the paper and the simulation model on one grid-connected inverter and multi grid-connected inverters are constructed. The simulation and experiments show that the method can adapt to the balance load, non-linear load and multi-inverters system, and there is no non-detection zone with this algorithm.