Robust frequency control design on micro-grid with unknown dead-zone nonlinearity

The frequency stability is very critical for the safe operation of power system. Meanwhile, distributed energy storages and unknown governor dead band (GDB) will cause system frequency deterioration when the load-frequency control (LFC) is not able to tackle these uncertainties. The stability of an island smart grid is a challenging topic because the less power sources can be regulated to handle power unbalance. In this paper, a neural network-based adaptive sliding mode controller is designed to be as the load frequency controller for an island smart grid with electrical vehicles (EVs), load disturbances and unknown governor dead band. The on-line neural compensation technology is employed to enable the sliding mode frequency control adaptive, thus the frequency stability of power system is improved. Simulation results on a benchmark island smart grid with governors, micro-turbine, EVs, load changes are provided to illustrate the competitive performance.