Getting parameters in power systems based on adaptive linear neural network

One of the key issues in the power system stability and control is to detect parameters quickly. The traditional fast fourier transform (FFT) and least square parameter estimation algorithms are of less practical significance owing to the slow speed caused by heavy computation burden. An approach is proposed using adaptive neural network to detect fault current at the time of vacuum interrupter synchronous breaking short-circuit, to estimate the extinguishing moment of arc for optimally breaking the contact. Taking the orthogonal filter to decrease the action of DC components so as to increase the convergence of the neurons. The step is adaptively changed based on the correlated error estimation. The approach can get the fault current after half period (10 ms). The MATLAB-based simulation shows the effectiveness and speediness of the proposed method.