Advanced Reactive Power Compensation of Wind Power Plant Using PMU Data

This paper introduces a new model to improve the wind power plant performance by modeling its reactive demand. It develops probabilistic based on prediction interval help better of demands unit which needs be compensated static VAr compensator (SVC). is made possible use non-parametric neural network (NN) using lower and upper bound estimation (LUBE) method. To avoid instability arising due nonlinear complex nature NN, idea combined intervals used here. Due highly non-stationary characteristics pattern consumed in plant, optimization algorithm $\theta $ -symbiotic organisms search ( -SOS) proposed train LUBE parameters polar coordinates. In addition, two-phase modification method developed enhance local ability SOS premature convergence issue. The experimental Phasor Measurement Unit (PMU) data shows that can SVC, effectively.