Active Power Control Strategies Of PV Station Considering Photovoltaic Power Prediction

With large-scale photovoltaic power integration into grid, the fluctuation of active power output of PV power station will become a big challenge for safe and stable operation of the power system. The influence of large-scale PV power station integration on system power balance is a main area in research. Therefore, detailed specifications are formulated by SGCC, that the grid-connected PV power station should have a certain ability to regulate active power. Formulating an appropriate active power control strategy has a positive significance in improving the schedulability of the PV power station and avoiding the adverse impact on system power balance due to the output fluctuation.According to the low conversion efficiency of the inverter and poor power quality output in traditional active power control strategies, a new strategy is proposed based on objective optimization, which takes the grid scheduling and the PV power prediction into account. The output power is adjusted to achieve the target value that is dispatched by the grid scheduling. When the target value is greater than the total power, use the Maximum Power Point Tracking, otherwise, through the proper allocation strategy to control the output range of each inverter and launching or stopping inverters in turn to tracking the target value.A variable step perturbation and observation MPPT control method is studied. Simulation results indicate that this method can track the maximum power point much better than the traditional methods in speed and stabilities. For the PV power prediction level, proposed PV system power prediction using support vector machines, trained the historical records with LIBSVM package. Further optimized the model parameters, established the prediction model, shows 24 hour prediction power curve of a PV power station in Gansu and compares with predictions of Elman neural network model, verified SVM superiority in the small sample prediction. Based on all the studies above, coordinated the output power of the PV plant, optimized active power allocation algorithm, minimized the number of running inverter, increasing the life span of PV inverter. Matlab/Simulink simulation and analysis shows that the control strategy this paper proposed can achieve the output power tracking the value that is dispatched by the grid scheduling and planning, and more reasonable and advanced than the existing power control strategy.