Energy storage scheduling design on friendly grid wind power

Abstract

Smooth and accurate schedule forecast of wind power may ensure healthy and reliable running of the power grid. Energy storage systems can participate in wind power scheduling due to the ability of space-time translation of electrical energy. The reasonable strategy design is important especially on smooth and accurate forecast of wind power. Firstly, based on control indexes of volatility and forecast error of wind power, Fourier mathematical analysis is used to design one-order low pass filter to smooth the curve of auto-regression ultra-short-term prediction of wind power. Secondly, the predicting curve after smooth is used as wind power actual control target, actual output is processed to track the power forecast by storage control. Thirdly, combined with the actual state of charge (SOC) of the lithium battery, forecast error coefficient adjustment is innovatively introduced as dynamic error band constraint factor. Finally, schemes of complete tracking and relaxed model control are analyzed respectively by simulation to ensure the results of smoothing and accurate forecast effects of the actual wind/storage output. The designed scheme of relaxation factor control can avoid overcharge and over discharge of energy storage, reducing the frequency of charge-discharge of lithium battery, achieving the feasibility and effectiveness of energy management strategy.