Fluctuation Reduction of Wind Power with Battery Energy Storage Systems

Abstract

Due to the characteristics of intermittency and uncertainty of wind power, the drastic wind power fluctuations have negative impacts on the safety and stability of power systems. In this paper, we focus on a wind farm with a battery energy storage system (BESS). The objective is to reduce the fluctuation of total output power by dynamically scheduling the charging and discharging power of BESS. The randomness of wind power is characterized by a periodic Markov chain from a long-term viewpoint. The variance of the power output is chosen as the indicator of long-term fluctuations. However, since the variance function is quadratic and non-additive, this scheduling problem does not fit the model of Markov decision processes (MDPs). We use the sensitivity-based optimization method to derive a difference formula to overcome this difficulty, which is the basis of the proposed iterative optimization algorithm. The optimal policy derived by the algorithm can determine the (dis)charging power of the BESS at each situation of the wind power, the battery energy level, and the hour of days. The performance of the proposed approach is verified through numerical experiments using real data from National Renewable Energy Laboratory (NREL). Simulation results demonstrate the efficiency of the proposed approach,