为了降低光伏集群输出力波动性与随机性、保障电力用户需求及电力系统的安全运行,采用历史数据驱动,提出一种基于光伏集群内部场站出力排序的有功出力预测控制方法。该方法首先利用改进的SSA-LSTM求解光伏集群的调度曲线,构建以集群实际出力对调度曲线的调度误差、实际出力对典型日出力的调度余量,建立光伏集群出力排序指标体系,以调度跟踪误差最小为目标,优化出力。其中为了提高调度跟踪精度,考虑场站实际出力与典型气象日曲线的出力符合度、余弦相似度、调度跟随度构建场站调控有效度指标,对集群内部场站进行排序控制,使其自适应选择出力模式,最终优化集群出力。仿真结果验证了所提方法能够合理分配各场站出力,使集群出力更加平滑。
时间: 2024-02-26 09:54:01 浏览: 33
To reduce the fluctuation and randomness of the output power of photovoltaic clusters, and to ensure the power demand of electricity users and the safe operation of the power system, a historical data-driven method based on the internal station output sorting of photovoltaic clusters is proposed for active power prediction and control. Firstly, an improved SSA-LSTM algorithm is used to solve the scheduling curve of the photovoltaic cluster, and an index system for the output sorting of the photovoltaic cluster is established based on the scheduling error between the actual output of the cluster and the scheduling curve, the scheduling margin between the actual output and the typical day output, and the optimization of the output is carried out with the goal of minimizing the scheduling tracking error. In order to improve the accuracy of the scheduling tracking, an effective control index for the station is constructed based on the conformity, cosine similarity, and scheduling follow-up degree between the actual output of the station and the output curve of the typical meteorological day. The station is sorted and controlled internally in the cluster, and the adaptive selection of the output mode is achieved to ultimately optimize the output of the cluster. The simulation results verify that the proposed method can reasonably allocate the output of each station and make the output of the cluster smoother.