一种跟踪风光互补发电系统计划出力的变时间尺度短期功率预测方法

摘要:

引入各发电组件的短期功率预测值有助于风光互补发电系统制定优化的能量分配策略，更好地跟踪计划出 力。针对风光互补发电系统能量分配周期多变的特点，提出了一种变时间尺度的短期功率预测算法。该算法将历史 数据与物理模型相结合，可获得各发电组件任何时间间隔的短期功率预测值，进而实现对于计划出力的最大程度跟 踪。基于某风光互补发电站数据的仿真结果表明，基于历史数据与物理模型相结合的预测方法可更准确地实现能量 分配变周期下的计划跟踪。

Abstract:

In wind － solar hybrid generation system，it is meaningful to introduce the predicted short － term power of each component to develop the optimal energy allocation strategy and track the generation scheduling in a better way． Considering the variable allocation cycle，a short － term power prediction method based on variable time scale is proposed． In the algorithm， the short － term power of different power components in any time scale could be predicted based on the histrionic data and physical model，and the tracking of generation scheduling in optimal mode is implemented． The simulation results based on the data of a wind － solar hybrid power supply system indicate that，the proposed method could be applied to hybrid generation system with variable allocation cycle，and tracks the generation scheduling in a better way，also the combined method outperforms the pure physical model in reducing the tracking error，and improves the stability of the system output．