采用分段离散化和高斯混合模型的多场景概率潮流计算

摘要

For the uncertainty of wind power and load on power system operation, a calculating method of multi-scenario probabilistic power flow based on piecewise wind power and Gaussian mixture model of load power is proposed.The output power of the wind farms is segmented using the directional power characteristic curves, and the wind power multi-scenario is set up.At the same time, the Gaussian mixture models are developed to set up the load power multi-scenario.Then, the multi-scenario of power injection and the corresponding probability are confirmed.In each scenario, the output power of each wind farm is constant, and the load power obeys Gaussian distribution.Finally, on the basis of the total probability theory, the probabilistic power flow results can be got by integrating probability distribution of the state variables and corresponding weight in each scenario.The modified IEEE 57 system is used to test the proposed method.The simulation results show that the probabilistic power flow calculation process is simplified greatly and the calculation efficiency is improved.%针对电网运行中风电和负荷的不确定性,提出了一种采用风电场输出功率分段离散化和负荷高斯混合模型的多场景概率潮流计算方法.在风电场有向功率特性曲线的基础上,分段离散化处理风电场输出功率,构造风电场输出功率多场景.同时,建立负荷的高斯混合模型,构造负荷功率多场景.然后,确定系统注入功率多场景及其对应的概率,在系统注入功率的每个场景中,风电节点输出功率为定值,负荷节点功率均服从高斯分布.最后,应用全概率公式,将系统注入功率的每个场景中计算所得状态变量的概率分布以该场景对应的概率作为权重,整合计算得到最终的概率潮流结果.以改进的IEEE 57节点系统进行仿真分析,结果表明所提方法简化了概率潮流求解过程,提高了计算效率.