基于传输点容量模型的发输电系统可靠性评估

A Method for Composite Power System Reliability Evaluation Based on Delivery Point Capability Model

利用发输电组合系统(简称组合系统)极限输电能力与负荷解耦(系统—负荷解耦)的特性,能有效提高组合系统可靠性的评估效率。目前已有的基于系统—负荷解耦分析的组合系统可靠性评估方法尚存在不足,其中的输电能力计算模型没有考虑异常状态下优化负荷削减的实际需求,也没有计及各不同传输点负荷水平变化的影响,此外,对于频率及持续时间型可靠性指标的计算方法,目前也鲜有研究。该文以系统—负荷解耦分析为出发点,提出基于传输点容量模型的组合系统可靠性评估方法。首先,建立一种适用于组合系统可靠性分析的传输点输电能力优化决策模型,并针对该模型目标函数具有斜率递增的特点,将其转换为线性规划模型进行求解,保证计算效率。然后,拓展传统发电系统可靠性评估中可用容量状态概率及增量频率的概念,基于马尔科夫链构建组合系统各传输点输电能力的概率—频率分布函数(即传输点容量模型),将其与相应负荷的概率—频率分布函数(即负荷模型)进行卷积和加法运算,不仅能得到组合系统及其传输点的概率型可靠性指标,而且还能得到频率及持续时间型等可靠性指标。以RBTS-6节点系统为例进行测试分析,说明该方法的有效性。

The efficiency of the composite power system （CPS） reliability evaluation can be improved effectively by using the decoupling characteristics between CPS transfer capability and corresponding load （system-load decoupling）. There are still some technical defects in the existing methods based on system-load decoupling. Existing transfer capability calculation models consider neither the optimal load shedding under abnormal conditions nor the change of the load level of each delivery point. In addition, there are few researches on the calculation methods of frequency and duration reliability index. By using the system-load decoupling characteristics, a method for CPS reliability evaluation based on delivery point capability model is proposed. Firstly, an optimaldecision-making model is presented for calculating delivery points transfer capability of CPS, which is suitable for reliability evaluation. Given that the objective function of the presented optimal model has successively increasing slopes, the corresponding nonlinear model is transformed into a linear programming model to be solved to enhance the computational efficiency. Then, the probability-frequency distribution function of delivery point transfer capability, also named delivery point capability model, is built based on Markov chain by extending the concept of state probability and incremental frequency already used for modeling available capability of generation. Not only the probability but also the frequency and duration reliability index of delivery points and CPS can be obtained by convolution and addition operation between delivery point capability models and corresponding load models. The analysis for RBTS-6 test system demonstrates the validity of the proposed method.