基于双层同构贝叶斯网络模型的配电网可靠性评估

提出了一种双层同构贝叶斯网络模型进行配电网的可靠性评估和再评估.在进行馈线分区的基础上,结合区域故障模式影响分析(Failure-Mode-and-Effect Analysis,FMEA)构成贝叶斯网络.通过贝叶斯网络的正向推理评估负荷点及配电网的可靠性,求取可靠性指标.通过网络的反向推理进行假设分析,识别网络中的薄弱环节,采用双层结构分别计算停运概率指标和停运频率指标.该方法可以处理带子馈线的复杂配电网可靠性评估问题,能够考虑自动装置不可靠动作、备用电源、线路计划检修等情况.算例分析结果表明了该方法的有效性.

A NEW BAYESIAN NETWORK MODEL FOR DISTRIBUTION SYSTEM RELIABILITY EVALUATION BASED ON DUAL ISOMORPHIC BAYESIAN NETWORK MODEL

The authors proposed a method for reliability evaluation and re-evaluation of distribution network by dual isomorphic Bayesian network model. On the basis of feeder partitioning and combining with failure-mode-and-effect analysis (FEMA) a Bayesian network was constituted. By means of forward reasoning of Bayesian network the reliability of load points and distribution network were evaluated and the reliability indices could be obtained. By means of backward reasoning of Bayesian network the what-if study was carried out and the weak points in distribution network were identified. The outage probability indices as well as outage frequency indices could be respectively calculated by dual model. The proposed method could be used to evaluate the reliability of complicated distribution network with sub-feeders while the situations such as unreliability of automatic isolation devices, alternative supply and feeder scheduling maintenance could be considered. The results of applying the proposed method to test system showed that this method was effective.