基于脆弱点的变电站风险评估方法研究与应用

发布时间：2018-06-26 08:21

本文选题：变电站自动化系统 + 复杂网络　；参考：《东北电力大学》2017年硕士论文

【摘要】：变电站是电力系统的中枢,随着其智能化程度的提高,变电站自动化系统间的数据交互也越发复杂。若变电站自动化系统中的节点因外界扰动而产生风险后,这些风险可能会对其相邻或相关联的节点造成一定程度的影响,进一步扩大风险范围,使得那些原本不具有风险的节点,最终可能会因风险的传播扩散而造成脆弱性改变,这些节点失效后极易引发连锁故障。由于变电站自动化系统可以抽象为由逻辑节点和逻辑连接组成的网络,因此应用复杂网络理论研究变电站的风险传播扩散现象和连锁故障演化机理具有一定的可行性。针对以往风险传播扩散研究中忽略风险逐层累积的问题,提出了一种考虑风险累积效应的风险传播扩散模型来描述节点的风险状况。该模型通过分析每条风险传播路径的概率,确定风险扩散至其相邻节点的先后顺序和传递的风险大小,从而判断初始节点受扰动后风险的传播扩散途径,同时,统计网络中各节点的累积风险,预测最具安全威胁的脆弱点。然后,针对连锁故障分析中不能根据网络演化调整节点负荷容许参数的问题,建立了一种适用于变电站自动化系统连锁故障分析的模型。该模型引入接近度概念来衡量节点的负荷承载量,用于动态调整节点负荷容许参数,同时,针对逻辑节点在站控层、间隔层和过程层的分布特点,改进网络平均有效性能的衡量方法,以更为准确地评估连锁故障的破坏程度。实验以T1-1小型变电站为例,依据IEC 61850标准中逻辑节点间的关联关系,构建成复杂网络,然后分析不同风险源,确定相应的脆弱点,对比不同脆弱点失效后引发连锁故障的情况。实验结果表明,本文提出的模型对变电站风险评估具有一定的有效性和适用性。最后,在变电站三维可视化场景的基础上,开发了动态风险评估仿真平台。平台集成了风险传播扩散算法和锁故障分析算法,以满足管理人员对运行数据直观展现和分析预测的需求,减少管理成本。
[Abstract]:Substation is the center of power system. With the improvement of its intelligence, the data exchange between substation automation system becomes more and more complicated. If the nodes in the substation automation system are exposed to external disturbances, these risks may have a certain degree of impact on the adjacent or associated nodes, and further expand the risk range. Those nodes that had no risk may eventually change their vulnerability due to the spread of risk, which can easily lead to cascading failures after the failure of these nodes. Because substation automation system can be abstracted as a network composed of logical nodes and logical connections, it is feasible to study the phenomenon of risk propagation and diffusion and the evolution mechanism of cascading faults by using complex network theory. Aiming at the problem that risk accumulation is neglected in previous risk propagation and diffusion studies, a risk propagation and diffusion model considering risk accumulation effect is proposed to describe the risk situation of nodes. By analyzing the probability of each risk propagation path, the model determines the sequence of risk diffusion to its adjacent nodes and the magnitude of the risk transmission, so as to judge the propagation and diffusion path of the risk after the initial node is disturbed, at the same time, The cumulative risk of each node in the network is counted to predict the most vulnerable point of security threat. Then, aiming at the problem that the allowable parameters of node load can not be adjusted according to the network evolution in the cascading fault analysis, a model suitable for the cascading fault analysis of substation automation system is established. In this model, the concept of proximity is introduced to measure the load carrying capacity of nodes, which is used to dynamically adjust the load allowable parameters of nodes. At the same time, the distribution characteristics of logical nodes in the station control layer, interval layer and process layer are analyzed. The method of measuring the average effective performance of the network is improved to evaluate the failure degree of cascading faults more accurately. Taking T1-1 small substation as an example, a complex network is constructed according to the relationship between logical nodes in IEC 61850 standard, and then different risk sources are analyzed, the corresponding vulnerability points are determined, and the cascading faults caused by the failure of different fragile points are compared. The experimental results show that the proposed model is effective and applicable to substation risk assessment. Finally, a dynamic risk assessment simulation platform is developed based on the three-dimensional visualization scene of substation. The platform integrates the risk propagation diffusion algorithm and the lock fault analysis algorithm to meet the needs of managers to display and analyze the operation data intuitively and reduce the management cost.
【学位授予单位】：东北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM63

【参考文献】