面向增强态势感知的输电系统安全性综合评估

发布时间：2018-03-06 06:26

本文选题：增强态势感知　切入点：安全性综合评估　出处：《天津大学》2014年博士论文　论文类型：学位论文

【摘要】：态势感知是提高电网自愈(Self-healing)功能和抗扰能力(Robustness)的关键。随着大规模连锁停电事故频繁发生和大规模风电的接入，安全性评估作为态势感知的核心分析工具面临着诸多挑战。本文以安全域方法学的研究成果为基础，围绕面向增强态势感知的输电系统安全性综合评估展开了研究。首先，研究了在注入功率空间上保证含双馈风机(DFIG)电力系统暂态稳定性的动态安全域(以下简称含DFIG的动态安全域)。通过大量暂态稳定仿真发现，在工程关心的范围内，含DFIG的动态安全域的边界仍可用一个或少数几个超平面表达式近似表示，其精度可满足工程实际需要。该研究结果使安全域方法的优越性在风电接入后的安全性监视与分析中进一步凸显，使其可作为面向增强态势感知的安全性监视、评估与控制的强有力的工具。其次，给出了安全转移概率的解析计算方法，证实了安全转移概率的有效性。基于超平面形式的安全域边界的表达式将安全转移概率计算中涉及的复杂体积分极大程度地简化，结合以半不变量为基础的级数展开法加以计算。该解析计算方法保证计算精度的同时，，计算速度快，为在安全性分析中计及风电和负荷等引起的节点注入不确定性提供了有效途径。同时指出所确定的安全转移概率不仅是最终的概率的静态和动态安全性综合评估的马尔可夫模型状态转移概率矩阵元素的重要成分，而且它本身就可作为评估系统安全性的有效指标，辅助运行人员监测系统状态以提前做好预防措施。再次，提出一个快速形成马尔可夫模型状态转移概率矩阵的方法。基于元件状态转移率矩阵和系统状态数组提出快速计算状态转移率的方法，并归纳出系统状态转移概率矩阵中非零元素的分布规律，进而快速实现系统状态转移概率矩阵的稀疏存储。对于两状态元件组成的系统，可基于给定的系统状态排序和服务状态集数组进一步提高计算效率。该方法的计算速度与应用规模均优于现有方法，为运用马尔可夫模型分析电网等大规模系统提供了有利工具。最后，完善了二层系统模型以计及风电的接入，以此为基础通过安全域和马尔可夫模型建立了概率的静态和动态安全性综合评估。模型计及了网络拓扑状态间的转移、系统的可再生能力、风电和负荷的不确定性、静态和动态安全性约束，以及天气和自然灾害等因素的影响，通过求解一组线性向量微分方程获得“到不安全时间”及其概率分布实现了对系统未来状态的有效估计并估算出可用于采取措施的时间。制定了有效的模型求解方案，首次将该模型成功应用于复杂系统，推进了模型的实用化进程。算例结果显示，由于整合了诸多影响电网安全的因素，模型给出了更符合实际的评估结果，可用于评估干扰因素对系统安全性的影响，辅助运行人员进行决策，为增强态势感知提供了有效的预测分析工具。
[Abstract]:Situation awareness (Self-healing) is to improve the self-healing function and anti disturbance ability (Robustness) key. With the large scale blackout occurred frequently and large-scale wind power access, safety assessment as the core of situation awareness analysis tool is facing many challenges. The results of this study to the security region methodology based on oriented enhancement comprehensive evaluation of transmission system security situation awareness is studied.
First of all, on the guarantee containing DFIG in power injection space (DFIG) dynamic security region of power system transient stability (hereinafter referred to as dynamic security domain containing DFIG). Through a large number of transient stability simulation results show that in the range of engineering is concerned, the dynamic security region of DFIG containing one or the boundary is still available a few hyperplanes approximate expression that the accuracy can meet the actual needs of the project. The results of this study. The advantages of the security region method in wind power safety monitoring and analysis of access after further highlighted, which can be used as for enhanced security surveillance situation perception, a powerful tool to assess and control.
Secondly, given the analytical method for calculating security transition probability, confirmed the validity of the safe transfer probability. The expression boundary hyperplane forms based on the safe transfer of simplified probability calculation of complex volume relates to the great extent, combined with the series with a semi invariant based expansion method to calculate the analytical calculation method. To ensure the accuracy and calculation speed, which provides an effective way for nodes with wind power and load analysis in safety into uncertainty. The safe transfer probability is not only an important component of the final state Markov model the probability of static and dynamic security assessment of transtion-probablity matrix elements at the same time points out, and it itself can be used as an effective index to evaluate the security of the system, the auxiliary operation personnel monitoring system to advance the prevention measures. Shi.
Again, put forward a method of rapid formation of Markov model state transtion-probablity matrix element. The state transition rate matrix and system state array fast calculation method proposed based on state transfer rate, and summed up the system state transition probability matrix of nonzero elements in the distribution rule, and then quickly realize the sparse storage system state transtion-probablity matrix. For the two state system components, system state sort and service status based on the given set of array to further improve computational efficiency. The calculation speed and scale of application the method is superior to the existing methods, providing a favorable tool for using the Markov model analysis of power grid and other large-scale systems.
Finally, to improve the two layer system model to account and access of wind power, on the basis of the security domain and Markov model to establish the probability of static and dynamic security assessment model. And the transfer network topology between the system of renewable capacity, wind power and load uncertainty, static and dynamic security constraints, influence factors and weather and natural disasters, "to secure time" and its probability distribution can effectively estimate the future status of the system and estimate can be used to take measures in time by solving a set of linear vector differential equations has been developed. The effective solving scheme of the model for the first time, this model is successful applied to the complex system, promote the practicability of the model. The simulation results show that the influence factors of integrated network security model, gives a more realistic assessment. It can be used to assess the impact of interference factors on the security of the system, assist the operators to make decisions, and provide an effective prediction and analysis tool for the enhancement of situational awareness.

【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM72

【参考文献】