基于网络拓扑数据的低频振荡扰动源定位研究

发布时间：2018-05-09 10:18

本文选题：低频振荡 + 扰动源定位　；参考：《华北电力大学》2017年硕士论文

【摘要】：随着电力工业的迅速发展,互联电网的规模越来越大,电网动态稳定性问题也日益复杂,低频振荡现象时有发生,既威胁着电网的安全稳定运行,也制约了电网的输电能力。及时准确定位低频振荡扰动源对于抑制振荡扩散、防止事故蔓延具有重要意义,为此,本文在前人研究的基础上,将低频振荡发生时的能量分布特性与电网拓扑结构相结合,提出了基于网络拓扑数据的扰动源定位新思路。首先以简单系统模型为例,对目前成熟的负阻尼机理及强迫振荡机理进行推导;然后以振荡能量为扰动源定位的依据,通过分析振荡发生时的能量分布特性,确定可以将图论应用于研究扰动源定位问题;随后在割集能量的基础上,提出了最大振荡能量流割集的概念,将扰动源定位问题转化为最大振荡能量流割集的搜索问题。由于电网规模较大,普通的割集搜索方法难以快速地搜索出最大振荡能量流割集,故本文提出从以下两方面提高搜索速度:1、缩小搜索空间;2、提高搜索的有效性。为了缩小搜索空间,本文对振荡发生时的能量分布特性进行深入分析,发现了最大振荡能量流割集的分布规律,提出了基于前K最短路径的扰动源定位方法,将最大振荡能量流割集的搜索限定在前K最短路径的搜索结果中,可以大大缩小搜索空间、提高搜索速度。本文进一步提出了K值选取的校正经验法,使基于前K最短路径的扰动源定位方法更具有工程实用性。为了提高搜索有效性,本文结合大规模网络割集搜索问题的特点,对网络中的支路进行01编码,利用遗传算法引导最大振荡能量流割集的搜索,并根据扰动源定位问题的具体情况,提出了通过预选染色体来加速遗传进化的措施,形成了基于遗传算法的扰动源定位方法。以上两种扰动源定位方法均同时适用于局部及区间振荡模式,并且能够在区域电网机组实时信息不全的情况下,只利用线路信息实现扰动源的定位。最后,蒙西电网的仿真算例验证了以上方法的有效性和实用性。
[Abstract]:With the rapid development of power industry, the scale of interconnected power network is becoming larger and larger, the dynamic stability of power network is becoming more and more complex, the phenomenon of low-frequency oscillation occurs frequently, which not only threatens the safe and stable operation of power grid, but also restricts the transmission ability of power grid. It is very important to locate the disturbance source of low-frequency oscillation in time and accurately to restrain the spread of oscillation and prevent the spread of accident. Therefore, based on the previous research, the energy distribution characteristics of low-frequency oscillation are combined with the topology of power network. A new idea of disturbance source location based on network topology data is proposed. Firstly, taking the simple system model as an example, the current mature negative damping mechanism and forced oscillation mechanism are deduced, and then the oscillation energy is used as the basis of the disturbance source location, and the energy distribution characteristics when the oscillation occurs are analyzed. It is determined that the graph theory can be applied to study the problem of disturbance source location, and then the concept of maximum oscillatory energy flow cut set is proposed based on the cut set energy, and the problem of disturbance source location is transformed into the search problem of maximum oscillatory energy flow cut set. Because of the large scale of the power network, it is difficult for the ordinary cut set search method to search the maximum oscillatory energy flow cut-set quickly. Therefore, this paper proposes to improve the search speed from the following two aspects: 1, reduce the search space and improve the efficiency of the search. In order to reduce the search space, this paper analyzes the energy distribution characteristics of oscillations, finds out the distribution law of the cut set of maximum oscillatory energy flow, and proposes a method of source location based on the shortest path of previous K. By limiting the search of the maximum oscillatory energy flow cut set to the search result of the first K shortest path, the search space can be greatly reduced and the search speed can be improved. In this paper, a correction empirical method for the selection of K value is proposed, which makes the disturbance source location method based on the first K shortest path more practical in engineering. In order to improve the efficiency of the search, this paper combines the characteristics of the large-scale network cut set search problem, carries out 01 coding of the branch of the network, and uses genetic algorithm to guide the search of the maximum oscillatory energy flow cut set. According to the specific situation of the problem of perturbed source location, the method of accelerating genetic evolution by pre-selecting chromosomes is put forward, and the method of perturbed source location based on genetic algorithm is formed. The above two methods are suitable for both local and interval oscillation modes, and can only use the line information to realize the location of the disturbance source when the real-time information of the units in the regional power network is not complete. Finally, the effectiveness and practicability of the above methods are verified by a simulation example of Mengxi Power Grid.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM712

【参考文献】