基于信号激励法的小扰动安全在线评估

基于信号激励法的小扰动安全在线评估 On-line small disturbance security assessment based on signal excitation Dissertation Submitted toTsinghua Universityin partial fulfillment of the requirement for the degree of Doctor of Engineering 摘 要 随着我国电力系统的互联和电力市场的发展，电力系统的小扰动安全事件发生的风险越来越大，给系统的安全运行带来了很大的隐患。对系统小扰动安全状态进行正确的评估至关重要。由于目前的各种小扰动评估方法主要是基于系统模型和参数，而准确的模型和参数难以获得，因此得到的分析结果与系统的实际安全状态有很大的差异。本文提出一种基于信号激励的小扰动安全在线评估方法：通过周期性的向系统注入特定的小扰动功率信号去激励系统的响应，并监测和分析系统对于扰动的响应信号，获得系统振荡模式的阻尼和频率信息，从而实现对系统的小扰动安全状态的评估，避免模型和参数的不准确对于评估结果的影响。 论文分几个部分对该在线评估方法进行了研究：首先设计了扰动功率信号，该信号与系统低频振荡的频率范围一致，能对该频段内的各频率点施加同等强度的激励，容易在系统中传输，所激励的响应随着系统的工作状态的变化而变化；之后，推导了以扰动信号为输入信号的系统状态空间表达式，说明了扰动信号激励所产生的频域响应与系统状态及振荡模式的对应关系；利用特征值对于负荷变化的灵敏度，对系统进行子系统的划分，确定了薄弱区域和扰动源的安装地点；通过建立各支路对应的可观性指标集合，确定了响应监测的合适位置；基于改进的Levy曲线拟合方法和3阶7变量的局部传递函数拟合模型，通过对系统频域响应曲线的拟合，获得了曲线局部极大值点对应振荡模式的阻尼和频率特性，形成了小扰动安全的在线评估算法，并通过EPRI-36节点系统和川渝电网的仿真说明了该评估算法的准确性和有效性；为了验证该评估方法的可行性，根据SMES具有灵活快速有功和无功调节能力以及小能量大输出功率的特点，选择SMES作为扰动功率源，利用一台电流源型SMES实验装置，通过动模实验说明，该方法的评估结果能反映系统小扰动安全状态的变化趋势。 关键词：小扰动安全，低频振荡，在线评估，SMES Abstract With the interconnection of Chinese power system and the development of power market, the risk of small disturbance instability becomes bigger and bigger, which threatens the power system secure operation. The accurate assessment of small disturbance security is quite important to the power system. However, most methods of disturbance security assessment are based on models and parameters, which are difficult to obtain accurately, so results from them may not accord with the real power systems. In this dissertation a method of on-line small disturbance security assessment based on signal excitation was proposed: by injecting periodic disturbance power signals with special waveform generated by a disturbance source into the power systems to excite low-frequency oscillation modes under this operating condition, and measuring and analyzing frequency response curves, the damping and frequency of modes could be obtained and the margin of the small disturbance security currently could be assessed. This method avoided the influences of inaccuracy of models and parameters. Systematical research on the method was done. Firstly, the disturbance power signal was designed, which had the same frequency band with the power system low frequency oscillation and could exert the same excitation on every frequency point. The disturbance power signal was easy to transmit in power systems and the responses excited by it varied with the operating points of power systems. The state-space representation of power systems with the disturbance signal was deduced. Based on it, the relationships of frequency response excited by disturbance signal with operating states and oscillation modes were explained. Partitioning the power system into subsystems by sensitivities of eigenvalues to loads, the weak small signal security area of power systems and the location of disturbance source could be determined. Through the observation index set of all branches, the suitable monitoring site could be chosen. Based on modified Levy curve fitting method, a local third-order transfer function fitting model was built, and then the damping and the frequency of modes were obtained by fitting the curves near the maximum values of frequency responses. Monitoring the variations of the damping and the frequency of modes, dynamic small signal security of power systems could be assessed. The simulation results of China-EPRI 36 buses and Chuan Yu Power System showed that the method was feasible and effective. Because SMES was able to control active and reactive power flexibly and quickly and output a large amount of power with a little energy, SMES was chosen as the disturbance source. To demonstrate the feasibility of this method, experiments were carried out with a grid model SMES system. The experimental results showed that the assessment results under the disturbance could reflect the trend of the small disturbance security of the power system. Key words: sm