【作者】 王东涛；

【导师】 余贻鑫；

【作者基本信息】 天津大学 ， 电力系统及其自动化， 2007， 博士

【摘要】 市场环境下的电力系统中出现的各种随机因素向传统的确定性安全监视与评估提出了挑战,因此急需发展概率安全和风险评估。本文应用安全域的最新成果,进行了输电系统概率动态和静态安全性评估以及风险安全性评估的研究。本文首先提出了一种基于实用动态安全域的离散和连续随机变量相结合的输电系统概率动态安全评估模型。模型中不仅计及了节点注入功率的不确定性、预想事故的不确定性、事故发生地点和故障清除时间的不确定性,还计及了稳定分析中负荷模型的不确定性。动态安全域的超平面数学描述使注入功率不确定性和电动机负荷比例不确定性的计及变得十分简便,而且使动态不安全概率的计算负担大大降低了。此外,采用截断正态分布作为节点注入功率随机变量的概率模型,并用以半不变量为基础的Gram-Charlier级数计算了随机变量的联合概率分布,使概率分布的考虑更加符合实际。其次,本文建立了基于随机潮流和割集空间静态电压稳定域的输电系统静态电压稳定安全分析模型。该模型采用基于牛顿—拉夫逊法的随机潮流确定割集静态电压稳定域中的线路潮流概率分布,利用半不变量法和Gram-Charlier级数,得到了割集静态电压稳定域中线路潮流的线性组合随机变量的概率分布函数,并根据割集电压稳定域的超平面方程计算系统的静态电压不稳定概率。本文提出了一种基于实用动态安全域的动态不安全风险评估模型。在动态不安全风险计算中计及了系统安全控制措施对风险的影响,而实用动态安全域的应用可以比较方便地计及暂态稳定约束。以系统动态不安全风险最小为目标,给出了一种风险控制的最优化模型。风险优化可分为两层进行,外层为对预想控制集合的优化,内层为针对一个事故子集进行的综合安全控制成本的优化。根据所提出的概率安全评估理论,本人所在课题组已针对我国实际电力系统开发了一个输电系统在线概率安全评估系统。该系统能够提供短期内动态不安全概率和静态电压不稳定概率的变化趋势。通过对概率指标的灵敏度分析,可以给出对系统不安全概率贡献大的故障和对系统静态电压不稳定概率影响大的节点等信息。该系统可用于指导运行人员根据系统运行状况决定是否需要采取预防措施以及制定优化的预防控制措施和紧急准备预案,使系统保持在较高的安全运行水平。更多还原

【Abstract】 Since uncertainties in a deregulated power system have put forward great challenges to traditional deterministic security assessment, security region based probabilistic dynamic and static security and risk assessment approaches are developed in this dissertation.Firstly, a practical dynamic security region (PDSR) based transmission system probabilistic dynamic security assessment model is developed combining discrete and continuous stochastic variables in this dissertation. In this model, not only the uncertainties in loads and generation power injections, probabilities of critical contingencies, uncertainties of fault occurrence locations, uncertainties of the fault-clearing time but also uncertainties of load model in stability analysis are considered. The hyperplane mathematic descriptions of PDSR make it easy to take account of the uncertainties of power injections and the uncertainty of motor loads proportion in total load, and can outstandingly reduce computation burden of dynamic insecurity probability. Furthermore, supposing that power injections follow truncated normal distribution, a method of Gram-Charlier progression based on cumulants is adopted to calculate the joint probabilistic distribution of truncated normal distribution, which is more up to the reality.A probabilistic steady voltage stability security assessment model based on stochastic load flow (SLF) and static voltage stability region in cut-set space (CVSR) is presented. The probabilistic distribution functions (PDFs) are obtained by SLF based on Newton-Raphson linear method. Then, the PDFs of linear combination of line flows in CVSR are deduced by Gram-Charlier progression and Cumulants. The probability of static voltage instability can be calculated by hyperplane equation of CVSR.This dissertation presents a dynamic insecurity risk assessment model based on practical dynamic security region. The influence of security control methods applied to transmission system is considered in the calculation of dynamic insecurity risk, and the transient stability constraints can be considered easily by an inequality in form of hyperplane. A risk control optimization model is given taking total risk cost as the objective function. The optimization model can be divided into two layers. The outer layer is contingency set optimization while the inner is an optimization problem of comprehensive security control cost to a given subset of contingency set. Finally, based on the theory of probabilistic security assessment stated above, an online probabilistic security assessment system for transmission systems is developed by our research group for a real power system in our country. The system provides the insecurity probability in the future up to several hours. By sensitivity analysis, the information, such as faults making the most contribution to the system insecurity probability and nodes influencing static voltage insecurity probability outstandingly, can be obtained. The system can be used to guide operators to decide whether it needs to take preventive control actions and the optimal comprehensive control actions that can be taken to keep system security degree in a relatively high level according to the state of power system.更多还原