【作者】 卓亮；

【导师】 何星；

【作者基本信息】 上海交通大学 ， 控制工程， 2012， 硕士

【摘要】 现代电力调度系统要求能够迅速、准确而全面地掌握电力系统的实时运行状态,做出及时有效的控制决策,从而保证运行的安全性和经济性。电力系统状态估计利用实时量测和伪量测数据估计出全网的母线电压幅值和角度,排除随机干扰引起的错误信息,从而估计和预报系统的实时运行状态,实现对系统当前运行状态的安全监控,在现代电力系统中发挥着重要的作用。传统的静态状态估计只能进行当前状态的估计,动态状态估计则可以提前估计下一时刻的状态。90年代以来,基于GPS技术的PMU装置的出现,为状态估计提供了新的发展契机。本文针对基于PMU的电力系统动态状态估计问题,主要做了以下几方面工作:(1)介绍了状态估计的研究意义、概念及研究现状。包括传统的静态状态估计算法、动态状态估计算法以及近几年计及PMU量测的状态估计算法的发展情况。重点介绍了动态状态估计的基本方法——扩展Kalman滤波法及相关改进算法,并指出算法各自的优缺点。(2)详细介绍了PMU的技术原理,构成WAMS系统的方式及引入状态估计时的相关问题。针对当前PMU量测和SCADA系统长期共存的情况,重点介绍了基于二者混合量测的动态状态估计算法。(3)提出一种基于混合量测的静动态结合的估计算法。该算法针对SCADA量测数据运用基本加权最小二乘法进行静态估计,动态估计时只选用PMU量测的节点电压相量与静态估计结果进行线性估计。传统的静态估计方法实用可靠,估计结果精度高有保障,动态估计时运用线性估计方法,速度快,保证了估计效率。IEEE-14节点系统仿真验证了算法的有效性和优越性。针对PMU的配置个数和位置对估计精度的影响做了仿真分析,结论表明:配置有PMU节点的估计精度偏高;临近配置甚至构成环路的配置方式可以提升整体估计精度;配置在邻接支路较多的节点,可以有效提升整体估计精度。(4)为了进一步提升估计时间和精度,提出基于快速分解的混合估计算法。该方法运用快速分解状态估计算法做静态估计以节省估计时间,动态估计时将PMU支路电流相量量测转换为节点电压相量量测参与估计,以达到增加量测冗余、提高估计精度的目的。IEEE-14节点系统仿真结果表明,该改进算法有效改善了滤波和预测的精度。同时给出了配置不同数目PMU时滤波误差的数值试验结果。更多还原

【Abstract】 Modern power system requests to monitor the real-time state to make effective control decisions, in order to keep its safety. Power system state estimation can calculate the bus voltage amplitude and phase from system measurements, so as to estimate and forecast the state, which plays an important role in modern power system. The traditional static state estimation can only give the current state, while dynamic state estimation can provide the state of next step. Since the 1990s, the emergence of PMU based on the GPS technology provides a new opportunity and method for state estimation.Aimed to the dynamic state estimation of power system based on PMU, the paper did some research work mainly on the below aspects:(1)The concept of state estimation, why to study the state estimation and how the research is going on were firstly introduced. The traditional static state estimation algorithms, dynamic state estimation algorithms and algorithms considering the PMU measurements were covered. The basic dynamic state estimation algorithm and some improved algorithms were given an emphasis and their merits and defects were pointed out after careful analysis and comparison.(2)PMU provides a new way for state estimation. Its technical theory, the way to form WAMS system and some relevant problems were detailed in the paper. Because the SCADA system and PMU measurements will exist together for a long time, some algorithms based on data fusion of these two measurements were introduced in detail.(3)Based on the study above, a new algorithm which combines the merits of the static and dynamic state estimation algorithm was brought up. The algorithm uses WLS algorithm on the SCADA measurements first, then mix the result with node voltage from PMU measurements as mixed measurements data to do further dynamic estimation. The traditional WLS algorithm provides reliable results and the linear dynamic algorithm has a quick performance. Simulations on an IEEE 14 test system showed that the new algorithm is effective and superior. Simulations and analysis were carried out to study the effects of PMU deployment to the estimation accuracy, results show that those lines with PMU have higher accuracy; near deployment improves all lines’accuracy; deployment on the node with more branches improves all lines’accuracy.(4)An improved algorithm was brought up to reduce estimation time and error. The fast decoupled method was applied in the process of static estimation. Then the results, together with the voltage phasor measured by PMU, and some relevant voltage phasor transformed from branch current measured by PMU, formed the measured data for the linear dynamic state estimation. Simulations on an IEEE 14 bus test system demonstrated that the method has a better validity in filtering and forecasting the voltage phasor. Numerical experiments results were given to show the effects of different amount of PMU to estimation accuracy.更多还原