【作者】 陈晓刚；

【导师】 曹一家；

【作者基本信息】 浙江大学 ， 电力系统及其自动化， 2008， 博士

【摘要】 随着全国1000kV特高压联网战略的实施,一个装机容量超过8亿kW、横跨数千公里的广域电力系统正在我国形成。尽管多大区互联电网的形成将给全国电力系统带来巨大效益,但大电网的安全性、稳定性和经济性也将面临着前所未有的巨大挑战。目前大规模停电事故的机理研究表明互联电网在大规模停电事故中表现出来的脆弱性与电网负荷水平及拓扑特性有密切的联系,在缺乏重大电网投资的前提下很难单独依靠加强电网建设解决该问题,因此构建可靠有效的广域安全监测系统具有非常重要的现实意义和工程价值。而现有电网监测系统在鲁棒性方面存在的缺陷却在历史上多次大停电事故的发生和扩大过程中起到了推波助澜的作用。在这样的工程背景下,本文对广域安全监测系统所存在的一些鲁棒性问题进行了研究并取得了一些创新性的研究成果。首先,提出了监测系统鲁棒性的定义,并通过对历史上多次大停电事故的分析指出监测系统的鲁棒性对电力系统安全稳定运行有十分重要的影响;介绍了目前国内电网公司普遍采用的SCADA/EMS系统的特点及其局限性,指出同步相量测量技术及基于该技术的广域测量系统是未来电网监测系统的发展方向;构建了未来若干年内浙江省电力公司可能采用的基于WAMS/SCADA混合测量的广域安全监测系统,并对其在鲁棒性方面可能存在的一些问题进行了分析。其次,针对现有电网参数辨识与估计方法所存在的数值稳定性差、易发散及易受残差污染干扰等问题,提出了一种基于WAMS/SCADA混合测量的电网参数辨识与估计方法。该方法首先利用广域测量系统的测量数据计算相对残差,从而对是否存在参数错误进行初步判断,然后利用相量测量单元能够测量电压和电流相量的特性建立支路两端变量之间的直接联系,对存在参数错误问题的支路进行辨识并在此基础上使用智能优化算法估计支路参数。该PMU配置方法只需安装约1/2总母线数的PMU即可对全网传输线路和变压器支路进行参数辨识与估计。第三,针对连锁故障会导致广域测量系统丧失对电力系统完全可观测能力的问题,提出了一种考虑高风险连锁故障的最优PMU配置方法。该方法首先使用隐性故障模型和风险理论对电力系统的连锁故障进行模拟仿真和统计分析,以辨识系统中的高风险连锁故障,然后运用改进模拟退火算法寻找在发生单一高风险连锁故障的情况下仍然能够保证WAMS对电网完全可观的最优PMU配置方案。仿真实验表明该方法能够在经济性和鲁棒性之间取得一个较好的平衡。第四,针对现有的PMU优化配置方法不能同时考虑多种约束条件的问题,提出了一种满足多种约束条件的最优PMU配置方法。该方法能在同时考虑已安装的PMU及对存在电压稳定、功角稳定、小干扰稳定和电磁环网等多种问题的重要环节进行实时监视的情况下,以最少数量的PMU实现对整个电网的完全可观测。最后给出了该方法在浙江省电力公司WAMS系统规划中的应用实例。第五,为了提高WAMS系统定位故障的能力,提出了一种利用电压相量进行故障定位的方法。该方法首先运用对称分量法和线性叠加原理建立故障后的附加正序网络并定义了故障点的匹配指标,进而基于该指标运用遍历搜索方法寻找故障点的位置。该方法仅利用电压相量进行计算因而能够避免由电流互感器饱和所带来的误差影响。基于IEEE 9节点测试系统的仿真实验表明该方法能够有效地定位故障,并且不受故障类型、过渡电阻等因素的影响。第六,针对网络拓扑改变及PMU故障可能造成WAMS对部分母线失去可观测能力的问题,提出了一种基于WAMS/SCADA混合测量的电压相量动态跟踪算法。该算法通过使用SCADA系统的历史数据对电网负荷建立静态负荷模型。一旦出现部分母线失去可观测性的情况,就结合PMU对相邻可观测母线电压相量的高精度测量使用高斯-塞德尔算法对非线性方程(组)进行求解,从而实现对不可观测区域母线电压相量的估计。最后,基于本文二至六章所提出的多种提高广域安全监测系统鲁棒性的措施,在未来若干年内浙江省电力公司可能会采用的WAMS/SCADA/EMS系统基础上提出了一种高鲁棒性广域安全监测系统架构。该监测系统架构综合WAMS、SCADA及FMS(故障信息系统)的优点,运用混合测量技术克服传统单一测量技术存在的局限性。并且对满足高鲁棒性广域安全监测系统需求的最优PMU配置方法进行了研究。更多还原

【Abstract】 With the implementation of 1000 kV EHV AC strategy, a wide area power system covering thousands of kilometers with installed capacity of 8×10~5 MW will be formed soon in China. Although the construction of national interconnected power grid benefits a great deal, it challenges the security, stability and economy of this power grid unprecedentedly. The present studies of large-scale blackout mechanism indicate that the vulnerability of power grids in blackouts has tight relationship with the load level and the network topology. As a result, there is a great practical significance and engineering value in the construction of reliable and effective wide area security monitoring system. Deficiency in robustness of the security monitoring systems played an important role in triggering and spreading the blackouts in history. Under such engineering background, this thesis conducts research on the robustness of wide area security monitoring system and obtains some innovative research results.Firstly, the monitoring system robustness is defined and the crucial impact of monitoring system robustness on power grid security is analyzed with the case study on multiple blackouts in the past. The characteristics and limitations of SCADA/EMS are presented and the WAMS is considered to be the future development highlight of power grid monitoring system. Problems in robustness which exist in the monitoring system based on the hybrid WAMS/SCADA measurement in the future are also analyzed.Secondly, a novel method is proposed to overcome the problems of numerical instability and divergence encountered in the traditional parameter identification methods. At first, the relative residual is employed to judge the existence of parameter error. Then the voltage and current phasors gathered by PMUs are used to establish the direct relationship among the variables on both ends of the branch. By this direct relationship, those branches with parameter errors can be identified and their parameters can be estimated by optimization method. Under this condition, all the parameters of transmission lines and transformers in the power grid can be estimated when only half of the buses are equipped with PMUs.Thirdly, an algorithm of PMU placement considering high risk cascading failures is proposed. Owing to the cascading failure, lines trip one by one, which makes the WAMS ultimately lose its observability. This method employs Risk Theory and Hidden Failure model to identify the high risk cascading failures and then optimally place the PMUs with modified simulating annealing method so as to ensure the full observability of WAMS when certain high risk cascading failure happens. The comparison simulations are carried out on IEEE 39 test system and the results justify that this new method can keep the balance between economy and robustness very well.Fourthly, an algorithm of optimal PMU placement considering multiple constraints is proposed. The minimal PMU placement set can be found when not only the installed PMUs are taken into consideration, but also the key infrastructures with certain problems, such as voltage stability, power angle stability, small signal stability and electromagnetic loop network are directly monitored. The practical application of this new method in Zhejiang power grid is also presented in this chapter.Fifthly, a novel fault location algorithm for transmission lines using synchronized voltage phasor is presented. On the basis of symmetrical component method and linear superposition principle, the additional positive sequence network is established and the matching index on fault point is proposed. Then with this index the fault can be located by traversing searching method. By only using synchronized voltage phasor, the fault location accuracy of this method is not affected by the saturation of current transformer. The simulations carried out on IEEE 9 test system have shown that the proposed technique locates the fault point accurately and adaptively, excluding influence from factors such as fault resistance, fault type and so on.Sixthly, a WAMS/SCADA based mixed dynamic voltage phasor tracing algorithm is proposed to estimate the voltage phasor of unobservable buses due to topology change or PMU malfunction in power system. Modeling load characteristic with historic data from SCADA, this method uses Gauss-Seidel iteration method to solve the nonlinear equations and estimates the voltage phasor of unobservable buses with high accuracy voltage phasor of neighboring buses measured by PMU.At last, an architecture of high robustness wide area security monitoring system is presented by integrating the robustness enhancement methods proposed from chapter 2 to chapter 6 into the WAMS/SCADA/EMS system that may be equipped by Zhejiang electric power corporation in the coming years. This architecture integrates the advantages of WAMS, SCADA and FMS and employs the hybrid measurement technology to overcome the limitation of the single measurement system. Moreover, the optimal PMU placement method that meets the requirement of high robustness wide area security motoring system is studied.更多还原