【作者】 钟慧荣；

【导师】 顾雪平；

【作者基本信息】 华北电力大学 ， 电力系统及其自动化， 2012， 博士

【摘要】 近年来不断发生的大停电事故表明,大停电是现代电力系统必须面对的严重威胁。随着电力系统规模的日益扩大及社会对电力供应依赖度的不断提高,大停电事故造成的后果也越来越严重。作为电力系统安全防御的重要措施之一,研究电力系统大停电后的黑启动恢复问题和与之相关的优化控制技术对于减轻恢复控制负担、加快系统恢复速度、减小大停电事故带来的经济损失及社会影响具有重要意义。本文在全面学习和借鉴已有研究工作的基础上,对黑启动恢复中的黑启动及网架重构优化技术进行了系统深入的研究,论文工作主要包括以下几方面：(1)考虑后续恢复效果的黑启动方案评估方法研究。通过分析黑启动与后续系统恢复过程的相互关系,提取反映黑启动方案后续恢复效果的评价指标,并对原有的评估指标进行整合,构建一套新的黑启动方案评估指标体系；在此基础上,采用模糊层次分析法对黑启动方案进行评估；对黑启动方案的评估结果进行灵敏度分析,计算导致黑启动方案排序结果变化的指标及权重的灵敏度临界值。基于河北南网的评估案例证明了本方法的有效性。(2)扩展黑启动策略及优化方法研究。深入分析电力系统黑启动恢复过程,提出一个黑启动电源同时启动多个被启动电厂的扩展黑启动策略；分析扩展黑启动方案的优化目标,综合考虑被启动机组的启动约束及系统的运行约束,建立了扩展黑启动方案优化模型；分析建立的优化模型,研究其求解方法,开发相应的算法对扩展黑启动方案进行优化求解。提出的扩展黑启动方法,有多台机组可以在黑启动阶段得到恢复,可显著加快系统的恢复进程。(3)网架重构阶段机组恢复的分层协调优化方法研究。鉴于黑启动恢复中机组恢复具有分层的特征,将机组恢复的优化分解为网络重构层和厂内机组恢复层两个恢复层次分层求解；通过研究两层机组恢复间的相互影响,确定机组恢复的优化目标,计及机组启动、调度运行、网络结构的各种约束,建立机组恢复的优化模型,并结合规则处理与优化计算进行求解,确定系统中全部机组的最优恢复序列。该方法克服了已有方法中把一个电厂按一台机组处理或每个电厂仅考虑一台机组恢复的缺点,对大停电后的系统恢复具有重要价值。(4)网架重构阶段的负荷恢复优化方法研究。深入分析网架重构阶段负荷恢复的特点,综合考虑负荷恢复代价、负荷重要性、负荷恢复对后续网架重构的影响等因素,采用模糊层次分析法对各负荷的综合重要性进行评价,得到各个负荷的综合重要性权重；再以恢复的加权负荷量最大为目标,考虑各种约束,建立了此阶段负荷恢复的优化模型,并研究相应的求解方法,对网架重构阶段的负荷恢复进行优化。(5)基于机组分层协调恢复的网架重构优化方法研究。将网架重构的连续时间过程转化为一个多时段按次序进行的恢复过程,提出“分时段决策、总体寻优”网架重构策略,并将网架重构转化为网络层机组恢复、厂站层机组恢复和重要负荷恢复的多目标分层优化问题；基于以上建立的机组分层恢复和重要负荷恢复的优化模型,构建每一时段的网架重构多目标分层优化模型,并采用多目标的分层序列法进行优化求解。所提出的网架重构方法考虑系统中所有机组的恢复并同时对网架重构阶段的负荷恢复进行优化,实现了网架重构过程的总体优化决策。更多还原

【Abstract】 The extensive blackout occurred in recent years has proved that the widespread blackout is a serious threat to the modern power systems. As the scale of power system becomes larger and the modern society is more dependent on the power supply excessively, the damage due to widespread blackout is more and more severe. Therefore, as one of self-rescue measures, the research on power system black-start restoration and its optimization technique is of great importance to relieving restoration burdens, speeding up the system restorationre process, reducing the economic loss and social influence resulting from faults.After thoroughly investigating the previous work about the black-start issue, the optimization technique about balck start stage and network reconfiguration has been studied systematically, the main work of this thesis is described as following:1. An assessment method of power system black-start schemes considering subsequent restoration efficiency is proposed and its sensitivity is analysed. First, the interdependence of black-start and follow-up system restoration process is analysed thoroughly, the indexes reflecting subsequent restoration efficiency is extracted, and all indexes is conformed to construct a new assessment index system. Then, Fuzzy Analytic Hierarchy Process algorithm is proposed to evaluate the black-start schemes. And last, the sensitivity of the assessment result is analysed, and the marginal weight coefficient and marginal index which can lead to the change of the rank of black-start schemes are calculated.2. An extended black-start optimization decision-making method is proposed. Firstly, the black-start restoration process is analysed thoroughly, a novel extended black-start strategy is proposed, in which a single black-start source can be used to start more than one power plant simultaneously. Secondly, the mathematic optimazation model of the extended black-start strategy is presented by analysing the restoration target and the consideration of the constraints of generator’s start-up and power system’s operation comprehensive. Thirdly, the solving method is studied by analysing the optimazation modle, and the solution algorhthem is develped to determine the best extended black-start scheme.3. A hierarchical optimization method of unit restarting during network reconfiguration is proposed. In view of the layered feature of units’restarting during network reconfiguration, the unit restarting process during the network reconfiguration is treated as a two-layer restoration process, the network-layer unit restarting, and the plant-layer unit restarting, respectively. First, the interaction between the network-layer unit restarting and the plant-layer unit restarting is analysed, the model of layered unit restarting is established by the study of optimization target and the consideration of the generator’s start-up, run scheduling and power system’s operating constraints comprehensive. Then, rule processing and optimization calculation are combined to determine the units’restarting sequence.4. An optimization method of loads recovery during network reconfiguration is proposed. First, the characteristics of loads recovery during network reconfiguration is analysed. Second, the loads’comprehensive importance is evaluated by comprehensivly considering the load recovery cost, load characteristics, load importance and the influence of the load restoration to the succeeding network reconstruction comprehensivly, and their comprehensive weights are calculated. Third, the loads’ recovery optimization model is presented, which the optimization target is to maximize the total weighted loads with all constraints being satisfied, then, the solving method is studied, and the loads’recovery optimization during network reconfiguration is realized.5. An optimization method of network reconfiguration is proposed by introducing a two-layer unit-restarting framework for power system restoration. First, the continuous restoring process of network reconfiguration is changed into a multi-step sequential process through partitioning the time axis for the system restoration, a global optimization strategy of "modeling on discrete time steps, optimizing on the whole process" is developed, and the optimization of the network reconfiguration is modeled as a multi-objective optimization problem of the network-layer unit restarting, the plant-layer unit restarting and the important loads recovery. Second, the overall multi-objective optimization model is bulit based on the built optimization modles of network-layer unit restarting, the plant-layer unit restarting and the important loads recovery. Third, the lexicographic optimization method is employed to solve the multi-objective optimization problem by integrating the solving processes of the three sub-problems. The proposed optimization method of network reconfiguration can determine the restarting sequence of all the units and realize the coordinative optimization of the unit restarting and restoration of the important loads.更多还原