【作者】 纪静；

【导师】 谢开贵；

【作者基本信息】 重庆大学 ， 电气工程， 2011， 博士

【摘要】 通常,增装变压器、线路、断路器等电力系统元件后,系统可靠性会得到改善,其称为可靠性同调现象。目前,很多可靠性评估的研究成果都是以该假设为前提。但是,由于电力系统及其可靠性评估模型的非线性特性,系统中存在着可靠性非同调现象,即:增装一个电力元件后,系统可靠性没有得到改善甚至发生恶化。该类元件的存在不仅增加了设备投资,而且降低了系统可靠性。本文受国家自然科学基金项目“电力系统可靠性非同调机理及非同调元件辨识研究(50777067)”的资助,开展了可靠性非同调的产生机理、非同调指标体系和非同调元件辨识方法等研究,重点分析了计及与不计变电站电气主接线的多回平行输电线路、配电变电站及发输电组合系统的可靠性非同调特性和非同调影响因素,这可为电力系统可靠性优化、运行等提供更充分的决策依据,对系统的可靠经济运行具有重要意义。本文主要内容如下:①分析了电力系统可靠性非同调机理,提出综合反映系统可靠性非同调特性的指标体系,并应用于多回平行输电线路、配电变电站等系统的非同调分析。分析了简单串并联系统以及复杂系统中可靠性非同调现象的产生机理,即将增装元件对系统可靠性的影响分为正效益和负效益两方面。当正效益大于负效益时,增装元件改善了系统可靠性,元件为同调元件;反之,为非同调元件。对于复杂电力系统,以期望缺供电量等指标作为可靠性非同调现象的判断依据,提出刻画电力系统可靠性非同调特性的指标体系,包括:可靠性非同调概率、非同调频率、非同调时间及非同调电量等。②分析不计与计及变电站电气主接线的多回平行输电线路的可靠性非同调特性,在可靠性评估模型基础上建立非同调元件辨识模型,并得到非同调临界负荷的解析模型。给出了两回、三回以及四回平行输电线路的可靠性评估解析模型,基于可靠性非同调定义分别建立其可靠性非同调辨识模型,应用线路容量、可靠性参数等建立解析的非同调临界负荷模型。若要判断某系统是否非同调系统,只需将系统实际负荷水平与临界负荷相比较即可,若系统负荷大于临界负荷,系统同调;否则,系统非同调。该辨识模型为解析模型,无需进行可靠性评估即可直接进行非同调辨识。同时,分析了元件可靠性参数及负荷水平等因素对系统的非同调特性、非同调临界负荷以及系统非同调指标的影响。③分析配电变电站单母线分段接线和桥形接线两种典型接线方式的可靠性非同调特性,分析了该类系统可靠性非同调现象产生的条件和影响因素,提出配电变电站可靠性非同调元件辨识模型。分析结果表明:负荷水平是影响可靠性非同调特性的最主要因素,元件可靠性参数的影响相对较小,且仅有断路器故障率会改变系统可靠性非同调特性,其它参数如断路器修复率、变压器的故障率和修复率等仅影响系统可靠性,不会改变非同调特性。④分析了发输电组合系统元件退出前后可靠性的变化关系,探讨了发输电组合系统可靠性非同调元件辨识指标,并提出基于辨识指标的可靠性非同调元件辨识算法。为了减少辨识可靠性非同调元件的计算复杂性,借鉴电力系统分析中的故障排序思想,分析了发输电组合系统中可能存在的可靠性非同调现象,提出用输电损耗、系统负载率、最小支路负荷安全裕度和带权重线路介数等指标单独作为非同调元件辨识的辅助指标或加权得到一个非同调辨识综合指标,以反映元件退出前后可靠性变化规律。基于此,提出发输电组合系统可靠性非同调辨识算法。以IEEE-RTS 79系统为例进行算例分析,分析表明:前3个指标与系统可靠性指标之间具有明显的相关关系,但指标的大小不能完全反映出元件退出前后可靠性的变化规律,综合指标也只能识别出系统中的部分可靠性非同调元件。因此,尽管该模型可实现发输电系统可靠性非同调元件的初步辨识,但非同调元件的辨识指标还需更加深入的研究。更多还原

【Abstract】 Generally, adding an electrical power component to a power system, such as generating unit, transformer and transmission line, can improve the system reliability performance, which is the so-called reliability coherence phenomenon in power systems. Many research achievements on the power system reliability evaluation are based on this rule. However, the reliability non-coherence (RN) phenomenon exists in power systems due to the nonlinearity of power systems and their reliability evaluation models, which refers to the fact that if one or more components are added to a power system, the system reliability would not be improved, or even become worse. This component is designated as the reliability non-coherence component (RNC). The RNC not only increases the component capital investment and maintenance cost, but also reduces the system reliability performance.Supported in part by the National Natural Science Foundation of China (“Study on the RN mechanism and recognizing the RNCs of power systems”, No. 50777067), this thesis studies the existence mechanism of the RN phenomenon, non-coherence indices, and methods for recognizing RNCs. It mainly focuses on the RN feature, influencing factors and RNC recognizing techniques of multiple parallel transmission lines (MPTL) with and without considering substation configurations, distribution substation configurations (DSC) and composite generation and transmission system (CGTS). This study can provide a useful decision information for power system reliability optimization and operations, and is of important significance to the reliability and economy of power system operations. This thesis mainly includes the following contents:①The power system RN mechanism is analyzed, and the index system used to describe the degree of the system RN feature is proposed and applied in the RN analysis of MPTLs and DSCs. The RN mechanism of the simple series, parallel systems and complex systems is analyzed, and the impact of adding a component on the system reliability is divided into two parts: positive effect and negative effect. If the positive effect is larger than the negative effect, adding a component can improve the system reliability and the component is the reliability coherence component. Otherwise, the component is a RNC. The Expected Energy Not Supplied (EENS) index or other indices is used to act as the RN judgment standard for the complex power systems, and the RN index system is proposed to describe the degree of the system RN feature, which includes: RN probability, RN frequency, expected RN time and expected RN energy.②The RN features of MPTLs with and without considering substation configurations are analyzed, the RN identification models are built based on the analytical reliability evaluation model, and the analytical model of the RN threshold load (RNTL) is obtained. The reliability analytical models for double-circuit transmission line (DCTL) system, three-circuit transmission line (TCTL) system and four-circuit transmission line (FCTL) system are deduced. The RN identification models for MPTLs are also built based on the definition of RN, and the analytical model of the RNTL is obtained based on the transmission line rated capacity and component reliability parameters. Without evaluating the system reliability, the RN feature of a MPTL can be judged using the comparison between the system load and the RNTL. If the system load is larger than the RNTL, the system has the reliability coherence feature; otherwise, the system has the RN feature. In addition, the impacts of component reliability parameters and load levels on the RN feature, RNTL and RN indices are also analyzed.③The RN features, the existing conditions and influence factors of the RN phenomenon for DSC including the sectionalized single bus and bridge connections are analyzed, and the RNC identification model for DSC is built. The simulation results show that load level is the most important influence factor of RN feature. The component reliability parameters have a little impact on the system RN features. Only the failure rate of breakers can change the system RN feature; other parameters, such as the repair rate of breakers, the failure rate and the repair rate of transformers, only affect the system reliability and can not change the non-coherence feature.④The reliability changing rule before and after removing a component from CGTS is analyzed. The identification indices of RNC in CGTS, which can be used to recognize the RNC, are discussed. In order to reduce the computation complexity of recognizing the RNCs, the thesis analyzes the RN phenomena of CGTS by using the experience of contingency ranking technique for power system, and proposes the indices, such as the transmission losses, system loadability, the minimum security margin of transmission lines and the weighted betweenness of transmission lines, or the comprehensive index which obtained by weighing the before-motioned indices, to recognize the RNCs of CGTS. The IEEE-RTS 79 is used as an example to illustrate the rationality of the proposed concepts. The analysis shows that there are significant correlation between the first three indices and the system reliability indices; however, the rank of the indices can not reflect perfectly the reliability changing rule, and the comprehensive index only recognizes some but not all RNCs in power systems. Therefore, although the algorithm can identify the RNCs preliminarily, the identification indices for CGTS still need more research.更多还原