节点文献

电网可靠性评估方法及可靠性基础理论研究

Research on Reliability Evaluation Approach of Power Grid and Reliability Basic Theory

【作者】 汪隆君

【导师】 王钢;

【作者基本信息】 华南理工大学 , 电力系统及其自动化, 2010, 博士

【摘要】 客观定量地评估电力系统可靠性能探明系统的薄弱环节,指导电力系统规划、生产与维护,以实现电力系统安全、稳定、可靠等运行目标,达到可靠性与经济性的有机统一。近年来,一方面,为实现更大范围的资源优化配置,促进我国能源集约化开发和高效率利用,促进经济与社会的可持续发展,多项高压/特高压直流输电工程陆续投运,受端电网形成了多直流馈入系统;输电网规模日益庞大且复杂。另一方面,为应对全球能源危机和气候变暖,清洁可再生能源的开发利用和智能电网的研究建设,已成为能源战略的重要组成部分。分布式电源出力的强随机性和间歇性,大大增加了电网规划与运行的不确定性;以高级资产管理体系为基石的智能电网尤为强调基于可靠性条件的状态检修技术。因此,电力系统可靠性研究与应用领域正面临着许多重大科学技术问题亟待攻关。为此,本论文针对高压直流输电系统概率风险评估、输电系统充裕度与安全性评估以及可靠性基础理论与应用研究中存在的若干问题进行科学性的探索,主要研究工作和成果如下:1.针对特高压/高压直流系统运行特性和结构特点,在对各子系统划分与建模的基础上,提出直流系统可靠性评估模型及其评估方法,计及直流控制保护系统和无功补偿对系统可靠性的影响。以换相电压时间面积为换相失败判据,在构建换相失败风险评价指标体系的基础上,提出了交流系统故障诱发多直流馈入系统换相失败风险评估方法,计及了多个不确定因素的综合作用。以南方电网实际直流系统为算例实现了对所提风险评估方法的验证。2.针对交直流混联系统运行条件复杂、运行约束多的特点和直流系统的运行特性,建立了以直流极功率方向系数为控制变量的最优负荷削减模型,有效计及了互联电网间直流系统功率支援与功率调制功能,以及由运行条件约束导致的直流极投退等工程实际问题。同时,为了克服平启动时原始对偶内点法收敛困难的局限性,引入了一种将初值寻优过程等效为求取动态系统稳定平衡点的思想,基于伪暂态连续法实现初值寻优。经IEEE RTS96交直流混联输电可靠性测试系统测试,验证了所提模型更加符合工程实际,求解方法收敛性能好、鲁棒性强,为大规模交直流混联输电系统充裕度评估提供了强有力工具。3.为避免确定性暂态稳定分析结果过于保守与不全面等问题,以安全域概率稳定模型为基础,基于动力系统理论确定电力系统暂态稳定主导不稳定平衡点附近的动态安全域线性边界,并采用以半不变量为基础的埃奇沃斯级数展开获得随机变量加权的联合概率分布,从而构建电力系统暂态稳定概率模型。该模型能有效计及发电机的出力和内电势、负荷的有功功率和无功功率、故障类型、故障位置和故障切除时间等多个不确定性因素。较传统方法,本文模型和方法不需要繁琐的公式推导,物理意义清晰,实现简单,计算效率高。通过IEEE 10机39节点系统对上述模型和方法进行了全面的验证。4.为克服电力系统可靠性应用研究中假设预防性检修间隔时间和修复持续时间为指数分布造成的模型失真,本文从可靠性数学基础理论出发,以电力系统继电保护为例,计及其预防性检修,研究建立了基于半马尔可夫过程的具有多停运模式的系统可靠性模型及其解析求解方法;提出了以保护年均经济损失最小为目标确定单一保护和双重化保护的最优预防性检修策略;研发了继电保护最优预防性检修分析决策系统。以某电网典型保护系统进行了验证,结果表明所提模型和方法具备计算速度快、精度高的优势。5.为彻底解决电力系统可靠性研究以状态转移函数服从指数分布为假设前提造成的问题,提高电力系统可靠性研究的精细化水平,以电力设备可靠性研究为例,建立了基于位相型分布的可靠性解析模型及其矩阵解析求解方法;以可靠性应用研究的电力设备检修优化为例,分别建立了以可靠性最优和经济性最优为目标的检修优化模型;通过电力变压器检修优化算例,实现了上述模型和方法的全面验证。为解决经典分布无法反映分布式电源有功出力间歇性问题,建立风力发电机与光伏发电系统出力的位相型分布随机模型,在此基础上构建了随机潮流的位相型分布模型,并采用以半不变量法为基础的埃奇沃斯级数实现其高效求解,另外还分析了分布式电源、无功补偿对支路负载安全性和电压质量的影响。本文研究工作得到了国家自然科学基金项目(50337010,50977032),国家重点基础研究发展计划项目(973项目)(2009CB219704),“十一五”国家科技支撑计划重大项目(2006BAA02A30),粤港关键领域重点突破项目(2009A091300011),广东省高新技术产业化重点项目(2010A010200005)的资助。

【Abstract】 Objective and quantitative reliability evaluation of power system can explore the weakness and point out a way for power system planning, operation and maintenance, in order to achieve the goal of keeping security, stability and reliability of power system operation, realizing the integration of reliability and economy. Recently, aiming at economical and social sustainable development, achieving greater scope of resource allocation optimization and promoting the development of China’s energy intensive and efficient utilization, the receiving-end grid forms a large scale multi-infeed high voltage direct current system thanks to several high voltage / ultra high voltage direct current transmission projects commissioning successively. Meanwhile, to cope with energy crisis and global warming, the development and utilization of renewable energy and the research of smart grid have been the important component of energy strategy. However, distribution generation is stochastic and intermittent so that it throws more uncertainty into power system planning and operation. Based on advanced asset management, smart grid attaches special importance to reliability condition-based maintenance. Therefore, power system reliability research and application are facing huge amounts of main technical problems to overcome. For this reason, the dissertation discussed probabilistic risk evaluation of HVDC transmission system, adequacy and security evaluation of transmission system, and the basic theory of reliability. The main work is as follows:1. According to the operation characteristics and structure features of UHVDC/HVDC system, the reliability model and evaluation of DC systems was presented based on modeling of subsystems, considering the effects of protection and reactive compensation on system reliability. This paper built the commutation failure risk index system and further proposed the risk evaluation approach of multi-infeed HVDC system under AC system fault conditions considering multiple uncertain factors, which took the commutation voltage-time area as criterion. The DC systems of China south grid is used to validate the risk evaluation.2. An optimal load shedding model, which took polar power-direction coefficient as controlling variable, was proposed, for the complicated conditions and multi-constraints of AC/DC hybrid transmission system and HVDC operation characteristic. The model not only takes into account the outage of DC polar due to operation restricting, but also considers the case of HVDC power reverse. Meanwhile, to overcome the shortcomings of convergence difficulties of current primal-dual interior point methods with flat starting, the paper applied the pseudo transient continuation method to seek the optimal initial point, introducing the idea of converting the problem of finding an initial point into that of finding a stable equilibrium point of active-set based dynamic system. The numerical results demonstrate that the proposed approach provides a novel powerful tool for the adequacy evaluation of AC/DC hybrid transmission systems.3. The linear boundary of dynamic security region of the controlling unstable equilibrium point was determined on the basis of dynamical system theory. Then, Edgeworth series expansion based on semi-invariant was employed to calculate the weighted joint probability distribution. Consequently, power system transient stability probabilistic model, which avoided conservative and incomplete results using determinate analysis of transient stability, was established. The proposed transient stability probabilistic model can effectively take many uncertain factors into account, such as generator output and internal potential, load active power and reactive power, fault location, fault type, fault-clearing time, et al. Finally, the numerical results of IEEE 10-machine 39-bus system indicated that the proposed model and its solution are effective and feasible, and have broad prospects for engineering application.4. To overcome the distortion caused by the assumption that preventive maintenance intervals and repair time are exponential, this paper studied the system reliability model with multi-outage states and its analytical solution based on semi-Markov process derived from the fundamental theory on reliability mathematics, took protection system and its preventive maintenance for the instance. The paper also proposed optimum routine maintenance intervals for protection by minimizing the annual average economic losses, and then developed optimum routine maintenance decision system for protection. The numerical results of typical protection systems showed that the advantages of the proposed method are also reflected in high precision and fast speed.5. To break the limitation of assuming state transfer functions follow exponential distribution and to improve the fineness level of reliability study, reliability studies of electrical devices as an instance, a reliability model based on phase-type distributions was proposed, deriving from the stochastic process fundamental theory, which was solved by matrix-analytic method. On the basis, the optimum maintenance models aiming at reliability and economic were designed, respectively. To solve intermittent problem of distributed generation that classical distribution can not reflect, the paper also built stochastic models of wind turbine and solar photovoltaic system based on phase-type distribution, employed Edgeworth series to solve probabilistic power flow model on the basis of semi-invariant. At the end, the effects of distributed generator and reactive compensation on branch load security and voltage quality were analyzed.This dissertation is supported by the nation natural science foundation of China (No. 50337010, 50977032), the national basic research program of China (973 Program) (No. 2009CB219704), the ministry of science and technology of the people’s republic of China (No. 2006BAA02A30), the crucial field and key breakthrough project in“Guangdong- Hongkong”(No. 2009A091300011), and high-tech industrialization key project in Guangdong (No. 2010A010200005).

节点文献中: 

本文链接的文献网络图示:

本文的引文网络