复杂大电网可靠性评估的均匀设计理论及应用

With the enlarging scale and expanding interconnection of bulk power systems (BPS), increasing of BPS operation modes, integrating of renewable energies and storage devices and growing of load demand response, modern BPSs have become increasingly large and complex. In general, computational complexity of BPS reliability evaluation increases 4-6 or even higher power with the number of buses. Decreasing computational complexity has, therefore, become one of the most difficult key problems in BPS reliability evaluation. In order to decrease the computational complexity, this project proposes a uniform design idea to produce system states instead of traditional methods, such as analytical enumeration methods and stochastic simulation methods, and then a uniform design theory system for BPS reliability evaluation can be built. This project analyzes reliability influence factors, such as operation modes and BPS structure, determines the level of each influence factor, and then proposes a uniform design technique for producing system states. Moreover, a method to improve uniformity is proposed based on a selected uniform criteria, and an approximately evaluation model of reliability indices is built. Relationship between uniformity and deviation, and method to reduce deviation of system states are also studied. This research project can be beneficial to improving BPS reliability evaluation efficiency and reducing its computational complexity, enhancing BPS reliability performance, reducing outage costs, and promoting of large-scale engineering applications including BPS reliability evaluation, planning, optimization and operations. This research project also provides theoretical and technical supports for operation and online reliability evaluation.

随着电网规模扩大、互联程度提高、运行模式增加、新能源和储能设备的接入、负荷需求响应增加等，现代大电网的规模日益庞大、结构日益复杂。通常，其计算复杂性随系统节点数成4-6次方甚至更高次方增长。因此，降低计算复杂性已成为大电网可靠性评估最棘手、最核心的问题之一。 为降低计算复杂性，本项目以"设计"思想代替解析枚举、随机模拟等系统状态产生方法，建立大电网可靠性评估的均匀设计理论体系：从运行和结构等角度分析可靠性影响因素，确定各因素水平，建立状态产生的均匀设计模型和算法；基于均匀性判断准则提出提高均匀性的方法，建立节点、系统可靠性指标模型，研究系统状态均匀性与误差的关联关系及降低误差（增强均匀性）方法。 通过项目研究，将有利于提高大电网可靠性评估的效率、降低计算复杂性，提高电网可靠性、减少停电损失，促进可靠性规划、优化、运行和改造等的大规模工程应用，为运行可靠性、在线可靠性等提供理论和技术支撑。

随着电网规模扩大、互联程度提高、运行模式增加、新能源和储能设备的接入等，现代大电网的规模日益庞大、结构日益复杂。通常，其计算复杂性随系统节点数成4-6次方甚至更高次方增长。因此，降低计算复杂性已成为大电网可靠性评估最棘手、最核心的问题之一。. 本项目分析了网络结构（线路、母线等的断路器引起的结构变换）、元件健康状况、外部和运行环境等对电力系统可靠性的影响；提出特高压直流输电系统的可靠性评估的“元件-子系统-系统”分层方法模型，降低其计算复杂性；基于电力设备的故障和运行两因素水平，提出可靠性评估中系统状态矩阵的均匀设计方法、系统状态的均匀性准则，建立系统可靠性评估的分层均匀设计模型和算法，大幅降低计算复杂性；探讨了系统状态均匀性与可靠性指标估计误差关联关系；针对风电、光伏等场站出力的多因素水平，建立含可再生能源电力系统可靠性评估的均匀设计模型和算法；给出均匀设计可靠性评估指标的区间估计和置信度分析；提出了大电网可靠性评估复杂性降低的最大风险方法；开展了均匀设计在风电场可靠性优化规划等研究。. 通过本项目研究，建立了大电网中特高压直流输电系统可靠性评估模型，给出了基于均匀设计的电网可靠性评估的系列理论及技术，实现了在可靠性规划、优化、运行和改造等的工程应用，提高了大电网可靠性评估的效率、降低计算复杂性为运行可靠性、在线可靠性等提供理论和技术支撑。