智能配电网运行可靠性多机制协调概率评估方法研究

Operational reliability of smart electric distribution system (SEDS) will become the fundamental issue faced with by the system operator responsible for controlling SEDS operational security in the near future. The operational reliability evaluation of SEDS differs from the traditional distribution system reliability evaluation in essence on the application scenarios, that is, the former serves the purpose of predicting system operational reliability levels, while the later of judging historical system reliability levels affected by recorded environmental factors. In this project, firstly, in response to the fundamental issue of developing evaluation models for SEDS operational reliability analyses, a general operational-reliability-predicting model of SEDS transmission components is to be proposed with the consideration of the factors of rare records of component failure events, component aging as well as load levels resulting from the power flow. The proposed model is to be used for sampling system operational states which then serve as the basic materials for SEDS operational reliability analyses. Secondly, in response to the rareness of operational contingencies, a multi-coordinated-mechanism simulation method is to be proposed to circumvent the low efficiency of the crude Monte Carlo method once applied to SEDS operational reliability evaluation for operational decision-making. Finally, in response to various correlation characteristics of the generation and load demand, a multi-coordinated-mechanism simulation method considering both the power-to-power and power-to-demand correlation characteristics as well as the foregoing rareness of operational contingencies is to be proposed, expectantly as another basic analysis tool, further to help make effective decisions for controlling SEDS operational security.

智能配电系统运行可靠性问题将成为未来智能配电网运行安全所面临的一个重要问题。智能配电网运行可靠性评估与传统配电网可靠性评估在应用场景上全然不同，前者服务于系统运行可靠性水平的预测，而后者则是考虑历史运行环境作用下的系统可靠性评判。本项目针对智能配电网运行可靠性概率评估的基础模型问题提出一种计及稀少故障样本、设备老化以及潮流影响的配网元件运行可靠性预测模型，为系统运行可靠性分析提供系统运行状态采样的依据。针对运行可靠性预测中有效预想故障稀有性的问题，提出考虑有效预想故障稀有性的混合协调机制智能配电网运行可靠性概率评估方法，用于克服粗糙蒙特卡洛方法仿真效率的不足。鉴于配电网小气候特征所引起的能源与负荷功率相关性现象，提出考虑源-荷-源相关性以及主动负荷的混合协调机制智能配电网运行可靠性概率评估方法，为运行控制决策中需要同时考虑源-荷-源相关性和有效预想故障稀有性的问题提供基础分析工具。

配电网智能化是利用先进控制手段解决未来清洁高效终端供电的必经途径。虽然目前对智能配电网的概念尚未达成共识，但从现有文献中不难归纳出智能配电网所具有的共性特征，即，整合先进的通讯控制技术实现可再生能源的就地消纳与供需互动。毋庸置疑，无论是智能配电网的运行分析亦或规划研究，系统可靠性始终是复杂系统建设和运行需要考量的首要问题。本课题着眼于智能配电网的运行可靠性，重点围绕1) 计及稀少故障样本、设备老化以及潮流影响的配网元件运行可靠性预测模型，2）考虑有效预想故障稀有性的智能配电网运行可靠性混合协调机制概率评估方法，以及3) 考虑源-荷-源相关性以及主动负荷的智能配电网运行可靠性混合协调机制概率评估方法等三个核心问题，对智能配电网的运行可靠性开展了持续深入的研究。三年来课题取得的成果主要体现在一次载能设备短期可靠性故障率模型以及基于重要度采样的运行可靠性指标的快速评估方法上，并发表了SCI/EI期刊论文6篇。在本课题的资助下，通过与实际电网部门的合作，实现了零散的架空线、电缆和变压器故障-环境关联数据的收集和整合。所收集的数据为后继深入研究和充分论证本课题所提的模型和算法创造了必要的数据条件。本课题的核心贡献在于，从理论上分析了智能配电网运行故障小概率事件的成因，探索了克服效率事件仿真的途径，并有针对性地提出了不同问题模型下的采样概率畸变方法，为研制通用的Monte Carlo方差减小技术以适应未来多角度的智能配电网运行可靠性评价提供了基础的理论支撑。