电网大规模实测过电压统计分布特性及其对典型绝缘影响的基础研究

长期以来，电力系统缺乏大规模过电压实时数据，目前采用分压器、套管末屏等难以普及推广的方法获取的实测过电压数据量少，导致实测过电压统计分布规律缺失，严重影响电力系统绝缘配合优化和安全运行，并将制约未来大电网的实时可观和系统建模。本项目以常规电磁式电压互感器为过电压监测手段，研究互感器铁芯磁滞和深度饱和物理过程的表征方法，建立电磁式电压互感器响应特性的非线性宽频模型，还原系统过电压幅值和波形；进而从电网已普遍安装的海量电磁式电压互感器中实时准确获取大规模过电压数据，研究并提出大规模实测过电压的数据清洗和分类识别方法，提取不同类型过电压的波形参数、频度和强度特征，揭示电网过电压时空分布规律和统计特性，构建表征电网过电压全信息的“过电压数据字典”；试验研究电力系统典型绝缘结构在实测过电压下的统计失效特性（放电概率曲线），提出统计绝缘配合改进方法，为电网科学绝缘配合和安全运行提供理论依据和方法支撑。

For a long period, large-scale real-time overvoltage data are never available. Nowadays, voltage dividers, bushing tap sensors, and other sensors have been successfully invented to obtain overvoltage data, but they cannot be widely applied, which leads to a lack of overvoltage data. Therefore, it is difficult to determine the statistical distribution of real-time overvoltage data, which greatly affects the optimization of insulation coordination and safe operation of power systems. It also restricts the real-time observability and system modeling of future big power grids. In this project, we plan to obtain overvoltage data using the widely used conventional electromagnetic voltage transformers. By representing the physical process of the core hysteresis and its deep saturation, a non-linear wide-band model of magnetic potential transformers is established to restore the amplitude and waveshape of the overvoltages. Then, massive overvoltage data can be precisely obtained through existing potential transformers in substations. Based on the obtained overvoltage data, data cleansing, classification, and identification methods are proposed. The waveform parameters and the characteristic quantities representing their strength and frequentness are also extracted. The time-space distribution and their statistical characteristics are extracted to establish an "overvoltage data dictionary" including all information from the massive overvoltages. The failure properties (discharge probability curves) of typical insulation structures under the practical overvoltages are experimentally obtained to support our proposed modified insulation coordination. The study in this project can provide a theoretical foundation and data/technical support for the scientific insulation coordination and safe operation of power grids.