基于决策树体系的预想故障集下风电场扰动风险测度评估

摘要

With the development of wind power and scale of wind farm, the spatial distribution difference between wind turbines also increase. Besides, wind turbine trip-off and damage accidents has occurred frequently because of the severe wind conditions, having adverse impacts on the stability and safety operation of power grid. Therefore, it is necessary to study the online risk assessment method for power system with wind energy. Considering the wind turbine spatial distribution difference, this paper proposed an online disturbance risk measure of wind farm based on decision trees, which can perform data mining on online information, and make fast judgement on voltage violation and wind turbine trip-off. Furthermore, according to the judgement of decision trees, disturbance risk measure indices are proposed, which are visualized and provide supportive information for wind farm and power system operators.%随着风力发电的大力发展及风电场规模的持续增加，风机的空间分布差异性愈发显著。此外，风机运行状态易受风电场元件故障、电网扰动等诸多因素的影响，因此，建立实时在线评估方法和预警机制已成为当务之急。该文考虑了风电场风机分布的离散特性，建立了风电场动态安全决策树体系，并提出风电场扰动风险测度指标。该决策树体系可利用在线信息进行数据挖掘，针对预想故障集下的风机电压越限与脱网状况进行快速分析判断，并根据判断结果输出扰动测度指标，为电网及风电场运行人员提供直观地风险水平及决策参考。通过风电场算例分析，验证了所提方法的有效性。