论文部分内容阅读
安全裕度低的电网关键断面是电网运行人员需重点关注的电网薄弱环节,对其进行准确预测是保证电网安全、稳定运行的重要技术手段。以广东电网为例,收集了该地区2014和2015年的电气量和气象数据。首先,将电气量与气象数据进行标准化和集成;其次,对特征全集进行特征选择,并利用神经网络模型进行训练,得到关键断面的神经网络预测模型。相比于传统方法,所提预测模型在电气量因素的基础上,引入了非电气量因素(气象因素),用以挖掘2种因素对电网安全运行中关键断面的影响。广东电网的算例测试表明,该模型预测准确性好、速度快,适应于复杂多变的实际电网。
The critical sections of the power grid with low safety margin are the weak points of the grid that the operators of the power grid should pay close attention to. The accurate prediction of the critical sections of the power grid is an important technical means to ensure the safe and stable operation of the power grid. Taking Guangdong Power Grid as an example, the electricity and weather data for the region in 2014 and 2015 were collected. Firstly, the electrical quantity and weather data are standardized and integrated. Secondly, feature selection is carried out on the feature corpus, and the neural network model is used to train the neural network prediction model. Compared with the traditional method, the proposed model introduces non-electrical factors (meteorological factors) on the basis of electrical factors to tap the influence of the two factors on the critical sections in the safe operation of power grid. A case study of Guangdong Power Grid shows that the model has good forecasting accuracy and fast speed and adapts to the complex and ever-changing actual power grid.