基于BP神经网络的极限传输容量预测方法

Prediction Method of Total Transfer Capability Based on BP Neural Network

传统断面极限传输容量计算采用线性回归的方法,将非线性电力系统转化为线性模型求解,计算过程复杂。故提出基于BP神经网络的非线性极限传输容量预测方法,把系统实际运行数据抽象转化为特征属性,采用基于K-means聚类的分组特征选择方法,降维后择优选取部分属性作为断面关键特征,再将其作为神经网络输入层向量,经过重复训练构建系统关键特征和极限传输容量的非线性对应关系。算例分析表明,基于BP神经网络的断面极限传输容量预测方法,既满足时效性要求又能提高预测的精确性。

The traditional cross-sectional total transfer capability calculation uses linear regression method to convertthe nonlinear power system into a linear model to solve, the calculation process is complicated. A nonlinear totaltransfer capability prediction method based on BP neural network is proposed. First, the actual operation data of thepower system is abstractly converted into feature attributes. A group feature selection method based on K-means clusteringis adopted. After dimensionality reduction,some attributes are selected as key features of the system cross section,and they are used as The neural network input layer vector,through repeated training,constructs the non-linearcorrespondence between the key feature quantity of the system and the total transfer capability of the cross-section.The analysis of the calculation example shows that the method can not only meet the requirements of timeliness butalso improve the accuracy of the prediction.