Prediction system for dynamic transmission line load capacity based on PCA and online sequential extreme learning machine
Jun Lin, Qinqin Zhang, Gehao Sheng, Yingjie Yan, Xiuchen Jiang · 2018
The accurate prediction and evaluation of the operating status of transmission line can provide technical support for the safe, economical and efficient operation of the power system. Most of the traditional methods just use a single parameter for the forecasting and analysis of transmission line. The running state of transmission line is also affected by weather conditions, operating conditions and many other factors. Due to the low quality of the monitoring data and the large randomness of the environmental conditions, the traditional methods have great limitations in the accuracy and timeliness of the prediction. In this paper, a load prediction method based on Principal Component Analysis (PCA) and online sequential extreme learning machine (OSELM) is proposed. Firstly, the principal component analysis is used to reduce the dimension of the original data, and the parameters with less influencing factors are eliminated. Main parameters are extracted to train the online sequential extreme learning machine. The practical examples show that the proposed method has better data fitting ability and higher prediction accuracy.