摘要
为了提高预测精度,提出了一种改进的最小二乘法线损预测技术。算法首次将核偏最小二乘回归算法应用于线损率预测。实验表明,提出的方法能够较好地克服变量相关性和非线性因素对预测模型的不利影响。
In this paper, in order to improve the forecast accuracy, put forward a kind of improved method of least squares loss forecast technology. Algorithm for the first time the kernel partial least squares regression algorithm is applied to the line loss rate prediction. Experiments show that, this method can overcome the variable correlation and prediction model of nonlinear factors on adverse effects.
出处
《科技通报》
北大核心
2012年第6期109-111,共3页
Bulletin of Science and Technology
关键词
线损率
核偏最小二乘回归
预测
line loss rate
kernel partial least squares regression
prediction
