摘要
气温波动是影响短期电力负荷预测准确率的主要因素之一,为了提高预测精度,将气温因素引入一种新型的非线性自回归模型中,构建一种基于气温因素的非线性自回归短期电力负荷预测模型,并提出该模型的实验定阶方法。以气温作为模型的外部输入量,基于Weierstrass定理推导了该模型的表达式,采用最小二乘法估计该模型的参数,根据所提出的实验定阶方法对模型进行定阶。对实际电力负荷样本进行预测,结果验证了模型实验定阶方法的可行性,表明该负荷预测模型预测精度较高,可应用于负荷短期预测之中。
Temperature fluctuation is one of the main factors affecting the accuracy of short-time load forecasting. To obtain a precise short-term load forecasting, in this paper, the temperature factor is introduced to a new nonlinear auto-regressive model, a nonlinear auto-regressive model for short-term load forecasting based on temperature factor is built and a experimenting order determination method is proposed for the model. Using temperature factor as the external input, this model is deduced based on Weierstrass theorem, the parameters of this model are identified using the least square algorithm, the order of the model is determined based on the proposed experimenting order determination method. Making load forecasting based on actual load samples, the obtained results verify the feasibility of the experiment order determination method of the model and show that the proposed model has a good forecasting capacity and can be applied to the short-time load forecasting.
出处
《控制工程》
CSCD
北大核心
2016年第3期346-352,共7页
Control Engineering of China
基金
国家自然科学基金项目(61174032)
国家自然科学基金项目(61104183)
高等学校博士学科点专项科研基金(20130093110011)
关键词
短期电力负荷预测
气温因素
非线性自回归模型
最小二乘法
实验定阶
Short-term load forecasting
temperature factor
nonlinear auto-regressive model
least square algorithm
experimenting order determination
