摘要
提出了一种新颖的电力空间负荷分布预测模型,该方法首先对各类负荷的影响因素进行分析并分别建模预测;而后将选定区域划分成等面积小区,利用主成分分析法对小区空间信息进行处理,从而形成支持向量机的训练样本集;在此基础上用训练好的支持向量机计算待预测区域小区的属性值,并按照各类用地类型排序。根据预测结果,结合待预测区域的整体发展规划,给出待预测区域各类负荷增量;最后,结合各类负荷密度预测值、各类用地发展总量、各类用地发展排序,给出空间负荷预测值。实例验证了本文方法的有效性。
A new spatial load forecasting （SLF） model for distribution network is proposed. First, the characteristics of different kinds of power load are analyzed, then the appropriate forecasting methods are chosen and the forecasting accuracy are improved. Dividing the selected field into same small areas, extracting the spatial information of the small area, using principal component analysis method to conduct the spatial information of the well-rounded small area to form the training sample of the support vector machine, the attribute values of the small areas in the forecasting field are then calculated by the trained support vector machine, so the order table of the different field are formed. Based on the forecasting results and the development plan of the forecasting field, the incremental load and the different load density are attained, so the total amount of different fields can be calculated, at last the spatial forecasting results can be obtained under the former order table. Finally, this model is implemented and the analysis result for practical calculation example by this model is satisfied.
出处
《电工电能新技术》
CSCD
北大核心
2008年第1期40-43,共4页
Advanced Technology of Electrical Engineering and Energy
关键词
空间负荷预测
负荷细分
主成分分析法
支持向量机
spatial load forecasting
load subdivision
principal component analysis
support vector machine
