摘要
针对电力系统中长期负荷预测会受到很多不定因素的影响,通过采用组合预测等维新息熵值法对中长期电力负荷进行建模,建立了基于等维新息熵值法组合预测数学模型。先是用最优加权几何平均法和灰色关联分析法算出单一预测模型的权重,接着由熵值法确定模型评价指标的相对权重,最终获得组合权重因子。在组合预测模型中引入了等维新息数据处理的思想,实现了变权重,使预测结果能够更加合理地反映负荷发展趋势;并通过寻找等维新息的最佳维数区,优化了等维新息熵值法组合预测模型,得到更高的预测精度。计算结果显示了采用等维新息熵值法对中长期电力负荷进行预测的有效性。
The mid-long term load forecasting for power system is usually influenced by uncertainties, the medium- and long-term load forecasting is modeled by combination forecasting on the basis of combined forecast equal dimen- sional innovation entropy method, The combined predictive model of equal dimensional new innovation entropy method in mid-long term load forecast is constructed. The optimal weighted geometric average method and grey correlational analysis are utilized to calculate the weights of each single model under each evaluation index, then the entropy method was introduced to compute the weights of model evaluation indexes. Finally, the combined weights are determined. In- novation and equal dimensionM operators was introduced in the combined forecasting model, thus the weight of combi- nation forecasting can be variable, and the predictive results reasonably reflect the variant tendency of power load, and seek the optimal dimension. At last, optimal combined foreeastlng model can be obtained and the accuracy of predic- tions can be enhanced. Testing results verifies the feasibility of equal dimensional innovation entropy method in mid- long term load forecasting.
出处
《电力系统及其自动化学报》
CSCD
北大核心
2014年第12期36-40,48,共6页
Proceedings of the CSU-EPSA
关键词
组合预测
熵值法
等维新息
最佳维数
变权重
combined forecast
entropy method
equal dimensional innovation
optimal dimension
variable weight