论文部分内容阅读
电力市场中的中长期电价受众多不确定因素影响,很难用传统的经验统计方法建模预测。现有的预测方法一般从系统模拟角度出发进行预测,需要较多的系统信息资料,难度很大。本文采用经验模式分解将电价分解成多个相对独立的分量,详细分析不同的因素对于这些分量的影响,从不同的时间尺度分析电价的影响因素。根据分量的特点有针对性地选择时间序列和支持向量机建模预测。实例研究表明该方法适合于分析和预测复杂的中长期电价。
The long-term electricity price in the power market is affected by many uncertainties and it is very difficult to model the forecast by traditional empirical statistical methods. The existing prediction methods are generally predicted from the system simulation point of view, requiring more system information and data, which is very difficult. In this paper, the empirical mode decomposition is used to decompose the electricity price into several relatively independent components. The effects of different factors on these components are analyzed in detail. The influencing factors of electricity prices are analyzed from different time scales. According to the characteristics of the components, the time series and SVM modeling are selected. Case studies show that this method is suitable for analyzing and forecasting complex long-term and medium-term electricity prices.