欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对电力市场中日前24点电价特性差异较大、采用单一模型很难描述的特点，建立多个模型分别对其进行预测，将数据空间按时点划分成24个子空间，然后根据这些子空间的相似性通过自组织映射对其进行自动聚类，并在不同类别的子空间分别建立支持向量机模型进行训练和预测。应用上述方法对PJM电力市场2005年8月的31天日前24点电价进行预测，结果表明该方法能够有效提高预测精度。
英文摘要：
According to the fact that in electricity market the features of electricity prices at day-ahead 24 points in time differ greatly and it is hard to describe these features by a single model, so multi-models are built to forecast them respectively. In accordance with points in time the data space is divided into 24 sub-spaces, then in the light of the similarity of these sub-spaces, they are automatically clustered by means of self-organizing mapping （SOM） and the support vector machine （SVM） models are established in different classified sub-spaces to perform training and forecasting. Using this method, the electricity prices at day-ahead 24 points of PJM electricity market on Aug. 3 1, 2005 are forecasted. Forecasting results show that the proposed method can improve forecasting accuracy effectively.
同期刊论文项目
同项目期刊论文
Multi-objective coordination optimal model for new power intelligence center based on hybrid algorit
Research on the wavelet neural networks based on data mining and its application in daily load forec
期刊信息
位置：