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：
影响中长期负荷因素众多,而且单一核函数支持向量机泛化或学习能力较弱,预测精度受限。提出一种结合粗糙集和支持向量机智能算法的负荷预测模型,通过属性约简算法筛选出影响长期电力负荷的核心影响因素,剔除冗余信息,选定全社会用电量、人均产值、产值单耗为输入变量,构建基于多项式核函数、径向基核函数的混合核函数支持向量机预测模型,有效提高函数的泛化及学习能力。算例结果表明,所提出的模型预测平均误差仅为0.59%,预测精度有了很大提高且适用于长期负荷预测。
英文摘要：
Many factors affect long-term load and the generalization and learning capability of single kernel function support vector machine are weak , which makes the prediction accuracy limited. A short-term load forecasting model based on rough sets and support vector machines is proposed to solve this problem. Through the rough set attribme reduction algorithm to find the core of the impact factor of load and remove redundant information, this paper selects the electricity consumption, per capita GDP, consumption per unit output value as input vector, then builds a mixed kernel function support vector machine prediction model based on polynomial kernel function and radial basis kernel function, which can effectively improve the function of generalization and learning ability. The simulation results show that the average error is only 0.59%, and compared with the traditional prediction model, prediction efficiency and accuracy with the new model are greatly improved and more applicable to long-term load forecasting.
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：