
基本信息:
- 专利标题: 基于决策形式背景的多因素自适应神经网络电力负荷预测方法及系统
- 专利标题(英):Multi-factor adaptive neural network power load forecasting method and system based on decision formal context
- 申请号:CN201810747257.1 申请日:2018-07-09
- 公开(公告)号:CN108805366A 公开(公告)日:2018-11-13
- 发明人: 李启昌 , 肖云东 , 宋强 , 李光肖 , 王琳 , 倪馨馨 , 何召慧 , 丁子甲 , 刘宗杰 , 邵士雯 , 刘庆华 , 杨峰 , 陆超 , 刘华利 , 张红兴 , 吴东 , 颜香梅 , 彭颖 , 刘莹 , 李怀花 , 谭媛
- 申请人: 国网山东省电力公司济宁供电公司 , 国家电网有限公司
- 申请人地址: 山东省济宁市高新技术开发区火炬路28号;
- 专利权人: 国网山东省电力公司济宁供电公司,国家电网有限公司
- 当前专利权人: 国网山东省电力公司济宁供电公司,国家电网有限公司
- 当前专利权人地址: 山东省济宁市高新技术开发区火炬路28号;
- 代理机构: 济南圣达知识产权代理有限公司
- 代理人: 董雪
- 主分类号: G06Q10/04
- IPC分类号: G06Q10/04 ; G06Q10/06 ; G06Q50/06
The invention discloses a multi-factor adaptive neural network power load forecasting method and system based on a decision formal context. The method comprises the following steps: obtaining historical power load values and factors influencing the historical power load values as input data for system forecasting; discretizing the input data; using an attribute reduction algorithm of the decisionformal context to reduce the input data, and extracting key historical load data values and key external influence factor values influencing load forecasting; and inputting the extracted key historical load data values and key external influence factor values into an adaptive neural network system, and obtaining the power load forecasting values through training. The multi-factor adaptive neural network power load forecasting method and system fully consider the influence of external factors on power load, and can make a power load forecasting structure more accurate.