基本信息:
- 专利标题： 一种用电量的预测方法及装置
- 专利标题（英）：Electricity consumption prediction method and device
- 申请号：CN201510854215.4 申请日：2015-11-30
- 公开(公告)号：CN105512751A 公开(公告)日：2016-04-20
- 发明人： 单体华 , 秦砺寒 , 王智敏 , 史智萍 , 李顺昕 , 朱全友 , 聂文海 , 黄毅臣 , 张海霞 , 张晋梅 , 赵炜炜 , 杨金刚 , 赵国梁 , 陈丹 , 杨敏 , 岳云力 , 李博 , 李笑蓉 , 吕昕 , 范荻 , 宋堃 , 朱正甲 , 李莉 , 岳昊 , 王旭冉 , 牛东晓 , 韩江磊 , 夏慧聪
- 申请人： 国家电网公司 , 国网冀北电力有限公司经济技术研究院
- 申请人地址： 北京市西城区西长安街86号
- 专利权人： 国家电网公司,国网冀北电力有限公司经济技术研究院
- 当前专利权人： 国家电网公司,国网冀北电力有限公司经济技术研究院
- 当前专利权人地址： 北京市西城区西长安街86号
- 代理机构： 北京三友知识产权代理有限公司
- 代理人： 王涛
- 主分类号： G06Q10/04
- IPC分类号： G06Q10/04 ; G06Q50/06
The invention provides an electricity consumption prediction method and device and belongs to the power load prediction technical field. The method comprises the following steps that: historical electricity consumption data of a region to be predicted in each period and corresponding iron and steel material historical price data are obtained; smoothing is carried out; an optimal influence in-advance period of iron and steel material prices to electricity consumption is determined according to a time difference correlation analysis model; a support vector regression machine model is constructed according to a kernel function, smoothed steel material historical price data and historical electricity consumption data; and the electricity consumption of the region to be predicted is determined according to the support vector regression machine model and iron and steel material prices under the optimal influence in-advance period. With the electricity consumption prediction method and device adopted, the problems of relatively large prediction error and low accuracy of prediction results of electricity consumption in the iron and steel industry can be solved.