摘要
智能电网已经是电网发展的必然趋势,随着新能源和分布式电源的并网以及其他智能设备的投入使用,将对发电计划、电力调度、电网经济运营、电力储存以及电网运营的安全性带来新的挑战。为了解决供电部门在负荷预测时需购买气象数据的问题,本文用神经网络模型预测了温度,并与气象部门的预报结果进行了比较,同时作为负荷预测的输入特征。负荷预测结果显示,两种方法的误差相同,在负荷预测方面,预测的温度已经可以取代气象预报温度。
Smart grid is the inevitable trend of electric grid enterprises. With the new energy and distributed energy accessed into electric grid, and smart devices coming into use, it is challenging to generation schedule, power switching, electricity storage and the economy and safety of grid operation. In order to solve the problem that the electricity sectors need to purchase meteorological data in the load forecasting, in this paper, the results of predicting the temperature by neural network model and meteorological department, as input characteristics of load forecasting, are compared. Load forecasting results show that the two methods of error are almost same, and meteorological forecast temperatures have can be replaced by forecast temperatures using neural network.
出处
《微计算机信息》
2011年第12期79-81,共3页
Control & Automation
关键词
智能电网
短期负荷预测
温度
模糊神经网络
smart grid
short-term load forecasting
temperature
neuro-fuzzy net
