基于大数据的短期负荷预测技术

Short-term Load Forecasting Technology Based on Big Data

利用大数据技术中的深度学习技术建立电力负荷预测模型,并将某地区一年中的数据分为训练数据集和测试数据集,比例设置为7∶3。利用TensorFlow平台对模型进行仿真分析,结果表明利用前日数据对当日负荷进行预测具有更高的精度。预测一周电力负荷时,如果可以考虑时间信息,能进一步提升预测精度。该研究对实现短期负荷的精确预测,促进电力领域技术水平的提升具有重要的现实意义。

Using the deep learning technology in big data technology,a power load forecasting model is established,and the data information of a certain area in one year is divided into training data set and test data set,and the ratio is set to 7∶3.The model is simulated and analyzed by TensorFlow platform,and the results show that it is more accurate to predict the load of the day using the data of the previous day.When forecasting one-week power load,if time information can be considered,the forecasting accuracy can be further improved.This research is of great practical significance to realize accurate short-term load forecasting and promote the improvement of technical level in power field.