摘要
电力负荷预测的准确性对于电力系统的安全经济运行具有重要的意义,但负荷的强随机性和气象相关性给负荷的预测带来了很大的困难。针对传统负荷预测方法存在效果不理想的问题,文章提出了基于改进深度神经网络的负荷预测方法。首先利用F-score特征评价准则选择出合适有效的特征量,再利用深度神经网络建立特征量与负荷之间的非线性映射预测模型。通过电力负荷预测实例的对比分析,结果表明本文方法在电力负荷预测中具有很好的预测精度和时效性。本文研究成果可为电力负荷的预测提供有效的技术参考和指导。
The accuracy of power load forecasting is of great significance to the safe and economic operation of power system,but the strong randomness of load and meteorological correlation make difficult to forecast the load.The traditional load forecasting method has the problem that the effect is not ideal,this paper presents a load forecasting method based on improved deep neural network,F-score feature evaluation criteria are used to select the appropriate and effective feature quantities firstly,then the deep neural network is used to the nonlinear mapping prediction model between characteristic quantity and load is established.Through the comparative analysis of power load forecasting example,the results show that the proposed method has good accuracy and timeliness in power load forecasting.The research results of this paper can provide effective technical reference and guidance for power load prediction.
作者
吕冰
白宇
L Bing;BAI Yu(State Grid Jibei Information&Telecommunication Company,Beijing 100053,China;Beijing Bowang China Science and Technology Co.,Ltd.,Beijing 100053,China)
出处
《微型电脑应用》
2023年第3期146-148,152,共4页
Microcomputer Applications