摘要
为了进一步提升电力系统短期负荷预测模型的精度,提出了计及储能调度因素的短期负荷预测模型。预测模型考虑了数量日益攀升的储能用户对负荷预测产生的影响,并针对基于电价和基于合同两种储能调度运行控制策略,分别构造与其对应的储能充放电模型。在改进后的负荷预测模型的构建过程中融入了与储能调度行为相关的电价因素和合同因素,并采用Elman神经网络(Elman-NN)进行短期负荷预测。仿真实例证明,计及储能调度因素的Elman-NN短期负荷预测模型预测精度优于传统的短期负荷预测模型,测试结果的平均相对误差和最大相对误差的平均值分别达到0.019 4和0.065 4,验证了该模型具有较好的预测性能及稳定性。
To further enhance the accuracy of short-term load forecasting model of power system,a short-term load forecasting model which takes energy storage dispatching factors into account is presented. By taking the influence of increasing energy storage users on load forecasting into account,the corresponding charge-discharge models are built under two control strategies of energy storage dispatching based on electricity price and contracts,respectively. The electricity price and contract factors related to energy storage dispatching are integrated into the improved load forecasting models,and the short-term load forecasting is carried out using Elman-neural networks(Elman-NN). The simulation result of an example shows that the prediction accuracy of the Elman-NN short-term load forecasting model considering the energy storage dispatching factors outperforms that of the traditional short-term load forecasting model,and the averages of normalized mean relative error and maximum relative error reach 0.019 4 and 0.065 4,respectively,which verifies that the proposed model has better prediction performance and stability.
作者
陈丽娜
撖奥洋
于立涛
张智晟
CHEN Lina;HAN Aoyang;YU Litao;ZHANG Zhisheng(College of Electrical Engineering,Qingdao University,Qingdao 266071,China;Qingdao Power Supply Company,State Grid Shandong Electric Power Company,Qingdao 266002,China;Key Laboratory of Smart Grid of Ministry of Education(Tianjin University),Tianjin 300072,China)
出处
《电力系统及其自动化学报》
CSCD
北大核心
2019年第7期57-63,共7页
Proceedings of the CSU-EPSA
基金
国家自然科学基金资助项目(51477078)
智能电网教育部重点实验室开放研究基金资助项目(2018)
关键词
电力系统
短期负荷预测
储能
ELMAN神经网络
实时电价
power system
short-term load forecasting
energy storage
Elman-neural networks(Elman-NN)
realtime electricity price
