摘要
短期电力负荷预测是电网平稳运行与安全调度的基础。然而,负荷序列的不稳定性增加了预测难度。为了提高短期电力负荷预测的精准度,本文利用CEEMDAN将原始负荷序列分解为若干固有模态函数和残差,并利用PE分别分析各个固有模态函数(IMF)的复杂度。最后,利用IPSO优化LSTM参数,将各子序列的预测结果进行累加,得到最终的预测负荷。本文通过对辽宁省沈阳市的实际负荷数据进行仿真模拟,将仿真结果与其他传统预测方法结果进行对比,结果证明该预测模型的误差更低,具有较高的预测精确度。
Short-term power load forecasting is the basis for smooth operation and safe dispatch of power grids.However,the instability of the load sequence increases the difficulty of prediction.In order to improve the accuracy of short-term power load forecasting,this paper uses CEEMDAN to decompose the original load sequence into several intrinsic mode functions and residuals,and uses PE to analyze the complexity of each intrinsic mode function(IMF)separately.Finally,IPSOis used to optimize the LSTMparameters,and the prediction results of each sub-sequence are accumulated to obtain the final predicted load.This paper simulates the actual load data of Shenyang City,Liaoning Province,and compares the simulation results with the results of other traditional forecasting methods.The results prove that the forecasting model has lower errors and higher forecasting accuracy.
作者
秦光宇
闫庆友
朱敬尧
Qin Guangyu;Yan Qingyou;Zhu Jingyao
出处
《价格理论与实践》
北大核心
2020年第2期75-78,174,共5页
Price:Theory & Practice
