摘要
为提升电采暖负荷的预测精度,提出CEEMDAN-PSO-Elman的电采暖短期负荷预测方法。针对电采暖负荷的非线性与波动性问题,采用自适应噪声完备经验模态分解(CEEMDAN)将历史电采暖负荷分解为若干个子序列,使得电采暖负荷更具规律性与平稳性。针对电采暖负荷序列的时间相关性特点,采用Elman神经网络预测模型,提高了对历史数据的敏感性可以充分挖掘电采暖负荷的时间关联性,并采用粒子群(PSO)算法优化Elman神经网络参数解决Elman神经网络易陷入局部极小值问题。最后叠加各子序列模型的预测结果得到最终预测结果。以新疆某地区的电采暖负荷实测数据进行仿真分析,与其它预测方法对比,结果表明上述模型有效的提高了电采暖短期负荷预测精度。
In order to improve the prediction accuracy of electric heating load,the CEEMDAN-PSO-Elman short-term electric heating load prediction method is proposed.Aiming at the problem of nonlinearity and volatility of electric heating load,the adaptive noise complete empirical mode decomposition(CEEMDAN) was used to decompose the historical electric heating load into several sub-sequences,making the electric heating load more regular and stable.Aiming at the time correlation characteristics of the electric heating load sequence,the Elman neural network prediction model was used to improve the sensitivity to historical data and the time correlation of the electric heating load can be fully explored.And the particle swarm optimization(PSO) algorithm was used to optimize the Elman neural network parameters to solve the problem of the Elman neural network easily falling into a local minimum.Finally,the prediction results of each sub-sequence model were superimposed to obtain the final prediction result.The simulation analysis was carried out based on the measured data of electric heating load in a certain area of Xinjiang,and compared with other forecasting methods,the results show that the model effectively improves the short-term load forecast accuracy of electric heating.
作者
陈中慧
王海云
常喜强
徐森
CHEN Zhong-hui;WANG Hai-yun;CHANG Xi-qiang;XU Sen(College of Electrical Engineering,Xinjiang University,Urumqi Xinjiang 830047,China;State Grid Xinjiang Electric Power Co.,Ltd,Urumqi Xinjiang 830011,China)
出处
《计算机仿真》
北大核心
2022年第5期432-437,共6页
Computer Simulation
基金
自治区教育厅重点项目(XJEDU2019I009)
自治区实验室开放课题(2018D04005)
国家自然科学基金(51667020)。
关键词
电采暖负荷预测
自适应噪声完备经验模态分解
粒子群优化
Electric heating load forecast
Ensemble empirical mode decomposition with adaptive noise
Particle swarm optimization(PSO)
