摘要
针对当前冷热电联供(Combined Cooling, Heating and Power, CCHP)系统优化调度研究缺少冷热电负荷预测的问题,提出了基于改进集成经验模态分解(Modified Ensemble Empirical Mode Decomposition, MEEMD)、模糊熵(Fuzzy Entropy, FN)和长短期记忆神经网络(Long Short Term Memory, LSTM)的CCHP系统负荷预测方法。使用MEEMD算法将原始数据分解为若干IMF分量,计算IMF分量的模糊熵并将模糊熵相近的分量相加,使用LSTM对相加后新的分量进行预测,最后将分量预测结果重构得到最终预测值。通过仿真并对比分析其他方法的负荷预测精度,证明所提预测方法具有良好的预测效果。
In order to solve the problem that the current research about day-ahead optimal scheduling of combined cooling,heating and power(CCHP)system is lack of load prediction,a method based on modified ensemble empirical mode decomposition(MEEMD),fuzzy entropy(FN)and long short term memory(LSTM)network was presented.MEEMD was used to decompose the original load series into several IMF components,calculate fuzzy entropy of each IMF components and merge the IMF components with the similar fuzzy entropy.LSTM network was then employed to predict each new component.Final prediction value was obtained by superposing each component prediction result.Through simulation and comparison with other methods,the superior prediction performance of the proposed is demonstrated.
作者
常雨芳
李金榜
段群龙
陈润
吴锋
黄文聪
CHANG Yufang;LI Jinbang;DUAN Qunlong;CHEN Run;WU Feng;HUANG Wencong(Hubei Key Laboratory for High-efficiency Utilization of Solar Energy and Operation Control of Energy Storage System,Hubei University of Technology,Wuhan 430068,China)
出处
《电工技术》
2021年第16期59-63,共5页
Electric Engineering
基金
国家自然科学基金资助(编号61903129,51977061)。
