基于Spearman相关性阈值寻优和VMD-LSTM的用户级综合能源系统超短期负荷预测

Ultra Short-term Load Forecasting of User Level Integrated Energy System Based on Spearman Threshold Optimization and Variational Mode Decomposition and Long Short-term Memory

由于用户级综合能源系统(integrated energy system,IES)的多元负荷序列之间复杂的耦合关系及易受外部因素影响等原因,综合能源系统多元负荷的精准预测面临很大困难。为此,提出一种基于Spearman相关性分析阈值寻优(threshold optimization,TO)和变分模态分解结合长短期记忆网络(variational mode decomposition based long short-term memory network,VMD-LSTM)的多元负荷预测方法。首先,使用斯皮尔曼等级(Spearman rank,SR)相关系数定量计算多元负荷间以及负荷与其他气候因素间的相关关系并通过循环寻优确定最优相关阈值,然后采用VMD算法将以最优阈值筛选出的负荷特征序列分解成更简单、平稳、有规律性的本征模态函数(intrinsic mode function,IMF)后与最优气象特征一起输入LSTM模型进行负荷预测。通过某用户级IES的实际数据对所提方法的有效性进行了验证,结果表明,所提方法能有效提高IES的多元负荷预测精度。

The integrated energy system(IES)faces great difficulties because of the strong complexity of the multivariate load series of IES at the user level,which is readily influenced by external factors.For that reason,this paper proposes a load forecasting way based on Spearman correlation threshold optimization,which integrates with variational mode decomposition(VMD)and long short-term memory network(LSTM).To start with,the Spearman rank(SR)correlation coefficient is used to quantitatively calculate the correlation between multiple loads and between loads and other climate factors,and the optimal correlation threshold is determined through cyclic optimization.Then,the VMD algorithm is used to decompose the load characteristic series screened based on the optimal threshold into simpler,more stable the regular intrinsic mode function(IMF)components are input into the LSTM model together with the optimal meteorological characteristics for load forecasting.The effectiveness of the proposed method is verified by the actual data of a user level IES,and the result was indicative of that the way can validly improve the accuracy of the multivariate load forecasting of IES.