针对热力系统参数运行数据预测困难，准确率低的问题。基于灰狼算法(Grey Wolf Optimizer，GWO)、变分模态分解(Variational Mode Decomposition ,VMD)、长短期记忆模型(Long Short Term Memory ,LSTM)提出一种单参数时序预测方法。首先使用改进适应度函数的GWO对VMD的分解层数和惩罚系数进行寻优，其次以最优参数对运行数据进行VMD并筛选出本征模态函数(intrinsic mode function ,IMF)分量作为原始数据趋势项，最后以此运行参数趋势项作为LSTM的训练集输入特征向量构建LSTM，LSTM超参数由北方苍鹰算法(Northern Goshawk Optimization，NGO)得到。经实际案例验证，该方法与原始数据直接作为训练集的LSTM相比，有效提高了LSTM对热力参数运行趋势预测的时间尺度与精度。

It is difficult to predict the operation data of thermal system parameters and has low accuracy. Based on gray wolf algorithm, variational modal decomposition and long short-term memory mode, a one-parameter time series prediction method is proposed. Firstly, the number of decomposition layers and penalty coefficients of VMD are optimized by GWO with improved fitness function, secondly, use VMD with optimal parameters for operation data and the intrinsic mode function components are screened out as the original data trend term, and finally the LSTM is constructed by using the running parameter trend term as the input feature vector of the training set which used for LSTM, and the LSTM hyperparameters are obtained by the northern goshawk algorithm. Verified by case studies, this method effectively improves the scale and accuracy of LSTM for the prediction of thermal parameter operation trend compared with LSTM directly used as the training set of the original data.

国家自然科学基金项目（面上项目，重点项目，重大项目）