1.广东工业大学;2.广东浩迪创新科技有限公司
国家自然科学基金资助项目（61571140）
1.Guangdong University of Technology;2.Guangdong Haodi Innovation Technology Co . , Ltd
Supported by National Natural Science Foundation of China (61571140)
为了挖掘负荷预测中不同输入特征的重要性，有效处理负荷数据中的线性成分和非线性成分，提高负荷预测的精度，文中提出一种基于TabNet和LSTNet（long and short-term temporal networks，长期和短期时间序列网络）的组合负荷预测模型。首先，通过引入自监督预训练来提高TabNet的预测精度，通过训练得到输入特征的全局重要性和预测结果，然后把重要性高的特征输入到LSTNet训练得出预测结果，最后通过方差-协方差组合方法得出TabNet-LSTNet模型的预测结果。通过仿真分析，与传统的LSTM（Long Short-Term Memory，长短期记忆网络）、极端梯度提升机（Xgboost）、轻量级梯度提升机（Lightgbm）和其它组合模型相比较，文中提出的TabNet-LSTNet模型具有更高的精度。
In order to explore the importance of different input features in load forecasting, effectively handle the linear and nonlinear components in load data, and improve the accuracy of load prediction, a combined load prediction model based on TabNet and LSTNet is proposed in this paper. First, the prediction accuracy of TabNet is improved by introducing self-supervised pre-training, and then the global importance of the input features and the prediction results are obtained by training. Next, the features with high importance are input to LSTNet, which is trained to get the prediction results. Finally, the prediction results of the combined model are derived by the method of variance-covariance combination. Through simulation analysis, the proposed combined model has higher accuracy compared with traditional LSTM, Xgboost, Lightgbm and other combined models.
