基于SVMD的改进BWO-TimesNet短期热负荷预测模型  

Improved BWO-TimesNet Short-term Heat Load Forecasting Model Based on SVMD

作　　者：段沁宇 薛贵军 谭全伟 谢文举 DUAN Qinyu;XUE Guijun;TAN Quanwei;XIE Wenju(School of Electrical Engineering,North China University of Science and Technology,Tangshan Hebei 063200,China;Intelligent Instrument Factory(North China University of Science and Technology),Tangshan Hebei 063000,China)

机构地区：[1]华北理工大学电气工程学院,河北唐山063200 [2]智能仪器厂(华北理工大学),河北唐山063000

出　　处：《广西师范大学学报（自然科学版）》2024年第6期101-116,共16页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金青年项目(61502143);河北省自然科学基金(E2020209121)。

摘　　要：精准高效的热负荷预测对于保障热力系统稳定运行和合理规划热力资源至关重要。为了提升热负荷预测的准确性,本文提出一种基于逐次变分模态分解(successive variational mode decomposition,SVMD)和改进白鲸优化算法(improved Beluga whale optimization,IBWO)的TimesNet短期热负荷预测模型。首先,利用SVMD将原始热负荷数据进行分解,去除噪声后得到若干个平稳且有规律的模态分量;其次,根据每个模态分量的特点选择合适的特征作为输入;然后,引入3种策略来改进白鲸优化算法,从而建立IBWO-TimesNet预测模型;最后,通过算例对模型的预测性能进行详细评估。结果表明:SVMD-IBWO-TimesNet模型的MAE、RMSE和R 2分别为0.647、1.190和99.1%。与其他主流预测模型相比,该模型具有更高的预测精度。同时,在减少训练样本的情况下,SVMD-IBWO-TimesNet模型仍能有效预测热负荷,具有较强的泛化能力。验证了所提出模型的有效性,为热力系统供热负荷的精准调控提供了参考。Accurate and efficient heat load forecasting is very important to ensure the stable operation of thermal system and rational planning of thermal resources.In order to improve the accuracy of heat load forecasting,a TimesNet short-term heat load forecasting model based on successive variational mode decomposition(SVMD)and an improved beluga optimization algorithm(IBWO)is proposed.Firstly,SVMD is used to decompose the original heat load data,and several stable and regular modal components are obtained after removing noise.Secondly,according to the characteristics of each modal component,the appropriate feature is selected as the input.Then,three strategies are introduced to improve the Beluga optimization algorithm,and the IBWO-TimesNet prediction model is established.Finally,the prediction performance of the model is evaluated in detail by an example.The results show that MAE,RMSE and R 2 of SVMD-IBWO-TimesNet model are 0.647,1.190 and 99.1%,respectively.Compared with other mainstream prediction models,this model has higher prediction accuracy.At the same time,the SVMD-IBWO-TimesNet model can still effectively predict the heat load and has strong generalization ability when the training samples are reduced.Therefore,the validity of the proposed model is verified,and the reference is provided for the precise regulation and control of the heating load of the thermal system.

关 键 词：热负荷预测 TimesNet 逐次变分模态分解 白鲸优化算法 

分 类 号：TM715[电气工程—电力系统及自动化]