基于改进变分模态分解与Informer组合模型的风电功率多步预测研究OA
Multi-step Prediction of Wind Power Based on Improved Variational Modal Decomposition and Informer Hybrid Model
保证风电功率预测的准确性是提高风能利用效率、实现电力系统可持续发展的关键工作.因此,该文提出一种基于改进变分模态分解与Informer组合模型的风电功率多步预测模型.首先,采用随机森林模型对风速、风向、压强等原始气象因素进行筛选.其次,通过鹈鹕优化算法改进后的变分模态分解算法对风电功率信号进行分解,从而提高风电序列预测精准性.第三,基于Informer模型对风电功率进行多步预测.最后,通过与其他模型进行对比分析,验证该模型在风电功率多步预测中的优越性.算例结果表明,基于改进变分模态分解与Informer组合模型的风电功率多步预测模型具有良好的预测性能,可为风电功率的预测提供参考.
The accurate predition of wind power is crucial for enhancing the efficiency of wind energy utilization and achieving sustainable development of the power system.In view of this,a multi-step wind power prediction model based on the improved variational modal decomposition(VMD)with Informer is proposed in this paper.Firstly,the original meteorological factors such as wind speed,wind direction,and pressure are filtered using the random forest model.Secondly,the wind power signal is decomposed by the pelican optimization algorithm-improved VMD algorithm to enhance the accuracy of wind power sequence prediction.Thirdly,multi-step wind power prediction is performed using the Informer model.Finally,the superiority of this model in multi-step wind power prediction is verified through multi-dimensional comparison with other models.The case results demonstrate that the wind power multi-step prediction model based on the improved VMD with Informer exhibits excellent prediction performance and can provide reference for wind power prediction.
郭晓鹏;赵琪;张国维
华北电力大学 经济与管理学院,北京市昌平区 102206华北电力大学 新能源电力与低碳发展研究北京市重点实验室,北京市昌平区 102206华北电力大学 经济与管理学院,北京市昌平区 102206
信息技术与安全科学
风电功率预测随机森林鹈鹕优化算法信号分解多步预测
wind power predictionrandom forest(RF)pelican optimization algorithm(POA)signal decompositionmulti-step prediction
《现代电力》 2026 (1)
20-29,10
国家自然科学基金(青年科学基金项目)(72301102).Project Supported by National Natural Science Foundation of China(Young Scientistic Program)(72301102).
评论