摘要
在“双碳”背景下，电力负荷受多重因素耦合影响，从传统的“纯负荷”逐渐演变为具有“负荷”+“电源”双重特性的广义负荷。广义负荷由于其复杂性和不确定性，传统的时间序列预测方法将不再适用。本文从图像处理的角度出发，提出了一种基于模态分解的广义负荷图像化短期预测方法。首先，通过XGBooste、GBDT、RF算法的结果对比，将数据集进行归一化处理和特征筛选。然后利用模态分解将广义负荷数据分解为3组模态，将其作为R、G、B三通道的像素值生成三基色（RGB）图像，并进行图像多样化处理，利用优化后的DenseNet神经网络进行训练和预测。最后，选取基础负荷和风光发电量数据，利用DBSCAN聚类算法得出不同风光渗透率下广义负荷场景的特征曲线，根据所提图像化预测方法进行预测，通过与传统的时间序列预测方法的对比，验证了广义负荷图像化预测方法的可行性。
Generalized load graphical forecasting method based on modal decomposition
基于模态分解的广义负荷图像化预测方法
Lizhen Wu, Peixin Chang, Wei Chen, Tingting Pei
1.College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,P.R.China
2.School of Electrical Data Engineering,University of Technology Sydney,NSW,2007,Australia
收听作者1分钟语音介绍
Abstract
Keywords
Load forecasting;Generalized load;Image processing;DenseNet;Modal decomposition
Fig. 1 Generalized load graphical forecasting method framework
Fig. 2 Feature contribution research based on different algorithms
Fig. 3 Decomposition process of complete ensemble empirical mode decomposition with adaptive noise
Fig. 4 Reciprocal diagram of RGB image and load curve
Fig. 5 DenseNet structure chart
Fig. 6 Dense Block structure chart
Fig. 7 Gaussian diversity processing sample diagram
Fig. 8 Reverse diversification processing sample diagram
Fig. 9 Characteristic cure of base load, wind power, and photovoltaic generation curve
Fig. 10 Characteristic curves of generalized load in scenarios with different permeabilities of renewable energy
Fig. 11 Load data for CEEMDAN decomposition
Fig. 12 Comparison of error indexes obtained with different processing methods
Fig. 13 Forecasting results under different generalized load scenarios
本文引文信息
Wu L Z, Chang P X, Chen W, et al. (2023) Generalized load graphical forecasting method based on modal decomposition, Global Energy Interconnection, 7(2): 166-178
吴丽珍，常培鑫，陈伟等 (2023) 基于模态分解的广义负荷图像化预测方法. 全球能源互联网（英文）, 7(2): 166-178
Biographies
Lizhen Wu
Lizhen Wu received the M.S.degree in the control theory and control engineering from Lanzhou University of Technology,Gansu,China,in 2004,and Ph.D.in control theory and control engineering from Lanzhou University of Technology in 2017.She is studied power systems and automation at the National Active Distribution Network Technology Research Center,Beijing Jiaotong University,Beijing,China,in 2015.Currently,she is an Associate Professor/Master Supervisor at the College of Electrical and Information Engineering,Lanzhou University of Technology,where she teaches courses on power electronics,control theory,and renewable energy systems.Her interests include distributed generation and microgrids,microenergy grid coordination control,power quality control,artificial intelligence and data-driven theory for smart grids,and networked control theory and its application.
Peixin Chang
Peixin Chang was born in Baiyin, China, in 1999. He received his B.Eng. degree in 2022. He is currently a master’s degree candidate at the Lanzhou University of Technology, Gansu, China. He interests include load modeling and control.
Wei Chen
Wei Chen received his M.S. degree in power systems and automation from Xi’an Jiaotong University, Xi’an, China, in 2005, and Ph.D. in control theory and control engineering from Lanzhou University of Technology in 2011. He is now a professor and doctoral supervisor at the College of Electrical and Information Engineering, Lanzhou University of Technology, where he teaches courses on power systems and automation as well as control theory. His interests include smart grids, intelligent control theory and applications, artificial intelligence, power system stability analysis, and power quality control technology.
Tingting Pei
Tingting Pei received her Ph.D. in renewable energy and smart grid from Lanzhou University of Technology, Lanzhou, China, in 2020. She is a lecturer at the College of Electrical and Information Engineering, Lanzhou University of Technology, since 2021. Her current research interests include fault diagnosis, reconfiguration, intelligent operation, and maintenance of photovoltaic power generation systems.
编辑：王彦博
审核：王 伟
郑重声明
根据国家版权局相关规定，纸媒、网站、微博、微信公众号转载、摘编本微信作品，需包含本微信名称、二维码等关键信息，并在文首注明《全球能源互联网》原创。
个人请按本微信原文转发、分享。
分享优秀论文，帮小编点个 “在看” 哦
