| 32 | 1 | 38 |
| 下载次数 | 被引频次 | 阅读次数 |
空间负荷预测(SLF)能够准确地预测未来负荷发生的时间段、空间位置和规模。地理信息系统(GIS)是一种结合计算机、数据库和信息系统优势的技术,能够对地理位置及其相关属性信息进行存储和处理。从GIS系统入手,通过整合负荷数据,利用机器学习生成对抗网络(GAN)模型对负荷数据集增维,实现动态时间窗地理空间负荷预测。利用所提方法结合某地区实际负荷数据,进行了算例分析和校验,证实方法的有效性。
Abstract:Spatial load forecasting(SLF) can accurately predict the time period, spatial location, and scale of future loads. Geographic information system(GIS) is a technology that combines the advantages of computer technology, databases, and information systems, capable of storing and processing geographic locations and their related attribute information. This paper starts from the GIS system, integrates load data, and uses generative adversarial networks(GAN) to expand the dimensionality of the load dataset, achieving dynamic time window geospatial load forecasting. The paper employs this method in conjunction with actual load data from a certain area for case analysis and verification.
[1] 江家丞.基于GIS地理信息系统的智能城市规划设计分析[J].智慧中国,2023(5):72-73.
[2] 高明.基于GIS的城市配电网规划设计研究[J].中国高新科技,2021(17):75-76.
[3] KOOCHALI A,SCHICHTEL P,DENGEL A,et al.Probabilistic Forecasting of Sensory Data with Generative Adversarial Networks-ForGAN[J].IEEE Access,2019,7:63868-63880.
[4] 宣菊琴,郑洁云,陈波,等.基于改进栈式自编码算法的中期负荷预测研究[J].电子器件,2021,44(5):1190-1197.
[5] RAVIKUMAR G,KHAPARDE S A,PRADEEP Y.CIM Oriented Database for Topology Processing and Integration of Power System Applications[C]//2013 IEEE Power & Energy Society General Meeting.July 21-25,2013,Vancouver,BC,Canada.IEEE,2013:1-5.
[6] FARZADI M,HERIS F N,SHAHIR F M,et al.The Role of the Intelligent Reconfiguration of Distribution Network on Reduction the Energy Not Supplied Costs in the Electricity Market through Case Studies Using Softwares NEPLAN & DIgSILENT Power Factory[C]//2015 9th International Conference on Electrical and Electronics Engineering(ELECO).November 26-28,2015,Bursa,Turkey.IEEE,2015:445-447.
[7] 陈佳瑜,杨剑,施灵.GIS图形异常对10 kV同期线损的影响分析[J].供用电,2021,38(4):64-69.
[8] 唐贤伦,杜一铭,刘雨微,等.基于条件深度卷积生成对抗网络的图像识别方法[J].自动化学报,2018,44(5):855-864.
[9] 麻英杰.基于深度学习的短期电力负荷预测研究[D].呼和浩特:内蒙古大学,2022.
[10] 王守相,陈海文,潘志新,等.采用改进生成式对抗网络的电力系统量测缺失数据重建方法[J].中国电机工程学报,2019,39(1):56-64.
[11] 张一珂,张鹏远,颜永红.基于对抗训练策略的语言模型数据增强技术[J].自动化学报,2018,44(5):891-900.
[12] 李晶晶,张永敏,田桂林,等.基于LSTM神经网络的数据驱动空间负荷预测方法[J].电子设计工程,2022,30(22):154-157.
基本信息:
中图分类号:TP181;TM715
引用信息:
[1]夏爱民,傅彬,国宗,等.基于机器学习的动态时间窗地理空间负荷预测方法[J].微型电脑应用,2025,41(09):88-92.
基金信息:
国家重点研发计划项目(2020YFB2104500)
2025-09-20
2025-09-20