Day-ahead electricity price prediction algorithm based on <italic style="font-style: italic">K</italic>NN-RF-VMD-CNN-BiLSTM

YU Lisu; DING Yao; XIONG Yongkang; LI Zipeng; ZHANG Wuxiong

doi:10.11959/j.issn.2096-3750.2025.00489

您当前的位置：

首页 >

文章列表页 >

Day-ahead electricity price prediction algorithm based on KNN-RF-VMD-CNN-BiLSTM

Theory and Technology | 更新时间：2026-01-04

- Day-ahead electricity price prediction algorithm based on KNN-RF-VMD-CNN-BiLSTM
- Chinese Journal on Internet of Things Vol. 9, Issue 4, Pages: 172-183(2025)
- 作者机构：
  
  1.南昌大学信息工程学院，江西南昌 330031
  2.先进信号处理与智能通信江西省重点实验室，江西南昌 330031
  3.中国科学院上海微系统与信息技术研究所，上海 200050
- 作者简介：
- 基金信息：
  
  The National Natural Science Foundation of China(62161024;52267016);Jiangxi Provincial Talent Project for Academic and Technical Leaders of Major Disciplines(20232BCJ23085);Jiangxi Provincial Natural Science Foundation(20224BAB212002;20242BAB20051;20252BAC240666;20252BAC240026);Jiangxi Province Science and Technology Development Programme(20242BCC32016)
- DOI：10.11959/j.issn.2096-3750.2025.00489
  CLC：
- Received：28 May 2025，
  
  Revised：2025-11-14，
  
  Published：10 December 2025
- 稿件说明：
移动端阅览
余礼苏,丁垚,熊永康等.基于KNN-RF-VMD-CNN-BiLSTM的日前电价预测算法[J].物联网学报,2025,09(04):172-183.

YU Lisu,DING Yao,XIONG Yongkang,et al.Day-ahead electricity price prediction algorithm based on KNN-RF-VMD-CNN-BiLSTM[J].Chinese Journal on Internet of Things,2025,09(04):172-183.
余礼苏,丁垚,熊永康等.基于KNN-RF-VMD-CNN-BiLSTM的日前电价预测算法[J].物联网学报,2025,09(04):172-183. DOI： 10.11959/j.issn.2096-3750.2025.00489.

YU Lisu,DING Yao,XIONG Yongkang,et al.Day-ahead electricity price prediction algorithm based on KNN-RF-VMD-CNN-BiLSTM[J].Chinese Journal on Internet of Things,2025,09(04):172-183. DOI： 10.11959/j.issn.2096-3750.2025.00489.

摘要

当前国内电力市场改革推进，市场主体须掌握电价变化趋势，以灵活调整生产计划与电力采购策略，因此对电价的准确预测需求日益增长。针对电价实际预测中存在的问题，如数据缺失、标错等数据异常导致的模型训练不平滑，首先，设计了

-近邻算法（

-nearest neighbors）-随机森林（RF

random forest）算法捕捉全局特征，精准识别并替换异常数据点；其次，通过变分模态分解（VMD

variational mode decomposition）将电价数据分解为多个子模态；最后，运用卷积神经网络（CNN

convolutional neural network）-双向长短期记忆（BiLSTM

bi-directional long short-term memory）网络组合模型进行预测，并得到最终的日前电价预测结果。经仿真验证，该组合电价预测算法相较于基础模型，在平均绝对误差（MAE

mean absolute error）、均方误差（MSE

mean square error）、均方根误差（RMSE

root mean square error）和平均绝对百分比误差（MAPE

mean absolute percentage error）的指标分别相对提升了15.8%、13.6%、1.54%和32.4%，且单个轮次推断时间在秒级内。该算法有效地兼顾了预测效率与精度。

Abstract

As the domestic electricity market reform advanced in China

market participants are required to understand the electricity price trends to flexibly adjust production plans and procurement strategies. Thus

the demand for accurate electricity price forecasting increases steadily. In actual forecasting

abnormal electricity price data

including missing values and mislabeling

cause non-smooth model training. In response to these problems

the

-nearest neighbors (

NN)- random forest (RF) method was first applied. This method

capable of capturing global features

was used to accurately identify and replace abnormal data points. Then

variational mode decomposition (VMD) was employed to decompose electricity price data into multiple sub-modes. Finally

a convolutional neural networks (CNN)-bi-directional long short-term memory (BiLSTM) combined model was utilized for forecasting

and the results were integrated to obtain the final day-ahead electricity price forecast. The Simulation results show that

compared with the basic model

the combined

NN-RF-VMD-CNN-BiLSTM electricity price forecasting algorithm achieves relative improvements of 15.8%

13.6%

1.54%

and 32.4% in mean absolute error (MAE)

mean square error (MSE)

root mean square error (RMSE)

and mean absolute percentage error (MAPE) indicators

respectively. Moreover

the inference time for a single epoch is within seconds. This algorithm effectively balances the forecasting efficiency and accuracy.

关键词

Keywords

references

陈国平 , 梁志峰 , 董昱 . 基于能源转型的中国特色电力市场建设的分析与思考 [J ] . 中国电机工程学报 , 2020 , 40 ( 2 ): 369 - 379 .

CHEN G P , LIANG Z F , DONG Y . Analysis and reflection on the marketization construction of electric power with Chinese charac-teristics based on energy transformation [J ] . Proceedings of the CSEE , 2020 , 40 ( 2 ): 369 - 379 .

YUCEKAYA A . Electricity trading for coal-fired power plants in Turkish power market considering uncertainty in spot, derivatives and bilateral contract market [J ] . Renewable and Sustainable Energy Reviews , 2022 , 159 : 112189 .

WANG K K , YU M , NIU D X , et al . Short-term electricity price fore-casting based on similarity day screening, two-layer decomposition technique and Bi-LSTM neural network [J ] . Applied Soft Computing , 2023 , 136 : 110018 .

AFANASYEV D O , FEDOROVA E A . On the impact of outlier filtering on the electricity price forecasting accuracy [J ] . Applied Energy , 2019 , 236 : 196 - 210 .

FRONDEL M , KAEDING M , SOMMER S . Market premia for renew-ables in Germany: The effect on electricity prices [J ] . Energy Eco-nomics , 2022 , 109 : 105874 .

LATTARI F , RUCCI A , MATTEUCCI M . A deep learning approach for change points detection in InSAR time series [J ] . IEEE Transactions on Geoscience and Remote Sensing , 2022 , 60 : 5223916 .

YOU D D , SHEN X C , LIU G J , et al . Signal anomaly identification strategy based on Bayesian inference for nuclear power machinery [J ] . Mechanical Systems and Signal Processing , 2021 , 161 : 107967 .

TABASSUM T , TOKER O , KHALGHANI M R . Cyber–physical anomaly detection for inverter-based microgrid using autoencoder neural network [J ] . Applied Energy , 2024 , 355 : 122283 .

纪德洋 , 金锋 , 冬雷 , 等 . 基于皮尔逊相关系数的光伏电站数据修复 [J ] . 中国电机工程学报 , 2022 , 42 ( 4 ): 1514 - 1522 .

JI D Y , JIN F , DONG L , et al . Data repairing of photovoltaic power plant based on Pearson correlation coefficient [J ] . Proceedings of the CSEE , 2022 , 42 ( 4 ): 1514 - 1522 .

张承圣 , 邵振国 , 陈飞雄 , 等 . 基于条件深度卷积生成对抗网络的新能源发电场景数据迁移方法 [J ] . 电网技术 , 2022 , 46 ( 6 ): 2182 - 2190 .

ZHANG C S , SHAO Z G , CHEN F X , et al . Renewable power generation data transferring based on conditional deep convolutions generative adversarial network [J ] . Power System Technology , 2022 , 46 ( 6 ): 2182 - 2190 .

林峻 , 严英杰 , 盛戈皞 , 等 . 考虑时间序列关联的变压器在线监测数据清洗 [J ] . 电网技术 , 2017 , 41 ( 11 ): 3733 - 3740 .

LIN J , YAN Y J , SHENG G H , et al . Online monitoring data cleaning of transformer considering time series correlation [J ] . Power System Technology , 2017 , 41 ( 11 ): 3733 - 3740 .

金容鑫 , 娄岱松 , 黄华德 , 等 . 水电机组状态监测数据清洗方法 [J ] . 中国农村水利水电 , 2022 ( 7 ): 187 - 192 .

JIN R X , LOU D S , HUANG H D , et al . Research on cleaning method of condition monitoring data of hydropower units [J ] . China Rural Water and Hydropower , 2022 ( 7 ): 187 - 192 .

闫亚男 , 陈小松 , 范慧敏 , 等 . 水电设备运行状态数据清洗方法研究 [J ] . 水电站机电技术 , 2021 , 44 ( 10 ): 14 - 16 .

YAN Y N , CHEN X S , FAN H M , et al . Study on data cleaning method for running state of hydropower equipment [J ] . Mechanical Electrical Technique of Hydropower Station , 2021 , 44 ( 10 ): 14 - 16 .

ONYEMA E M , DALAL S , ROMERO C A T , et al . Design of intrusion detection system based on cyborg intelligence for security of cloud network traffic of smart cities [J ] . Journal of Cloud Computing , 2022 , 11 ( 1 ): 26 .

闫亚男 , 韩长霖 , 陈小松 , 等 . 融合时间序列特性的水电遥测数据清洗优化框架 [J ] . 水力发电 , 2021 , 47 ( 11 ): 79 - 83, 95 .

YAN Y N , HAN C L , CHEN X S , et al . Improved data cleaning framework of hydropower telemetry dataset based on time series characteristics [J ] . Water Power , 2021 , 47 ( 11 ): 79 - 83, 95 .

刘云鹏 , 王权 , 许自强 , 等 . 基于多层架构的油中溶解气体数据清洗与异常识别方法研究 [J ] . 华北电力大学学报(自然科学版) , 2022 , 49 ( 1 ): 81 - 89 .

LIU Y P , WANG Q , XU Z Q , et al . Research on data cleaning and abnormal recognition method of dissolved gas in oil based on mul-ti-layer architecture [J ] . Journal of North China Electric Power Uni-versity (Natural Science Edition) , 2022 , 49 ( 1 ): 81 - 89 .

廖若愚 , 刘友波 , 沈晓东 , 等 . 基于双向循环插补网络的分布式光伏集群时序数据耦合增强方法 [J ] . 电网技术 , 2024 , 48 ( 7 ): 2784 - 2794 , I0042 - I0048 .

LIAO R Y , LIU Y B , SHEN X D , et al . Time series data coupling en-hancement method of distributed photovoltaic cluster based on bi-directional recurrent imputation network [J ] . Power System Technology , 2024 , 48 ( 7 ): 2784 - 2794 , I0042 - I0048 .

徐昊 , 王永生 , 许志伟 , 等 . 基于生成对抗网络多变量风电时间序列异常值处理 [J ] . 太阳能学报 , 2022 , 43 ( 12 ): 300 - 311 .

XU H , WANG Y S , XU Z W , et al . Outlier processing of multivariable wind power time series based on generative adversarial network [J ] . Acta Energiae Solaris Sinica , 2022 , 43 ( 12 ): 300 - 311 .

彭曙蓉 , 李彬 , 彭君哲 , 等 . 基于CEEMDAN分解的深度学习分位数回归电价预测 [J ] . 电力系统及其自动化学报 , 2020 , 32 ( 4 ): 138 - 144 .

PENG S R , LI B , PENG J Z , et al . Electricity price forecasting based on CEEMDAN decomposition and deep learning quantile regression [J ] . Proceedings of the CSU-EPSA , 2020 , 32 ( 4 ): 138 - 144 .

XU Y Z , HUANG X , ZHENG X D , et al . VMD-ATT-LSTM electricity price prediction based on grey wolf optimization algorithm in electricity markets considering renewable energy [J ] . Renewable Energy , 2024 , 236 : 121408 .

WANG P , XU K Q , DING Z H , et al . An online electricity market price forecasting method via random forest [J ] . IEEE Transactions on In-dustry Applications , 2022 , 58 ( 6 ): 7013 - 7021 .

MUKHERJEE A , ABRAHAM S , SINGH A , et al . From data to cure: a comprehensive exploration of multi-omics data analysis for targeted therapies [J ] . Molecular Biotechnology , 2025 , 67 ( 4 ): 1269 - 1289 .

LI Q , ZHANG L M , ZHANG G H , et al . Simultaneous detection for multiple anomaly data in Internet of energy based on random forest [J ] . Applied Soft Computing , 2023 , 134 : 109993 .

HUANG N E , SHEN Z , LONG S R , et al . The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis [J ] . Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences , 1998 , 454 ( 1971 ): 903 - 995 .

LIU H L , LI Z Y , LI C , et al . Research and application of short-term load forecasting based on CEEMDAN-LSTM modeling [J ] . Energy Reports , 2024 , 12 : 2144 - 2155 .

DRAGOMIRETSKIY K , ZOSSO D . Variational mode decomposi-tion [J ] . IEEE Transactions on Signal Processing , 2014 , 62 ( 3 ): 531 - 544 .

WANG S X , WANG X , WANG S M , et al . Bi-directional long short-term memory method based on attention mechanism and rolling update for short-term load forecasting [J ] . International Journal of Electrical Power Energy Systems , 2019 , 109 : 470 - 479 .

PROBST P , WRIGHT M N , BOULESTEIX A L . Hyperparameters and tuning strategies for random forest [J ] . WIREs Data Mining and Knowledge Discovery , 2019 , 9 ( 3 ): e1301 .

QIN R , WANG Y . ImputeGAN: generative adversarial network for multivariate time series imputation [J ] . Entropy , 2023 , 25 ( 1 ): 137 .

ZHOU H Y , ZHANG S H , PENG J Q , et al . Informer: beyond efficient transformer for long sequence time-series forecasting [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2021 , 35 ( 12 ): 11106 - 11115 .

GUO F , DENG S , ZHENG W , et al . Short-term electricity price forecasting based on the two-layer VMD decomposition technique and SSA-LSTM [J ] . Energies , 2022 , 15 ( 22 ): 8445 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on time series generation and anomaly detection model based on variational autoencoder

Related Author

Guo Changlin

Zhou Zaiyan

Liu Chunxiu

Li Wanbin

Liu Yimin

Li Longtan

Jin Guiyue

Xue Zhiwei

Related Institution

State Grid Shandong Electric Power Company Dezhou Power Supply Company

⁰

Day-ahead electricity price prediction algorithm based on KNN-RF-VMD-CNN-BiLSTM

DOI：10.11959/j.issn.2096-3750.2025.00489

摘要

Abstract

关键词

Keywords

references