Load forecasting for integrated energy systems based on CNN-BiLSTM-RF-KDE

doi:10.3969/j.issn.2097-0706.2025.09.007

Abstract

Abstract:

To address the challenges of insufficient multi-source heterogeneous data fusion and uncertainty quantification in load forecasting for integrated energy systems， a hybrid CNN-BiLSTM-RF-KDE model was proposed. The model utilized convolutional neural network（CNN） to extract local features of load data， bidirectional long short-term memory （BiLSTM） to capture bidirectional temporal dependencies， random forest（RF） to handle high-dimensional nonlinear relationships， and kernel density estimation（KDE） to quantify prediction uncertainty. Additionally， an electricity-heat-gas multi-energy flow coupling model was established to analyze the influence of different carbon price intervals on scheduling strategies. Case studies demonstrated that the coefficient of determination（R²） for electrical and heating load forecasting reached 0.93 and 0.96 on the training set， and 0.79 and 0.84 on the test set， respectively. The predicted power generation and heat output of each device closely aligned with the mean trend， indicating that the model provided more accurate load predictions. Based on this data， more reliable analysis and scheduling of integrated energy systems could be achieved.

Key words: integrated energy system, load forecasting, convolutional neural network, bidirectional long short-term memory network, kernel density estimation, random forest, carbon price

CLC Number:

TK01

DOU Xiang, LI Zhuoqun, ZHANG Zhe, WEN Xin, ZHAO Bo, HAN Yan, ZHONG Sheng. Load forecasting for integrated energy systems based on CNN-BiLSTM-RF-KDE[J]. Integrated Intelligent Energy, 2025, 47(9): 60-70.

Figures/Tables 13

Fig.1

Fig.2

Table 1

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Fig.12

References 26

[1]	李菲菲, 崔金栋, 朱增陈, 等. 我国支柱产业碳中和路径的数字化驱动与协同机制研究[J]. 情报科学, 2024, 42(3):64-70.
	LI Feifei, CUI Jindong, ZHU Zengchen, et al. The digital driving and collaborative mechanism of carbon neutrality path in China's pillar industries[J]. Information Science, 2024, 42(3): 64-70.
[2]	陆王琳, 陆启亮, 张志洪. 碳中和背景下综合智慧能源发展趋势[J]. 动力工程学报, 2022, 42(1): 10-18. doi: 10.19805/j.cnki.jcspe.2022.01.002
	LU Wanglin, LU Qiliang, ZHANG Zhihong. An overview of the integrated energy systems' development under the background of carbon neutralization[J]. Journal of Chinese Society of Power Engineering, 2022, 42(1): 10-18. doi: 10.19805/j.cnki.jcspe.2022.01.002
[3]	ABU-RAYASH A, DINCER I. Development of an integrated energy system for smart communities[J]. Energy, 2020, 202:117683.
[4]	NAVESI B R, JADIDOLESLAM M, SHAHRBABAK M Z, et al. Capability of battery-based integrated renewable energy systems in the energy management and flexibility regulation of smart distribution networks considering energy and flexibility markets[J]. Journal of Energy Storage, 2024, 98: 113007.
[5]	TUAN A H, VIET V P, PHUONG X N. Integrating renewable sources into energy system for smart city as a sagacious strategy towards clean and sustainable process[J]. Journal of Cleaner Production, 2021, 305:127161.
[6]	周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017(6): 1229-1251.
	ZHOU Feiyan, JIN Linpeng, DONG Jun. Review of convolutional neural network[J]. Chinese Journal of Computers, 2017(6): 1229-1251.
[7]	杨澜倩, 郭锦敏, 田慧丽, 等. 基于CNN-LSTM-Self attention的园区负荷多尺度预测研究[J]. 综合智慧能源, 2025, 47(2): 79-87. doi: 10.3969/j.issn.2097-0706.2025.02.008
	YANG Lanqian, GUO Jinmin, TIAN Huili, et al. Research on multi-scale load prediction in parks based on CNN-LSTM-Self attention[J]. Integrated Intelligent Energy, 2025, 47(2): 79-87. doi: 10.3969/j.issn.2097-0706.2025.02.008
[8]	许远东. 基于CNN-BILSTM-Attention模型的光伏发电预测研究[J]. 现代工业经济和信息化, 2024, 14(11): 171-172.
	XU Yuandong. Research on photovoltaic power generation prediction based on CNN-BILSTM-Attention modeling[J]. Modern Industrial Economy and Informationization, 2024, 14(11): 171-172.
[9]	ZHANG Y L, SHU X, MA Y X. Urban traffic flow prediction based on the integration of ARIMA and CNN[C]// Proceedings of the Second International Conference on Artificial Intelligence and Communication Technologies(ICAICT), 2024: 299-313.
[10]	HAI N V, NGUYEN D B, QUY T Q, et al. Using opals program system and sparse CNN model in processing and classifying airborne laser scanning data[C]// International Conference on Advances in Information and Communication Technology. Springer, 2025.
[11]	曲景影, 孙显, 高鑫. 基于CNN模型的高分辨率遥感图像目标识别[J]. 国外电子测量技术, 2016, 35(8): 45-50.
	QU Jingying, SUN Xian, GAO Xin. Remote sensing image target recognition based on CNN[J]. Foreign Electronic Measurement Technology, 2016, 35(8): 45-50.
[12]	赵志宏, 赵敬娇, 魏子洋. 基于BiLSTM的滚动轴承故障诊断研究[J]. 振动与冲击, 2021, 40(1): 95-101.
	ZHAO Zhihong, ZHAO Jingjiao, WEI Ziyang. Rolling bearing fault diagnosis based on BiLSTM network[J]. Journal of Vibration and Shock, 2021, 40(1): 95-101.
[13]	吴俊, 程垚, 郝瀚, 等. 基于BERT嵌入BiLSTM-CRF模型的中文专业术语抽取研究[J]. 情报学报, 2020(4): 409-418.
	WU Jun, CHENG Yao, HAO Han, et al. Automatic extraction of Chinese terminology based on BERT embedding and BiLSTM-CRF model[J]. Journal of the China Society for Scientific and Technical Information, 2020(4): 409-418.
[14]	SAFARI A, GHAEMI S. NeuroFuzzyMan: A hybrid neuro-fuzzy BiLSTM stacked ensemble model for financial forecasting and analysis: Dataset case studies on JPMorgan, AMZN and TSLA[J]. Expert Systems with Applications, 2025, 266: 126037.
[15]	MISHRA S, JAIN V, SARAF A Y, et al. Deep neuro-fuzzy system for violence detection[J]. Neurocomputing, 2025, 619: 129007.
[16]	张淑清, 李君, 姜安琦, 等. 基于FPA-VMD和BiLSTM神经网络的新型两阶段短期电力负荷预测[J]. 电网技术, 2022, 46(8): 3269-3279.
	ZHANG Shuqing, LI Jun, JIANG Anqi, et al. A novel two-stage model based on FPA-VMD and BiLSTM neural network for short-term power load forecasting[J]. Power System Technology, 2022, 46(8): 3269-3279.
[17]	钟吴君, 李培强, 涂春鸣. 基于EEMD-CBAM-BiLSTM的牵引负荷超短期预测[J]. 电工技术学报, 2024, 39(21): 6850-6864.
	ZHONG Wujun, LI Peiqiang, TU Chunming. Traction load ultra-short-term forecasting framework based on EEMD-CBAM-BiLSTM[J]. Transactions of China Electrotechnical Society, 2024, 39(21): 6850-6864.
[18]	SAAD S, ABDUL N, RASHID B, et al. Effect of climate change on thermal loads in concrete box girders[J]. Journal of Bridge Engineering, 2025, 30(3):6835.
[19]	ABASS J P, MUTHULINGAM S. Comprehensive assessment of PCM integrated roof for passive building design: A study in energo-economics[J]. Energy & Buildings, 2024, 317:114387.
[20]	李洋, 董红斌. 基于CNN和BiLSTM网络特征融合的文本情感分析[J]. 计算机应用, 2018, 38(11): 3075-3080. doi: 10.11772/j.issn.1001-9081.2018041289
	LI Yang, DONG Hongbin. Text sentiment analysis based on feature fusion of convolution neural network and bidirectional long shortvterm memory network[J]. Journal of Computer Applications, 2018, 38(11): 3075-3080. doi: 10.11772/j.issn.1001-9081.2018041289
[21]	朱凌建, 荀子涵, 王裕鑫, 等. 基于CNN-BiLSTM的短期电力负荷预测[J]. 电网技术, 2021, 45(11): 4532-4539.
	ZHU Lingjian, XUN Zihan, WANG Yuxin, et al. Short-term power load forecasting based on CNN-BiLSTM[J]. Power System Technology, 2021, 45(11): 4532-4539.
[22]	LAWAL B Y, OWOLAWI A P, TU C, et al. The kernel density estimation technique for spatio-temporal distribution and mapping of rain heights over South Africa: The effects on rain-induced attenuation[J]. Atmosphere, 2024, 15(11):1354.
[23]	曾忠平, 王雅丽, 彭浩轩. 基于SOPARC和KDE的游客游憩行为研究: 以武汉东湖绿道为例[J]. 中国园林, 2019, 35(12): 58-62.
	ZENG Zhongping, WANG Yali, PENG Haoxuan. Research on tourist recreation behavior based on SOPARC and KDE: Taking Wuhan East Lake greenway as an example[J]. Chinese Landscape Architecture, 2019, 35(12): 58-62.
[24]	薛海燕, 张旺锋, 马文亚, 等. 基于KDE的渭河流域镇驻地及居民点空间分布格局研究[J]. 安徽农业科学, 2014, 42(34): 12350-12352, 12359.
	XUE Haiyan, ZHANG Wangfeng, MA Wenya, et al. A KDE-based research on spatial distribution pattern of towns and settlements in Wei River basin[J]. Journal of Anhui Agricultural Sciences, 2014, 42(34): 12350-12352, 12359.
[25]	吴潇雨, 和敬涵, 张沛, 等. 基于灰色投影改进随机森林算法的电力系统短期负荷预测[J]. 电力系统自动化, 2015, 39(12): 50-55.
	WU Xiaoyu, HE Jinghan, ZHANG Pei, et al. Power system short-term load forecasting based on improved random forest with grey relation projection[J]. Automation of Electric Power Systems, 2015, 39(12): 50-55.
[26]	赵腾, 王林童, 张焰, 等. 采用互信息与随机森林算法的用户用电关联因素辨识及用电量预测方法[J]. 中国电机工程学报, 2016, 36(3): 604-614.
	ZHAO Teng, WANG Lintong, ZHANG Yan, et al. Relation factor identification of electricity consumption behavior of users and electricity demand forecasting based on mutual information and random forests[J]. Proceedings of the CSEE, 2016, 36(3): 604-614.

参数	数值
CNN卷积层层数	2
第1,2层卷积核大小	3，1
第1,2层卷积步幅	1，1
第1层CNN输出通道大小	16
第2层CNN输出通道大小	32
BiLSTM层数	1
BiLSTM隐藏单元数	64
RF数	100
训练轮数	500
初始学习率	0.001