NON-INTRUSIVE LOAD DECOMPOSITION APPROACH BASED ON RFAM-TCN MODEL AND MULTI-MODAL FEATURE OPTIMIZATION

Wu Zhengtao; Li Zichen; Xia Yanghong; Ni Yini

doi:10.19912/j.0254-0096.tynxb.2025-0180

PDF(1948 KB)

Acta Energiae Solaris Sinica ›› 2026, Vol. 47 ›› Issue (6) : 109-119. DOI: 10.19912/j.0254-0096.tynxb.2025-0180

NON-INTRUSIVE LOAD DECOMPOSITION APPROACH BASED ON RFAM-TCN MODEL AND MULTI-MODAL FEATURE OPTIMIZATION

Wu Zhengtao¹, Li Zichen², Xia Yanghong², Ni Yini²

Author information +

History +

Abstract

To achieve accurate load awareness and efficient device control, this study proposes a novel non-intrusive load decomposition （NILD） method by combining multi-modal feature selection techniques with an improved time convolutional network to enhance identification accuracy and efficiency. In this method, the load multi-scale maximum overlapping discrete wavelet transform results are introduced as input features, and the feature is combined with the mRMR algorithm to ensure the effective integration of multi-modal features. At the same time, the adaptive receptive field attention mechanism （RFAM） is added to the TCN method,improving the ability to extract features at different time scales. Finally, the algorithm is validated on datasets such as PLAID. The results show that the proposed method can achieve more accurate and efficient load decomposition with fewer features,which is higher than the regression accuracy of the traditional BiLSTM model of 5.31%,and the detection rate of new energy household equipment such as photovoltaic power has also reached over 96.6%.

Key words

electric load management / wavelet transforms / feature selection / non-intrusive load decomposition / receptive field attention mechanism / temporal convolutional network

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Wu Zhengtao, Li Zichen, Xia Yanghong, Ni Yini. NON-INTRUSIVE LOAD DECOMPOSITION APPROACH BASED ON RFAM-TCN MODEL AND MULTI-MODAL FEATURE OPTIMIZATION[J]. Acta Energiae Solaris Sinica. 2026, 47(6): 109-119 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0180

References

[1] 程祥, 李林芝, 吴浩, 等. 非侵入式负荷监测与分解研究综述[J]. 电网技术, 2016, 40(10): 3108-3117.
CHENG X, LI L Z, WU H, et al.A survey of the research on non-intrusive load monitoring and disaggregation[J]. Power system technology, 2016, 40(10): 3108-3117.
[2] 李利娟, 刘海, 刘红良, 等. 融合外部注意力机制的序列到点非侵入式负荷分解[J]. 上海交通大学学报, 2024, 58(6): 846-854.
LI L J, LIU H, LIU H L, et al.Non-intrusive load disaggregation using sequence-to-point integrating external attention mechanism[J]. Journal of Shanghai Jiao Tong University, 2024, 58(6): 846-854.
[3] ZHANG Y M, YANG G H, MA S D.Non-intrusive load monitoring based on convolutional neural network with differential input[J]. Procedia cirp, 2019, 83: 670-674.
[4] 林顺富, 詹银枫, 李毅, 等. 基于CNN-BiLSTM与DTW的非侵入式住宅负荷监测方法[J]. 电网技术, 2022, 46(5): 1973-1981.
LIN S F, ZHAN Y F, LI Y, et al.Non-intrusive residential load monitoring method based on CNN-BiLSTM and DTW[J]. Power system technology, 2022, 46(5): 1973-1981.
[5] ZHANG Y S, QIAN W Y, YE Y J, et al.A novel non-intrusive load monitoring method based on ResNet-seq2seq networks for energy disaggregation of distributed energy resources integrated with residential houses[J]. Applied energy, 2023, 349: 121703.
[6] YIN L F, GE W.Mobileception-ResNet for transient stability prediction of novel power systems[J]. Energy, 2024, 309: 133163.
[7] ZHOU X X, FENG J R, LI Y.Non-intrusive load decomposition based on CNN-LSTM hybrid deep learning model[J]. Energy reports, 2021, 7: 5762-5771.
[8] HWANG H, KANG S.Nonintrusive load monitoring using an LSTM with feedback structure[J]. IEEE transactions on instrumentation and measurement, 2022, 71: 2509111.
[9] 周亚军. 面向单体非侵入式负荷监测的神经网络数据特征分析[D]. 北京: 华北电力大学, 2023.
ZHOU Y J.Analysis of neural network data characteristics for non-invasive load monitoring of single unit[D]. Beijing: North China Electric Power University, 2023.
[10] SIROJAN T, PHUNG B T, AMBIKAIRAJAH E.Deep neural network based energy disaggregation[C]//2018 IEEE International Conference on Smart Energy Grid Engineering (SEGE). Oshawa, ON, Canada, 2018: 73-77.
[11] 莫浩杰, 彭勇刚, 蔡田田, 等. 一种基于时频特征融合和极限学习机的非侵入式负荷识别方法[J]. 电工电能新技术, 2023, 42(3): 85-96.
MO H J, PENG Y G, CAI T T, et al.Nonintrusive load identification method based on time-frequency feature fusion and extreme learning machine[J]. Advanced technology of electrical engineering and energy, 2023, 42(3): 85-96.
[12] 邓旭晖, 陈中, 杨凯, 等. 基于多任务学习卷积网络的非侵入式负荷监测方法[J]. 电力系统自动化, 2023, 47(8): 189-197.
DENG X H, CHEN Z, YANG K, et al.Non-intrusive load monitoring method based on multi-task learning convolutional network[J]. Automation of electric power systems, 2023, 47(8): 189-197.
[13] 刘仲民, 侯坤福, 高敬更, 等. 基于时间卷积神经网络的非侵入式居民用电负荷分解方法[J]. 电力建设, 2021, 42(3): 97-106.
LIU Z M, HOU K F, GAO J G, et al.Non-intrusive residential electricity load disaggregation based on temporal convolutional neural network[J]. Electric power construction, 2021, 42(3): 97-106.
[14] 李想, 刘宇航, 张琪, 等. 小波检测和特征图谱决策的非侵入电动自行车充电实时监测系统[J]. 电力系统自动化, 2023, 47(19): 177-186.
LI X, LIU Y H, ZHANG Q, et al.Non-intrusive real-time monitoring system for electric bicycle charging based on wavelet detection and feature graph decision[J]. Automation of electric power systems, 2023, 47(19): 177-186.
[15] REN Y H, ZHANG S, WANG J, et al.TSC-MI: a temporal spatial convolution neural network fused with mutual information for motor imagery based EEG classification[C]//Brain Informatics. Cham: Springer, 2021: 474-485.
[16] 王磊, 马佳琪, 韩肖清, 等. 考虑多状态特征的非侵入式负荷识别方法[J]. 电网技术, 2024, 48(11): 4720-4728.
WANG L, MA J Q, HAN X Q, et al.Non-intrusive load monitoring method with multi-state characterization of loads[J]. Power system technology, 2024, 48(11): 4720-4728.
[17] NI Y N, XIA Y H, LI Z C, et al.A non-intrusive identification approach for residential photovoltaic systems using transient features and TCN with attention mechanisms[J]. Sustainability, 2023, 15(20): 14865.
[18] 马乐乐, 孔小兵, 郭磊, 等. 基于最大重叠离散小波变换和深度学习的光伏功率预测[J]. 太阳能学报, 2024, 45(5): 576-583.
MA L L, KONG X B, GUO L, et al.Photovoltaic power forecasting based on maximum overlap discrete wavelet transform and deep learning[J]. Acta energiae solaris sinica, 2024, 45(5): 576-583.
[19] KONONENKO I.Estimating attributes: analysis and extensions of RELIEF[C]//Machine Learning: ECML-94. Berlin, Heidelberg: Springer, 1994: 171-182.
[20] 温廷新, 郭晓赛. 基于特征组合的ECA-TCN光伏发电功率预测模型[J]. 太阳能学报, 2024, 45(12): 94-100.
WEN T X, GUO X S.Eca-tcn photovoltaic power prediction model based on feature combination[J]. Acta energiae solaris sinica, 2024, 45(12): 94-100.
[21] LIANG J, NG S K K, KENDALL G, et al. Load signature study: part I: basic concept, structure, and methodology[J]. IEEE transactions on power delivery, 2010, 25(2): 551-560.
[22] MONTEMURRO A, ISAKSEN J L, HANSEN T, et al.Temporal convolutional networks for automatic ECG features extraction[J]. Journal of electrocardiology, 2019, 57: S106-S107.
[23] 时培明, 郭轩宇, 杜清灿, 等. 基于TCN-BiLSTM-Attention-ESN的光伏功率预测[J]. 太阳能学报, 2024, 45(9): 304-316.
SHI P M, GUO X Y, DU Q C, et al.Photovoltaic power prediction based on TCN-BILSTM-Attention-ESN[J]. Acta energiae solaris sinica, 2024, 45(9): 304-316.
[24] 刘云翔, 马海力, 朱建林, 等. 基于感受野注意力卷积的自动驾驶多任务感知算法[J]. 计算机工程与应用, 2024, 60(20): 133-141.
LIU Y X, MA H L, ZHU J L, et al.Autonomous driving multi-task perception algorithm based on receptive-field attention convolution[J]. Computer engineering and applications, 2024, 60(20): 133-141.