To address the challenges of multi-type composite disturbance feature coupling and the fragmentation between static and dynamic characteristics in single-image representations for PQD (power quality disturbance) identification in new-type power systems, a novel composite PQD identification method based on dual-modal feature fusion and a collaborative attention-driven architecture was proposed. Time-series signals were first encoded into complementary images using GASF (Gramian angular summation fields) and MTF (Markov transition fields), overcoming the information limitations inherent in single-encoding modes. Subsequently, a dual-branch collaborative attention-driven network was designed to perform parallel feature extraction from the two types of encoded images, achieving complementary enhancement and synergistic integration of static trends and dynamic details through dual-modal feature fusion. Furthermore, a multi-SNR mixed training strategy was introduced to enable the model to learn noise-invariant features, thereby significantly improving its robustness under 20~40 dB noise interference. Experimental results demonstrated that the proposed method attained an average classification accuracy of 97.42% on a test set containing 10 single and 22 composite disturbances under 20 dB noise, confirming its high accuracy and strong robustness even in noisy environments.
CHENG Jiang-zhou, HU Xing-yu, BAO Gang, MENG Jia-lin.
Power Quality Composite Disturbance Identification Method Based on Dual-modal Feature Fusion and Collaborative Attention-driven[J].
Science Technology and Engineering, 2026, 26(6): 2430-2441 DOI:10.12404/j.issn.1671-1815.2503247
ZhangZhigang, KangChongqing. Challenges and prospects of constructing new-type power system under carbon neutrality target[J]. Proceedings of the CSEE, 2022, 42(8): 2806-2819.
WangQianggang, ZhouYiyao, LiaoJianquan, et al. Review on power quality in DC distribution network[J]. Automation of Electric Power Systems, 2023, 47(7): 193-207.
[5]
AmirM, DeshmukhR G, KhalidH M, et al. Energy storage technologies: an integrated survey of developments, global economical/environmental effects, optimal scheduling model, and sustainable adaption policies[J]. Journal of Energy Storage, 2023, 72.DOI:10.1016/J.EST.
LiuDapeng, LuoJiabin, LiuYong, et al. Power quality disturbance classification based on improved multi-channel deep residual network with Wave-ViT[J]. Acta Metrologica Sinica, 2025, 46(5): 629-637.
HeCaijun, LiKaicheng, DongYufei, et al. Recognition of composite power quality disturbance based on knowledge distillation and RP-MobileNetV3[J]. Power System Protection and Control, 2023, 51(14): 75-84.
QianBeiqi, ChenQian, LiZongyuan, et al. Classification method for power quality disturbances based on Markov transition field and multi-head attention mechanism[J]. Power System Technology, 2024, 48(2): 721-733.
[12]
KhetarpalP, NagpalN, Al-NumayM S, et al. Power quality disturbances detection and classification based on deep convolution auto-encoder networks[J]. IEEE Access, 2023, 11: 46026-46038.
HuangJie, LiJianmin, ZhuBingfan, et al. A multi-label learning method for power quality disturbance recognition and classification based on conditional network and knowledge injection[J]. Transactions of China Electrotechnical Society, 2025. DOI:10.19595/j.cnki.1000-6753.tces.242088.
[15]
WangJ, ZhangD, ZhouY. Ensemble deep learning for automated classification of power quality disturbances signals[J]. Electric Power Systems Research, 2022, 213. DOI:10.1016/J.EPSR.2022.108695.
[16]
周凯翔. 基于深度学习的电能质量扰动信号分类方法研究[D]. 长春: 长春工业大学, 2024.
[17]
ZhouKaixiang. Research on classification method of power quality disturbance signals based on deep learning[D]. Changchun: Changchun University of Technology, 2024.
[18]
ZhangL, JiangC, PangA, et al. Super-efficient detector and defense method for adversarial attacks in power quality classification[J]. Applied Energy, 2024.DOI:10.1016/J.APENERGY.2024.122872.
GongZheng, ZouYang, JinTao, et al. Power quality disturbance classification method based on feature fusion parallel optimization model[J]. Proceedings of the CSEE, 2023, 43(3): 1017-1027.
JinXing, ZhouKaixiang, YuHaizhou, et al. Power quality disturbance classification method based on convolutional neural network and Transformer[J]. Science Technology and Engineering, 2024, 24(16): 6726-6733.
HeCaijun, LiKaicheng, YangWangwang, et al. Composite power quality disturbance recognition based on dual-channel GAF and deep residual network[J]. Power System Technology, 2023, 47(1): 369-379.
WangJidong, WangZeping, ZhangDi. Power quality disturbance classification based on recurrence plot and pre-trained transfer learning[J]. Southern Power System Technology, 2025, 19(2): 48-56, 114.
LanMingyang, LiuYulong, JinTao, et al. Identification of composite power quality disturbance types based on visual trajectory circle and ResNet18[J]. Proceedings of the CSEE, 2022, 42(17): 6274-6286.
ZhangXi, ZhengJianyong, MeiFei, et al. An edge-cloud collaborative identification framework for power quality disturbances based on image analysis[J]. Proceedings of the CSEE, 2025, 45(12): 4593-4608.
BiGuihong, BaoTongyu, ChenChenpeng, et al. Composite power quality disturbance recognition based on hybrid decomposition multi-scale time-frequency map and Res-GRU-AT[J]. Power System Protection and Control, 2024, 52(4): 12-25.
WangShouxiang, ZhaoNing, WangTongxun, et al. A lightweight recognition method for power quality disturbances based on trajectory color coding and knowledge distillation[J]. Electric Power Automation Equipment, 2024, 44(12): 85-91.
[35]
LiuZ, LinY, CaoY, et al. Swin transformer: Hierarchical vision transformer using shifted windows[C]// Proceedings of the IEEE/CVF international conference on computer vision. 2021: 10012-10022.
[36]
LiuY, ShaoZ, HoffmannN. Global attention mechanism: retain information to enhance channel-spatial interactions[J]. arxiv Preprint, arxiv:2112.05561, 2021.
FuKuan, WangHongxin, LiuJie, et al. Composite power quality disturbance recognition based on Markov transition field and EfficientNet[J]. Power System and Clean Energy, 2024, 40(4): 74-83.
ShiShuai, ChenZiwen, HuangDongmei, et al. Power quality disturbance recognition algorithm based on MTF visualization and improved DenseNet neural network[J]. Journal of Electric Power Science and Technology, 2024, 39(4): 102-111.
MaWei. Power quality disturbance recognition method based on GAF-MTF and multi-stream convolutional neural network[D]. Qinhuangdao: Yanshan University, 2023.
ChenDongjing, BiGuihong, BaoTongyu, et al. Research on composite power quality disturbance recognition using multi-channel combined time-frequency images[J]. Journal of Yunnan University (Natural Sciences Edition), 2025, 47(3): 443-453.