Smart City Gnosys

Smart city article details

Title	Hp-Mobilenetv3: A Lightweight Neural Network For Non-Intrusive Load Monitoring
ID_Doc	29531
Authors	Li S.; Zhao S.; Gao P.; Ge S.; Huang R.; Shu L.
Year	2025
Published	Engineering Research Express, 7, 1
DOI	http://dx.doi.org/10.1088/2631-8695/ada66a
Abstract	Non-intrusive load monitoring (NILM) constitutes an extremely promising technology within the realm of intelligent energy management. Its application possesses the potential to accelerate the construction of smart cities and power ecology, encourage consumers to adjust their electricity consumption behavior, and reduce carbon emissions. Despite the fact that recent deep learning neural network models have demonstrated excellent performance in NILM tasks, these models often require a large amount of computation and sufficient hardware memory to handle the electricity parameters from user edge devices as well as ensure the security of communication data. This makes it difficult to use them on edge devices with limited hardware resources. To overcome these limitations, this study proposes a network architecture named HP-MobileNetV3, which significantly reduces the model size and computational requirements through hierarchical pointwise convolution and network structure pruning, enabling its deployment on edge devices. Experimental results, conducted on the publicly available PLAID dataset, demonstrate that HP-MobileNetV3 achieves a reduction of 96.74% in the number of parameters, 83.51% in computational load, and 35.18% in inference time, while maintaining performance, thus facilitating the practical application of NILM technology. © 2025 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.
Author Keywords	deep learning; Edge NILM; feature fusion; lightweight NILM; pointwise convolution; residual block

Similar Articles

Id	Similarity	Authors	Title	Published
8906	0.891	Fazio P.; Mehic M.; Mannone M.; Astorino F.; Voznak M.	An Overview Of A New Statistical Non-Intrusive Load Monitoring (Nilm) Analysis And Recognition Approach For Domestic Environments: Denardo	2024 IEEE International Workshop on Metrology for Living Environment, MetroLivEnv 2024 - Proceedings (2024)
39732	0.88	Shi Y.; Li W.; Chang X.; Yang T.; Sun Y.; Zomaya A.Y.	On Enabling Collaborative Non-Intrusive Load Monitoring For Sustainable Smart Cities	Scientific Reports, 13, 1 (2023)
38055	0.867	Mughal A.H.; Tahir A.; Javed F.	Mtopsoredc: M Tops Knn For Online Reinforced Electric Device Classification	HONET 2020 - IEEE 17th International Conference on Smart Communities: Improving Quality of Life using ICT, IoT and AI (2020)
35428	0.865	Fazio P.; Mehic M.; Voznak M.	Load Monitoring And Appliance Recognition Using An Inexpensive, Low-Frequency, Data-To-Image, Neural Network, And Network Mobility Approach For Domestic Iot Systems	IEEE Internet of Things Journal, 11, 8 (2024)
34947	0.865	He N.; Liu D.; Zhang Z.; Lin Z.; Zhao T.; Xu Y.	Learning-Based Non-Intrusive Electric Load Monitoring For Smart Energy Management	Sensors, 24, 10 (2024)
44869	0.861	Jiang Y.; Liu M.; Li J.; Zhang J.	Reinforced Mcts For Non-Intrusive Online Load Identification Based On Cognitive Green Computing In Smart Grid	Mathematical Biosciences and Engineering, 19, 11 (2022)
7647	0.859	Kommey B.; Tamakloe E.; Kponyo J.J.; Tchao E.T.; Agbemenu A.S.; Nunoo-Mensah H.	An Artificial Intelligence-Based Non-Intrusive Load Monitoring Of Energy Consumption In An Electrical Energy System Using A Modified K-Nearest Neighbour Algorithm	IET Smart Cities, 6, 3 (2024)
26320	0.856	Alzamil I.	Federated Deep Learning For Scalable And Explainable Load Forecasting In Privacy-Conscious Smart Cities	IEEE Access (2025)
7417	0.856	Sunder R.; R S.; Paul V.; Punia S.K.; Konduri B.; Nabilal K.V.; Lilhore U.K.; Lohani T.K.; Ghith E.; Tlija M.	An Advanced Hybrid Deep Learning Model For Accurate Energy Load Prediction In Smart Building	Energy Exploration and Exploitation, 42, 6 (2024)
12172	0.85	So D.; Oh J.; Jeon I.; Moon J.; Lee M.; Rho S.	Bigta-Net: A Hybrid Deep Learning-Based Electrical Energy Forecasting Model For Building Energy Management Systems	Systems, 11, 9 (2023)