154

Optimising energy efficiency enhancing NILM through high-resolution data analytics

C. Rodriguez-Navarro(1), F. Portillo(1), L. Castro-Santos(2), A. Filgueira-Vizoso(3), F. G. Montoya(1) and A. Alcayde(1)

1. Departamento de Ingeniería Escuela Superior de Ingeniería, Universidad de Almería, (Spain)

2. Departamento de Enxeñaría Naval e Industrial, Centro de Investigación en Tecnoloxías Navais e Industriais (CITENI), Escola Politécnica de Enxeñaría de Ferrol, Universidade da Coruña Campus Industrial de Ferrol, (Spain)

3. Departamento de Química Centro de Investigación en Tecnoloxías Navais e Industriais (CITENI), Escola Politécnica de Enxeñaría de Ferrol, Universidade da Coruña, Campus Industrial de Ferrol, Ferrol (Spain)

2026-01-20

This study introduces a novel approach toenhancing energy efficiency by building a high-resolution dataset for Non-Intrusive Load Monitoring, addressing the challenges of monitoring a wide range of devices. To achieve optimal energy efficiency, it is essential to have advanced monitoring of electrical variables. In this context, the second version of openZmeter is presented, which supports up to 4 devices, each capable of 160 measurements, including voltage, current and power harmonics on each channel. To carry out this purpose, the Non-Intrusive Load Monitoring Toolkit is adapted for the new openZmeter v2. The main objective of this study is to offer a new pragmatic approach to generating artificial intelligence models to optimise energy use, analysing the results through an exhaustive experimental analysis under various casuistry and conditions. Numerous comparative studies are provided using classical disaggregation algorithms with different requirements. Conclusively, the research emphasises the transformative potential of artificial intelligence for energy efficiency strategies, offering insights for scholars and practitioners.

[1] G. W. Hart, ‘Nonintrusive Appliance Load Monitoring’, Proceedings of the IEEE, vol. 80, no. 12, 1992, doi: 10.1109/5.192069.

[2] B. Nipun, K. Jack, P. Oliver, D. Haimonti, and K. William, ‘NILMTK: An Open Source Toolkit for Non-intrusive Load Monitoring’, Apr. 2014, doi: 10.1145/2602044.2602051.

[3] R. G. Pablo Salvador, ‘Implementación del algoritmo de análisis de demanda eléctrica por usos finales en las curvas de UVECASE’, 2020. doi: DOI: 10.13140/RG.2.2.27860.22405.

[4] K. Jack, B. Nipun, P. Oliver, D. Haimonti, and K. William, ‘NILMTK v0.2: A non-intrusive load monitoring toolkit for large scale data sets’, in BuildSys 2014 - Proceedings of the 1st ACM Conference on Embedded Systems for Energy-Efficient Buildings, Association for Computing Machinery, Nov. 2014, pp. 182–183. doi: 10.1145/2674061.2675024.

[5] W. Huijuan, Y. Wenrong, C. Tingyu, and Y. Qingxin, ‘An Optimal Load Disaggregation Method Based on Power Consumption Pattern for Low Sampling Data’, doi: 10.3390/su11010251.

[6] C. Kwok Tai, L. Miltiadis D., and V. Anna, ‘Energy sustainability in smart cities: Artificial intelligence, smart monitoring, and optimization of energy consumption’, Energies (Basel), vol. 11, no. 11, Nov. 2018, doi: 10.3390/en11112869.

[7] G. Feixiang, L. Chang, J. Linru, L. Hao, L. J Y, and Y. Bo, ‘Load Disaggregation in Non-Intrusive Load Monitoring Based on Random Forest Optimized by Particle Swarm Optimization’.

[8] Q. Bing, L. Liya, and W. Xin, ‘Low-Rate Non-Intrusive Load Disaggregation with Graph Shift Quadratic Form Constraint’, doi: 10.3390/app8040554.

[9] Y. Quanbo, W. Huijuan, W. Botao, S. Yaodong, and W. Hejia, ‘A fusion load disaggregation method based on clustering algorithm and support vector regression optimization for lowsampling data’, Future Internet, vol. 11, no. 2, 2019, doi: 10.3390/fi11020051.

[10] F. Albert, ‘Algorithms for Energy Disaggregation’, 2016. Accessed: Apr. 21, 2022. [Online]. Available: https://upcommons.upc.edu/bitstream/handle/2117/89937/108724.pdf

[11] J. Lei, L. Jiaming, L. Suhuai, W. Sam, and P. Glenn, ‘Power Load Event Detection and Classification Based on Edge Symbol Analysis and Support Vector Machine’, Applied Computational Intelligence and Soft Computing, vol. 2012, 2012, doi: 10.1155/2012/742461.

[12] B. G, B. C., C. D., F. L., and M. G., ‘Preliminary Sensitivity Analysis of Combinatorial Optimization (CO) for NILM Applications: Effect of the Meter Accuracy; Preliminary Sensitivity Analysis of Combinatorial Optimization (CO) for NILM Applications: Effect of the Meter Accuracy’, 2021, doi: 10.1109/RTSI50628.2021.9597362.

[13] L. Hui, Non-intrusive Load Monitoring, 2020th ed., vol. 1. Central South University Chansha,China: Science press beijing, 2020. Accessed: Apr. 21, 2022. [Online]. Available: https://link.springer.com/book/10.1007/978-981-15-1860-7?noAccess=true

[14] S. Arias, ‘AriasSilva/DEPS_NILM_Dataset: DEPS: Dataset de la Escuela Politénica Superior’. Accessed: Jun. 03, 2022. [Online]. Available: https://github.com/AriasSilva/DEPS_NILM_Dataset

[15] C. Rodriguez-Navarro, A. Alcayde, V. Isanbaev, L. Castro-Santos, A. Filgueira-Vizoso, and F. G. Montoya, ‘DSUALMH- A new high-resolution dataset for NILM’, Renewable Energy and Power Quality Journal, vol. 21, no. 1, pp. 238–243, Jul. 2023, doi: 10.24084/repqj21.286.

[16] C. Rodríguez-Navarro, F. Portillo, F. Martínez, F. Manzano-Agugliaro, and A. Alcayde, ‘Development and Application of an Open Power Meter Suitable for NILM’, Inventions, vol. 9, no. 1, p. 2, Dec. 2023, doi: 10.3390/inventions9010002.

[17] V. Eduardo, A. Alfredo, M. Francisco G., B. Raul, and A.-C. Francisco M., ‘OpenZmeter: An efficient low-cost energy smart meter and power quality analyzer’, Sustainability (Switzerland), vol. 10, no. 11, Nov. 2018, doi: 10.3390/su10114038.