TELEWORKING MONITORING SYSTEM USING NILM AND K-NN ALGORITHMS: A STRATEGY FOR SUSTAINABLE SMART CITIES

Chuan Choong Yang; ADRIANA NOH; SITI NOORJANNAH IBRAHIM; ANI LIZA ASNAWI; NOR FADHILLAH MOHAMED AZMIN

doi:10.31436/ijiok.v1i2.16

Authors

Chuan Choong Yang International Islamic University Malaysia
Adriana Noh International Islamic University Malaysia
S. Noorjannah Ibrahim International Islamic University Malaysia
Ani Liza Asnaw International Islamic University Malaysia
N.F. Azmin International Islamic University Malaysia

DOI:

https://doi.org/10.31436/ijiok.v1i2.16

Keywords:

Non-Intrusive Load Monitoring (NILM), K-Nearest Neighbors (k-NN), Teleworking, Sustainability, Smart Cities

Abstract

Working from home or teleworking has become a common practice for most office employees during certain special situations such as pandemic. One of the challenges faced by employers, however, is monitoring workers who are working from home. Webcam, live video feed, or mobile phone tracking deemed to be intrusive. Therefore, in this work, a non-intrusive monitoring approach is used to effectively help employers to keep track of teleworking employees through specific electrical appliances operating condition while maintaining users’ privacies. This strategy uses non-intrusive load monitoring (NILM) approach to recognize four electrical appliances’ switching events used during teleworking measured from a single power point. Together with an event classification method known as K-Nearest Neighbor (k-NN) algorithm, the teleworking event and duration can be identified. The results were presented using classification metrics that consist of confusion matrix and accuracy score. An accuracy of up to 62% has been achieved for the classifier. It is observed that the similarity of appliances’ power usage affects the model accuracy and confusion matrix is constructed to help identify the number of events that are correctly classified as well as wrongly classified. Results from NILM and k-NN strategy can be implemented in the smart city towards sustainability to create a sustainable and employees well-being. It is also useful for an organization to evaluate an employee’s performance who opt for teleworking.

ABSTRAK: Bekerja dari rumah telah menjadi amalan biasa bagi kebanyakan pekerja-pekerja pejabat semasa situasi khas tertentu seperti wabak penyakit. Salah satu cabaran yang dihadapi oleh para majikan, adalah memantau para pekerja yang bekerja dari rumah. Kamera web, suapan video langsung atau penjejakan telefon mudah alih adalah dianggap mengganggu privasi. Oleh itu, dalam kajian ini, pendekatan pemantauan tidak mengganggu privasi digunakan untuk membantu para majikan dengan berkesan menjejak para pekerja yang bekerja dari rumah melalui keadaan operasi peralatan-peralatan elektrik tertentu sambil mengekalkan privasi pengguna. Strategi ini menggunakan pendekatan pemantauan beban elektrik tanpa gangguan (NILM) untuk mengenali empat situasi pensuisan peralatan-peralatan elektrik yang digunakan semasa bekerja dari rumah diukur dari satu titik kuasa. Bersama-sama dengan kaedah-kaedah pengkelasan situation yang dikenali sebagai algoritma K-Nearest Neighbor (k-NN), acara bekerja dari rumah dan tempoh boleh dikenal pasti. Keputusan telah dibentangkan menggunakan metrik klasifikasi yang terdiri daripada matriks kekeliruan dan skor ketepatan. Ketepatan sehingga 62% telah dicapai untuk pengkelasan. Adalah diperhatikan bahawa persamaan penggunaan kuasa peralatan-peralatan elektrik mempengaruhi ketepatan model dan matriks kekeliruan dibina untuk membantu mengenal pasti bilangan peristiwa yang dikelaskan dengan betul serta dikelaskan secara salah. Hasil daripada strategi NILM dan k-NN boleh dilaksanakan di bandar pintar ke arah kemampanan untuk mewujudkan kesejahteraan para pekerja dan mampan. Ia juga berguna untuk organisasi menilai prestasi para pekerja yang memilih untuk bekerja dari rumah.

References

Toli AM and Murtagh N (2020) The Concept of Sustainability in Smart City Definitions. Front. Built Environ. 6:77. doi: 10.3389/fbuil.2020.00077 DOI: https://doi.org/10.3389/fbuil.2020.00077

Hafez F.S., et al. (2023) Energy efficiency in sustainable buildings: A systematic review with taxonomy, challenges, motivations, methodological aspects, recommendations, and pathways for future research. Energy Strategy Rev., 45. DOI: https://doi.org/10.1016/j.esr.2022.101013

H. -H. Chang, C. -C. Yang and W. -J. Lee, (2021), Fault Location Identification in Power Transmission Networks: Using Novel Nonintrusive Fault-Monitoring Systems," in IEEE Industry Applications Magazine, vol. 27, no. 2, pp. 76-89, doi: 10.1109/MIAS.2020.3024493. DOI: https://doi.org/10.1109/MIAS.2020.3024493

D. H. Green, D. W. Quinn, S. Madden, P. A. Lindahl and S. B. Leeb, (2022), Nonintrusive Measurements for Detecting Progressive Equipment Faults, IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1-12, Art no. 3518112, doi: 10.1109/TIM.2022.3193178. DOI: https://doi.org/10.1109/TIM.2022.3193178

Lee, P., Yang, C. C. and Kwong, W. C. (2023) “An Exploratory Study on Students’ Electricity Consumption in RCE Gombak For Sustainable Communities”, Malaysian Journal of Social Sciences and Humanities (MJSSH), 8(3), p. e002170. doi: 10.47405/mjssh.v8i3.2170. DOI: https://doi.org/10.47405/mjssh.v8i3.2170

R. Gopinath, Mukesh Kumar, C. Prakash Chandra Joshua, Kota Srinivas, (2020) Energy management using non-intrusive load monitoring techniques – State-of-the-art and future research directions, Sustainable Cities and Society, Volume 62, 102411. doi.org/10.1016/j.scs.2020.102411. DOI: https://doi.org/10.1016/j.scs.2020.102411

Shi, Y., Li, W., Chang, X. et al. (2023) On enabling collaborative non-intrusive load monitoring for sustainable smart cities. Sci Rep 13, 6569. https://doi.org/10.1038/s41598-023-33131-0 DOI: https://doi.org/10.1038/s41598-023-33131-0

Yassine Himeur, Mariam Elnour, Fodil Fadli, Nader Meskin, Ioan Petri, Yacine Rezgui, Faycal Bensaali, Abbes Amira, (2022) Next-generation energy systems for sustainable smart cities: Roles of transfer learning, Sustainable Cities and Society, Volume 85, 104059. doi.org/10.1016/j.scs.2022.104059. DOI: https://doi.org/10.1016/j.scs.2022.104059

Reinhardt, A., & Klemenjak, C. (2020). Device-Free User Activity Detection using Non-Intrusive Load Monitoring: A Case Study. Proceedings of the 2nd ACM Workshop on Device-Free Human Sensing. DOI: https://doi.org/10.1145/3427772.3429391

Hernández, Á., Ruano, A., Ureña, J., Ruano, M. G., & Garcia, J. J. (2019). Applications of NILM techniques to energy management and assisted living. IFAC PapersOnLine, 52(11), 164-171 DOI: https://doi.org/10.1016/j.ifacol.2019.09.135

Himeur, Y., Alsalemi, A., Bensaali, F., & Amira, A. (2021). Smart non-intrusive appliance identification using a novel local power histogramming descriptor with an improved k-nearest neighbors‘ classifier, Sustainable Cities and Society, 67, 102764. DOI: https://doi.org/10.1016/j.scs.2021.102764

Baets, L.D., Develder, C., Dhaene, T., & Deschrijver, D. (2019). Detection of unidentified appliances in non-intrusive load monitoring using siamese neural networks. International Journal of Electrical Power & Energy Systems. DOI: https://doi.org/10.1016/j.ijepes.2018.07.026

Azizi, E., Beheshti, M. T., & Bolouki, S. (2021). Event matching classification method for non-intrusive load monitoring. Sustainability, 13(2), 693. DOI: https://doi.org/10.3390/su13020693

Akpinar, M. Fatih Adak and G. Guvenc, (2021) SVM-based anomaly detection in remote working: Intelligent software SmartRadar", Applied Soft Computing, 109. 107457, 10.1016/j.asoc.2021.107457. DOI: https://doi.org/10.1016/j.asoc.2021.107457

Yang, C. C., Soh, C. S., & Yap, V. V. (2017). A systematic approach in appliance disaggregation using K-nearest neighbours and naive Bayes classifiers for Energy Efficiency. Energy Efficiency, 11, 239–259. DOI: https://doi.org/10.1007/s12053-017-9561-0

Yang, C.C., Soh, C.S. & Yap, V.V. (2019) A systematic approach in load disaggregation utilizing a multi-stage classification algorithm for consumer electrical appliances classification. Front. Energy 13, 386–398. https://doi.org/10.1007/s11708-017-0497-z DOI: https://doi.org/10.1007/s11708-017-0497-z

Linge, N., Liu, X., Liu, Q., & Lu, M. (2019). Non-intrusive load monitoring and its challenges in a NILM system framework. International Journal of High Performance Computing and Networking, 14(1), 102. DOI: https://doi.org/10.1504/IJHPCN.2019.10021110

Kim, J., Le, T.-T.-H., & Kim, H. (2017). Nonintrusive load monitoring based on Advanced Deep Learning and novel signature. Computational Intelligence and Neuroscience, 2017, 1–22. DOI: https://doi.org/10.1155/2017/4216281

Zhou, Y., Li, F., Liu, L., Wang, T., Cheng, Z., Li, R., & Gao, J. (2022). Non-intrusive load monitoring method based on the time-segmented state probability. Energy Reports, 8, 1418-1423. DOI: https://doi.org/10.1016/j.egyr.2022.02.021

Wilhelm, S., & Kasbauer, J. (2021). Exploiting Smart Meter Power Consumption Measurements for Human Activity Recognition (HAR) with a Motif-Detection-Based Non-Intrusive Load Monitoring (NILM) Approach. Sensors (Basel,Switzerland), 21(23), 8036. DOI: https://doi.org/10.3390/s21238036

Rashid, Haroon & Singh, Pushpendra & Stankovic, Vladimir & Stankovic, Lina, (2019). "Can non-intrusive load monitoring be used for identifying an appliance’s anomalous behaviour?," Applied Energy, Elsevier, vol. 238(C), pages 796-805. doi.org/10.1016/j.apenergy.2019.01.061. DOI: https://doi.org/10.1016/j.apenergy.2019.01.061

Muhidin, Faizatul Huda, Yang, Chuan Choong, Mohd Hamzi, Muhammad Ikmal Hamzi, Ibrahim, Siti Noorjannah, Brunzell, Lena and Lee, Pei May, (2022). "Sustainable energy use and Consumption feedback through an Internet of Things (IOT) Based Mobile Application for Energy Monitoring System – A Case Study" 2nd Global CleanUp Congress on 20-21 September 2022, Kuala Lumpur Malaysia.