Implementation of non-intrusive appliances load monitoring (NIALM) on k-nearest neighbors (k-NN) classifier

Amleset Kelati; Hossam Gaber; Juha Plosila; Hannu Tenhunen; Amleset Kelati; Hossam Gaber; Juha Plosila; Hannu Tenhunen

doi:10.3934/ElectrEng.2020.3.326

AIMS Electronics and Electrical Engineering

2020, Volume 4, Issue 3: 326-344. doi: 10.3934/ElectrEng.2020.3.326

Previous Article Next Article

Research article Special Issues

Implementation of non-intrusive appliances load monitoring (NIALM) on k-nearest neighbors (k-NN) classifier

1.
Royal Institute of Technology (KTH), Division of Electronics and Embedded Systems, EECS/ELE, Electrum 229, 16440 Kista, Sweden
2.
Ontario Tech University, Faculty of Energy Systems and Nuclear Science, and Faculty of Engineering and Applied Science, 2000 Simcoe Street North Oshawa, ON L1G0C5, Canada
3.
University of Turku, Department of Future Technologies, FI-20014 Turun yliopisto, Finland

Received: 23 April 2020 Accepted: 08 September 2020 Published: 16 September 2020

Nonintrusive Appliance Load Monitoring (NIALM) is used to analyze individual’s house energy consumption by distinguishing variations in voltage and current of appliances in a household. The method identifies load consumption of each appliance from the aggregated home energy consumption. NIALM will also provide information of load consumptions of each appliance by indirectly detecting the abnormal changes of appliance usage. The proposed NIALM approach is based on features extraction from load consumptions measurements of electrical power signals in order to classify appliance’s state of operation. In this work, we have improved the identification accuracy and the detection of appliances based on their operational state by employing Machine Learning (ML) technique; namely k-nearest neighbor (k-NN) classification algorithm. The dataset used to perform this process is from the publicly available (PLAID) of power, voltage and current signals of appliances from several houses. This is used as benchmark data set. The PLAID dataset is collected and processed for each appliance and our classification results based on k-NN algorithm achieved high accuracy and is able to gain cost-effective solution. In addition, the result shows that k-NN classifier is a proven as an efficient method for NIALM techniques when compared with other proposed different ML options. Based on the used dataset, the average F-score measure obtained using the k-NN classifier is 90%. Possible reasons behind these findings are discussed and areas for further exploration are proposed.
- non-intrusive load monitoring (NIALM),
- smart meter,
- k-nearest neighbor (k-NN),
- appliance classification,
- PLAID dataset,
- V-I trajectory
Citation: Amleset Kelati, Hossam Gaber, Juha Plosila, Hannu Tenhunen. Implementation of non-intrusive appliances load monitoring (NIALM) on k-nearest neighbors (k-NN) classifier[J]. AIMS Electronics and Electrical Engineering, 2020, 4(3): 326-344. doi: 10.3934/ElectrEng.2020.3.326

Related Papers:

Abstract

Nonintrusive Appliance Load Monitoring (NIALM) is used to analyze individual’s house energy consumption by distinguishing variations in voltage and current of appliances in a household. The method identifies load consumption of each appliance from the aggregated home energy consumption. NIALM will also provide information of load consumptions of each appliance by indirectly detecting the abnormal changes of appliance usage. The proposed NIALM approach is based on features extraction from load consumptions measurements of electrical power signals in order to classify appliance’s state of operation. In this work, we have improved the identification accuracy and the detection of appliances based on their operational state by employing Machine Learning (ML) technique; namely k-nearest neighbor (k-NN) classification algorithm. The dataset used to perform this process is from the publicly available (PLAID) of power, voltage and current signals of appliances from several houses. This is used as benchmark data set. The PLAID dataset is collected and processed for each appliance and our classification results based on k-NN algorithm achieved high accuracy and is able to gain cost-effective solution. In addition, the result shows that k-NN classifier is a proven as an efficient method for NIALM techniques when compared with other proposed different ML options. Based on the used dataset, the average F-score measure obtained using the k-NN classifier is 90%. Possible reasons behind these findings are discussed and areas for further exploration are proposed.

References

[1]	Jian M, Wu J, Chen J, et al. (2017) IOT base smart home appliances by using Cloud Intelligent Tetris Switch. 19th International Conference on Advanced Communication Technology (ICACT), 589-592.
[2]	Lobaccaro G, Carlucci S, Löfström E (2016) A Review of Systems and Technologies for Smart Homes and Smart Grids. Energies 9: 348. doi: 10.3390/en9050348
[3]	EL Jaouhari S, Jose Palacios-Garcia E, Anvari-Moghaddam A, et al. (2019.) Integrated Management of Energy, Wellbeing and Health in the Next Generation of Smart Homes. Sensors 19: 481.
[4]	Kelati A, Plosila J, Tenhunen H (2018) Analysis of Smart Meter Design for e-Health Monitoring on the Smart Grid System. 8th International Workshop on Integration of Solar Power into Power Systems. Energynautics GmbH.
[5]	Nourhan TM, Piechnick M, Falkenberg J, et al. (2017) Detection of muscle fatigue using wearable (MYO) surface electromyography based control device. 8th International Conference on Information Technology (ICIT), 44-49.
[6]	Kelati A and Tenhunen H (2018) Wearable in Cloud. Proceeding of IEEE/ACM International Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE), 7-8.
[7]	Nk A, Bhat G, Park J, et al. (2019) Sensor-Classifier Co-Optimization for Wearable Human Activity Recognition Applications. Proceeding of IEEE International Conference on Embedded Software and Systems (ICESS), 1-4.
[8]	Chalmers C, Hurst W, Mackay M, et al. (2016) Smart Monitoring: An Intelligent System to Facilitate Health Care across an Ageing Population. The Eighth International Conference on Emerging Networks and Systems Intelligence, 34-39.
[9]	Kelati A, Plosila J and Tenhunen H (2019) Smart Meter Load Profiling for e-Health Monitoring System. IEEE 7th International Conference on Smart Energy Grid Engineering (SEGE), 97-102.
[10]	Chalmers C, Hurst W, Mackay M, et al. (2019) Identifying behavioral changes for health monitoring applications using the advanced metering infrastructure. Behav Inform Technol 38: 1154-1166. doi: 10.1080/0144929X.2019.1574900
[11]	Aftab M and Chau CK (2017) Smart Power Plugs for Efficient Online Classification and Tracking of Appliance Behavior. Proceedings of the 8th Asia-Pacific Workshop on Systems, 1-7.
[12]	Burbano D (2015) Intrusive and Non-Intrusive Load Monitoring (A Survey). Latin American Journal of Computing 2: 45-53.
[13]	Hart GW (1992) Nonintrusive appliance load monitoring. Proceedings of the IEEE 80: 1870-1891. doi: 10.1109/5.192069
[14]	Du L, Yang Y, He D, et al. (2014) Feature Extraction for Load Identification Using Long-Term Operating Waveforms. IEEE T Smart Grid 6: 819-826.
[15]	Alcalá J, Ureña J, Hernández A, et al. (2017) Event-Based Energy Disaggregation Algorithm for Activity Monitoring From a Single-Point Sensor. IEEE T Instrum Meas 66: 2615-2626. doi: 10.1109/TIM.2017.2700987
[16]	Le TTH, Kim H (2018) Non-Intrusive Load Monitoring Based on Novel Transient Signal in Household Appliances with Low Sampling Rate. Energies 11: 3409. doi: 10.3390/en11123409
[17]	Meziane MN, Ravier P, Lamarque G, et al. (2017) High accuracy event detection for Non-Intrusive Load Monitoring. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2452-2456.
[18]	Gao J, Kara EC, Giri S, et al. (2015) A feasibility study of automated plug-load identification from high-frequency measurements. IEEE Global Conference on Signal and Information Processing (GlobalSIP), 220-224.
[19]	Wójcik A, Winiecki W, Łukaszewski R, et al. (2019) Analysis of Transient State Signatures in Electrical Household Appliances. 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS) 2: 639-644.
[20]	Kang S and Yoon JW (2016) Classification of home appliance by using Probabilistic KNN with sensor data. Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), 1-5.
[21]	Zhang B, and Srihari SN (2004) Fast k nearest neighbor classification using cluster-based trees. IEEE T Pattern Anal 26: 525-528. doi: 10.1109/TPAMI.2004.1265868
[22]	Kalaivani P and Shunmuganathan KL (2014) An improved K-nearest-neighbor algorithm using genetic algorithm for sentiment classification. IEEE International Conference on Circuits, Power and Computing Technologies, 1647-1651,
[23]	Gao J, Giri S, Kara EC, et al. (2014) PLAID: a public dataset of high-resolution electrical appliance measurements for load identification research: demo abstract. Proceedings of ACM conference on embedded systems for energy-efficient buildings, 198-199.
[24]	Medico R, De Baets L, Gao J, et al. (2020) A voltage and current measurement dataset for plug load appliance identification in households. Sci Data 7: 1-10.
[25]	Dong M, Meira PC, Xu W, et al. (2014) An event window based load monitoring technique for smart meters. IEEE T Smart Grid 3: 787-796.
[26]	Kelati A, Dhaou IB, Kondoro A, et al. (2019) IoT based Appliances Identification Techniques with Fog Computing for e-Health. Proceedings of IEEE IST-Africa Week Conference (IST-Africa), 1-11.
[27]	Yang CC, Soh CS, Yap VV (2014) Comparative Study of Event Detection Methods for Non-intrusive Appliance Load Monitoring. Energy Procedia 61: 1840-1843. doi: 10.1016/j.egypro.2014.12.225
[28]	Barsim KS; Mauch L, Yang B (2016) Neural Network Ensembles to Real-time Identification of Plug-level Appliance Measurements. Proceedings of the 3rd International Workshop on Non-Intrusive Load Monitoring.
[29]	Norford LK, Leeb SB (1996) Non-intrusive electrical load monitoring in commercial buildings based on steady-state and transient load detection algorithm. Energy Buildings 24: 51-64. doi: 10.1016/0378-7788(95)00958-2
[30]	Chang HH, Lin CL, Yang HT (2008) Load Recognition for Different Loads with the Same Real Power and Reactive Power in a Nonintrusive Load-morning System. 12th International Conference on Computer Supported Cooperative Work in Design, CSCWD, 1122-1127.
[31]	Hoyo-Montaño JA, León-Ortega N, Valencia-Palomo G, et al. (2018) Non-Intrusive Electric Load identification using Wavelet Transform. Ingeniería e Investigación 38: 42-51.
[32]	Zoha A, Gluhak A, Iman MA, et al. (2012) Non-Intrusive Load Monitoring Approaches for Disaggregated Energy Sensing: Survey. Sensors 12: 16838-16866. doi: 10.3390/s121216838
[33]	Zhang B, Srihari SN (2004) Fast k nearest neighbor classification using cluster-based trees. IEEE T Pattern Anal 26: 525-528. doi: 10.1109/TPAMI.2004.1265868
[34]	Dalianis H (2018) Evaluation Metrics and Evaluation. Clinical Text Mining 45-53.
[35]	Pedregosa F, Varoquaux G, Gramfort A, et al. (2011) Scikit-learn: Machine Learning in Python. J Mach Learn Res 12: 2825-2830.
[36]	Friedman J, Hastie T, Tibshirani R, et al. (2001) The Elements of Statistical Learning. Springer series in statistics 1.
[37]	Mitchell TM (1997) Machine Learning. McGraw-Hill, Inc.
[38]	Solutions to "Pattern Classification" by Duda et al. (2018) Available from: https://tommyodland.com/files/edu/duda_solutions.pdf.
[39]	Dudani SA (1976) The distance-weighted k-nearest-neighbor rule. IEEE T Syst Man Cy 4: 325-327.
[40]	Thanh Noi P, Kappas M (2018) Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery. Sensors 18: 18.
[41]	De Baets L, Ruyssinck J, Develder C, et al. (2018) Appliance classification using VI trajectories and convolutional neural networks. Energy Buildings 158: 32-36. doi: 10.1016/j.enbuild.2017.09.087

Reader Comments

Your name:*

Email:*
© 2020 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)