K-Means Kümeleme Yöntemi ile Konutlarda Kullanılan Elektrikli Cihazların Güç Tüketimi ve Harmonik Bileşenlerinden Karakter Analizi

Öz

Sanayileşmenin artması ve teknolojik gelişmeler, elektrik enerjisine olan ihtiyacı ve enerji kalitesine olan etkiyi artırmaktadır. Bu durum, elektriksel yüklerin izlenmesi ve kontrol edilmesi ihtiyacını doğurmaktadır. Konutlarda kullanılan elektrikli cihazların ölçüm ve denetimi amacıyla farklı akıllı ölçüm uygulamaları ve makine öğrenmesi algoritmaları denenmektedir. Bu çalışmada, evsel cihazların, temel güç tüketim parametreleri ve ürettikleri harmonik bileşenler dikkate alınarak her bir cihazın güç tüketim karakterleri incelenmiştir. Ölçümler, K-Means kümeleme algoritması ile analiz edilmiştir. Analiz sonucunda, yeterli sayıda öznitelik dikkate alınması durumunda her bir cihazın güç tüketim karakterlerine ulaşılabileceği gözlemlenmiştir.

Anahtar Kelimeler

• Yük izleme
• Harmonikler
• Kümeleme
• K-Means

Characteristics of Power Consumption and Harmonic Components of Electrical Appliances Used in Residences with K-Means Clustering Method

Öz

Industrialization and technological developments increase the need for electrical energy and the impact on energy quality. Therefore, it is needed to monitoring and controlling of electrical loads. Different intelligent measurement applications and machine learning algorithms are tried for measurement and control of electrical devices used in residences. In this study, the power consumption characteristic of each residential device are measured by considering the basic power consumption parameters and harmonic components. Measurements were analyzed by K-Means clustering algorithm. As a result of the analysis, it is observed that the power consumption characteristics of each device can be achieved if sufficient number of features are taken into account.

Anahtar Kelimeler

• Load monitoring
• Harmonics
• Clustering
• K-Means

Kaynakça

1. Kekezoğlu B, Bozkurt A, Arıkan O, Kocatepe C, Yumurtacı R, Baysal M. Assessment of Power Quality Terms on Energy Distribution Systems: A Case Study of Istanbul. Prz. Elektrotechniczny 2012; 88(11): 157–160.
2. Utley JI, Shorrock LD. Domestic energy fact file 2008. United Kingdom: Department of Energy and Climate Change 2008.
3. Ehrhardt-Martinez K, Donnelly KA. Advanced Metering Initiatives and Residential Feedback Programs: A Meta-Review for Household Electricity-Saving Opportunities. USA: American Council for an Energy-Efficient Economy 2010.
4. Hart GW. Nonintrusive Appliance Load Monitoring. Proceedindg of the IEEE 1992; 80(12): 1870-1891.
5. Zoha A, Gluhak M, Imran MA, Rajasegarar S. Non-intrusive Load Monitoring approaches for disaggregated energy sensing: A survey. Sensors 2012; 12(12): 16838–16866.
6. Selvam MM,Gnanadass R, Padhy NP. Fuzzy based clustering of smart meter data using real power and THD patterns. Energy Procedia 2017; 117: 401–408.
7. Huang SJ, Hsieh CT, Kuo LC, Lin CW, Chang CW, Fang SA. Classification of home appliance electricity consumption using power signature and harmonic features. IEEE PEDS 2011; 5-8 December 2011; Singapore. pp. 596–599.
8. Yang J, Yang Y, Chen J, Fu L, He Z. Determining the harmonic contributions of multiple harmonic sources using data clustering analysis. 2015 International Symposium on Smart Electric Distribution Systems and Technologies (EDST); 8-11 September 2015; Vienna, Austria. pp. 405–411.

9. Kelly J, Knottenbelt W. Neural NILM: Deep neural networks applied to energy disaggregation. Neural and Evolutionary Computing; 4-5 November 2015; Seoul, South Korea. pp. 55–64.
10. Altrabalsi H, Stankovic L, Liao J, Stankovic V. A low-complexity energy disaggregation method: Performance and robustness. 2014 IEEE Symposium on Computational Intelligence Applications in Smart Grid (CIASG); 9-12 December 2014; Orlando, FL, USA. pp. 1–8.
11. Kim HJ, Cho Y, Kim J, Cho J, Kim J. Demonstration of the LVDC distribution system in an island. CIRED - Open Access Proceedings Journal 2017; 2017(1): 2215–2218.
12. Egarter D, Bhuvana VP, Elmenreich W. PALDi: Online load disaggregation via particle filtering. IEEE Transactions on Instrumentation and Measurement 2015; 64(2): 467–477.
13. Machlev R, Levron Y, Beck Y. Modified Cross-Entropy Method for Classification of Events in NILM Systems. IEEE Transactions on Smart Grid 2019; 10(5): 4962–4973.
14. Çakmak Z, Uzgören N, Keçek G. Kümeleme Anali̇zi Tekni̇kleri̇ ile İlleri̇n Kültürel Yapılarına Göre Sınıflandırılması ve Değişimlerin İncelenmesi̇ 2005; 2005(12).
15. Loohach R, Garg K. Effect of Distance Functions on Simple K-means Clustering Algorithm. International Journal of Computer Applications 2012; 49(6): 7–9.
16. Omran MGH, Engelbrecht AP, Salman A. An overview of clustering methods. Intelligent Data Analysis 2007; 11(6): 583–605.
17. Arillaga J, Watson NR. Power System Harmonics. 2nd ed. New York, NY, USA: Wiley, 2003.
18. Çayır A, Yenidoğan I, Dağ H. Konutların günlük elektrik güç tüketimi tahmini için uygun model seçimi. Fırat Üniversitesi Mühendislik Bilimleri Dergisi 2018; 30(3): 15-21.
19. Sun M, Nakoty FM, Liu Q, Liu X, Yang Y, Shen T. Non-Intrusive load monitoring system framework and load disaggregation algorithms: A survey. 2019 International Conference on Advanced Mechatronic Systems (ICAMechS); 26-28 August 2019; Kusatsu, Shiga, Japan. pp. 284-288.
20. Rashid H, Singh P, Stankovic V, Stankovic L. Can non-intrusive load monitoring be used for identifying an appliance’s anomalous behaviour?. Applied Energy 2019; 238, 796-805.