# Automatic Identification of Different Types of Consumer Configurations by Using Harmonic Current Measurements

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Impact Factors on Power Quality

#### 2.1. Electrical Environment

#### 2.2. Types of Variation

## 3. Measurement Campaign

- 18 measurements in residential areas,
- 6 measurements in commercial areas,
- 6 measurements in office areas and
- 10 measurements in mixed areas.

^{(h)}, independent of changes in the fundamental current magnitude, I

^{(1)}:

## 4. Harmonic Emission Profiles

#### 4.1. Characteristic Emission Profiles

#### 4.2. Grouping of Emission Profiles

#### 4.2.1. Min-Max Normalization and Clustering of Emission Profiles

#### 4.2.2. Definition of Profile Classes

## 5. Classification of Harmonic Emissions Profiles

#### 5.1. Binary Tree Structure Classifier

#### 5.2. Feature Selection

#### 5.2.1. Features for Classifier SVM-1

#### 5.2.2. Features for Classifier SVM-2

#### 5.2.3. Features for Classifier SVM-3

#### 5.2.4. Features for Classifier SVM-4

#### 5.3. Implementation of Support Vector Machines

#### 5.3.1. Radial Basis Kernel Function

#### 5.3.2. Determination of Optimal Parameters

## 6. Classification of Harmonic Emission Profiles

#### 6.1. Performance Evaluation

#### 6.2. Robustness of Performance

## 7. Measure of Misclassification

## 8. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Santoso, S.; McGranaghan, M.F.; Dugan, R.C.; Beaty, H.W. Electrical Power Systems Quality, 3rd ed.; McGraw-Hill Professional: New York, NY, USA, 2012. [Google Scholar]
- Arrillaga, J.; Watson, N.R. Power System Harmonics; Wiley: Hoboken, NJ, USA, 2003. [Google Scholar]
- Yazdani-Asrami, M.; Mirzaie, M.; Akmal, A.A.S. No-load loss calculation of distribution transformers supplied by nonsinusoidal voltage using three-dimensional finite element analysis. Energy
**2013**, 50, 205–219. [Google Scholar] [CrossRef] - Zavoda, F. The key role of intelligent electronic devices (IED) in advanced Distribution Automation (ADA). In Proceedings of the 2008 China International Conference on Electricity Distribution, Guangzhou, China, 10–13 December 2008; pp. 1–7. [Google Scholar]
- Elphick, S.; Ciufo, P.; Drury, G.; Smith, V.; Perera, S.; Gosbell, V. Large Scale Proactive Power-Quality Monitoring: An Example from Australia. IEEE Trans. Power Deliv.
**2017**, 32, 881–889. [Google Scholar] [CrossRef] [Green Version] - Kilter, J.; Elphick, S.; Meyer, J.; Milanovic, J.V. Guidelines for Power quality monitoring—Results from CIGRE/CIRED JWG C4.112. In Proceedings of the 2014 16th International Conference on Harmonics and Quality of Power (ICHQP), Bucharest, Romania, 25–28 May 2014; pp. 703–707. [Google Scholar] [CrossRef]
- Meyer, J.; Schegner, P.; Eberl, G. Increasing the reliability of indices for power quality assessment in distribution networks. In Proceedings of the 2008 13th International Conference on Harmonics and Quality of Power, Wollongong, NSW, Australia, 28 September–1 October 2008; pp. 1–6. [Google Scholar]
- Domagk, M.; Meyer, J.; Schegner, P. Seasonal variations in long-term measurements of power quality parameters. In Proceedings of the 2015 IEEE Eindhoven PowerTech, Eindhoven, The Netherlands, 29 June–2 July 2015; pp. 1–6. [Google Scholar]
- Domagk, M.; Meyer, J.; Schegner, P. Characterization of public low voltage grids by clustering time series of power quality parameters. In Proceedings of the 12th International Conference, Istanbul, Turkey, 10–14 June 2012; pp. 558–563. [Google Scholar]
- Salles, D.; Jiang, C.; Xu, W.; Freitas, W.; Mazin, H.E. Assessing the Collective Harmonic Impact of Modern Residential Loads—Part I: Methodology. IEEE Trans. Power Deliv.
**2012**, 27, 1937–1946. [Google Scholar] [CrossRef] - Devarapalli, H.P.; Dhanikonda, V.S.S.S.S.; Gunturi, S.B. Non-Intrusive Identification of Load Patterns in Smart Homes Using Percentage Total Harmonic Distortion. Energies
**2020**, 13, 4628. [Google Scholar] [CrossRef] - Meyer, J.; Blanco, A.-M.; Domagk, M.; Schegner, P. Assessment of Prevailing Harmonic Current Emission in Public Low-Voltage Networks. IEEE Trans. Power Deliv.
**2016**, 32, 962–970. [Google Scholar] [CrossRef] - Domagk, M.; Zyabkina, O.; Meyer, J.; Schegner, P. Trend identification in power quality measurements. In Proceedings of the 2015 Australasian Universities Power Engineering Conference (AUPEC), Wollongong, Australia, 7–30 September 2015; pp. 1–6. [Google Scholar]
- Gil-De-Castro, A.; Rönnberg, S.; Bollen, M.H.J.; Moreno-Munoz, A.; Pallares-Lopez, V. Harmonics from a domestic customer with different lamp technologies. In Proceedings of the 2012 IEEE 15th International Conference on Harmonics and Quality of Power, Hong Kong, China, 17–20 June 2012; pp. 585–590. [Google Scholar]
- Electromagnetic Compatibility (EMC)—Part 4-30: Testing and Measurement Techniques—Power Quality Measurement Methods; IEC: Geneva, Switzerland, 2015.
- Bodnar, R.; Otcenasova, A.; Regul’A, M.; Szabo, D. Measurement of harmonics in low-voltage network on the border between SVK and CZE. In Proceedings of the 2014 15th International Scientific Conference on Electric Power Engineering (EPE), Brno, Czech Republic, 2–14 May 2014; pp. 217–222. [Google Scholar]
- Kutt, L.; Saarijärvi, E.; Lehtonen, M.; Mõlder, H.; Vinnal, T. Harmonic load of residential distribution network Case study monitoring results. In Proceedings of the 2014 Electric Power Quality and Supply Reliability Conference (PQ), Rakvere, Estonia, 11–13 June 2014; pp. 93–98. [Google Scholar]
- Kouveliotis-Lysikatos, I.; Kotsampopoulos, P.; Hatziargyriou, N. Harmonic Study in LV networks with high penetration of PV systems. In Proceedings of the 2015 IEEE Eindhoven PowerTech, Eindhoven, The Netherlands, 29 June–2 July 2015; pp. 1–6. [Google Scholar]
- Zhou, K.-L.; Yang, S.-L.; Shen, C. A review of electric load classification in smart grid environment. Renew. Sustain. Energy Rev.
**2013**, 24, 103–110. [Google Scholar] [CrossRef] - Chicco, G.; Ionel, O.-M.; Porumb, R. Electrical Load Pattern Grouping Based on Centroid Model with Ant Colony Clustering. IEEE Trans. Power Syst.
**2012**, 28, 1706–1715. [Google Scholar] [CrossRef] - Jiang, Z.; Lin, R.; Yang, F. A Hybrid Machine Learning Model for Electricity Consumer Categorization Using Smart Meter Data. Energies
**2018**, 11, 2235. [Google Scholar] [CrossRef] [Green Version] - Domagk, M. ‘Identifikation und Quantifizierung korrelativer Zusammenhänge zwischen elektrischer sowie klimatischer Umgebung und Elektroenergiequalität’. Ph.D. Thesis, Technische Universität Dresden, Dresden, Germany, 2015. [Google Scholar]
- Meier, C.H.; Adam, F.T.; Schieferdecker, B. Repräsentative VDEW-Lastprofile; VDEW: Frankfurt, Germany, 1999. [Google Scholar]
- Chakrabarti, S.; Neapolitan, R.E.; Pyle, D. Data Mining: Know It All; Elsevier Science: Amsterdam, The Netherlands, 2008. [Google Scholar]
- Lee, J.-S.; Oh, I.-S. Binary classification trees for multi-class classification problems. In Proceedings of the Seventh International Conference on Document Analysis and Recognition, Proceedings, Edinburgh, UK, 3–6 August 2003. [Google Scholar]
- Wu, J.; Yang, H. Linear Regression-Based Efficient SVM Learning for Large-Scale Classification. IEEE Trans. Neural Networks Learn. Syst.
**2015**, 26, 2357–2369. [Google Scholar] [CrossRef] [PubMed] - Jindal, A.; Dua, A.; Kaur, K.; Singh, M.; Kumar, N.; Mishra, S. Decision Tree and SVM-Based Data Analytics for Theft Detection in Smart Grid. IEEE Trans. Ind. Inform.
**2016**, 12, 1005–1016. [Google Scholar] [CrossRef] - Baghaee, H.R.; Mlakic, D.; Nikolovski, S.; Dragicevic, T.D. Support Vector Machine-Based Islanding and Grid Fault Detection in Active Distribution Networks. IEEE J. Emerg. Sel. Top. Power Electron.
**2019**, 8, 2385–2403. [Google Scholar] [CrossRef] - Naderian, S.; Salemnia, A. Method for classification of PQ events based on discrete Gabor transform with FIR window and T2FK-based SVM and its experimental verification. IET Gener. Transm. Distrib.
**2017**, 11, 133–141. [Google Scholar] [CrossRef] - Thirumala, K.; Prasad, M.S.; Jain, T.; Umarikar, A.C. Tunable-Q Wavelet Transform and Dual Multiclass SVM for Online Automatic Detection of Power Quality Disturbances. IEEE Trans. Smart Grid
**2018**, 9, 3018–3028. [Google Scholar] [CrossRef] - Sha, H.; Mei, F.; Zhang, C.; Pan, Y.; Zheng, J. Identification Method for Voltage Sags Based on K-means-Singular Value Decomposition and Least Squares Support Vector Machine. Energies
**2019**, 12, 1137. [Google Scholar] [CrossRef] [Green Version] - Bravo-Rodríguez, J.C.; Torres, F.J.; Borrás, M.D. Hybrid Machine Learning Models for Classifying Power Quality Disturbances: A Comparative Study. Energies
**2020**, 13, 2761. [Google Scholar] [CrossRef] - Tang, Q.; Qiu, W.; Zhou, Y. Classification of Complex Power Quality Disturbances Using Optimized S-Transform and Kernel SVM. IEEE Trans. Ind. Electron.
**2019**, 67, 9715–9723. [Google Scholar] [CrossRef] - Duda, R.O.; Hart, P.E.; Stork, D.G. Pattern Classification, 2nd ed.; John Wiley & Sons: New York, NY, USA, 2012. [Google Scholar]
- Bollen, M.H.J.; Gu, I.Y.-H. Signal Processing of Power Quality Disturbances; Wiley: Hoboken, NJ, USA, 2006. [Google Scholar]
- Abe, S. Support Vector Machines for Pattern Classification; Springer Science and Business Media LLC: Berlin/Heidelberg, Germany, 2010. [Google Scholar]
- Chang, C.-C.; Lin, C.-J. LIBSVM. ACM Trans. Intell. Syst. Technol.
**2011**, 2, 1–27. [Google Scholar] [CrossRef] - Athimethphat, M.; Lerteerawong, B. Binary classification tree for multiclass classification with observation-based clustering. In Proceedings of the 2012 9th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, Phetchaburi, Thailand, 16–18 May 2012. [Google Scholar] [CrossRef]
- Tukey, J.W. Exploratory Data Analysis; Addison-Wesley: Boston, MA, USA, 1977. [Google Scholar]

**Figure 1.**Classification of variation types and major impact factors in public LV networks [8].

**Figure 2.**Profile of the 3rd harmonic current for an office building; median (red), 5th and 95th percentile (blue), daily time series (grey); aggregation interval: 10 min; measurement duration: 320 days from January to December 2012.

**Figure 3.**Profile of the 3rd harmonic current for an electronic market; median (red), 5th and 95th percentile (blue), daily time series (grey); aggregation interval: 10 min; measurement duration: 353 days from January to December 2012.

**Figure 4.**Example for typical emission profiles of the 3rd harmonic order for different consumer configurations: residential (RES), commercial (COM), office (OFF), and mixed (MIX) areas; min-max scaled 50th percentile time series for different day types: working days (G1), Saturdays (G2), and Sundays/holidays (G3); phase L1; aggregation interval: 10 min.

**Figure 5.**Classes of emission profiles for the 3rd harmonic current; min–max normalized 50th percentile time series for 40 measurements in LV grids within three groups: working days, Saturdays, and Sundays/holidays for all phases L1 to L3.

**Figure 7.**Features of the emission profiles for the first classifier support vector machine (SVM)-1 to distinguish class D0 and classes (D1, D2, D3, D4).

**Figure 8.**Grid search for the optimal parameters of the classifier SVM-1 using a cross-validation with 10 subgroups (k = 1); error rate in p.u.

**Figure 9.**Training for the classifier SVM-1, with the resulting hyperplane using optimal parameters (

**a**), and suboptimal parameters (

**b**).

**Figure 10.**Robustness for the classification of 360 emission profiles; 100 repetitions, with random splits into 50% training samples and 50% test samples; box whisker plots of average and per class classification rates for training samples (white) and test samples (grey).

**Figure 11.**Measures of misclassification for the classes D1 and D2, and emission profiles of training samples (grey), and test samples (blue/red).

**Table 1.**Number of emission profiles per class using a random separation into training samples and test samples.

Class | D0 | D1 | D2 | D3 | D4 |
---|---|---|---|---|---|

Training samples | 27 | 85 | 15 | 15 | 38 |

Test samples | 23 | 79 | 15 | 15 | 48 |

**Table 2.**Classification results for the training samples of Table 1.

Class | D0 | D1 | D2 | D3 | D4 | Average |
---|---|---|---|---|---|---|

Number of misclassifications | 1 | 4 | 0 | 1 | 3 | 1.8 |

Classification rate in % | 96.3 | 95.3 | 100 | 93.3 | 92.1 | 95.4 |

**Table 3.**Classification results for the test samples of Table 1.

Class | D0 | D1 | D2 | D3 | D4 | Average |
---|---|---|---|---|---|---|

Number of misclassifications | 0 | 4 | 1 | 1 | 2 | 1.6 |

Classification rate in % | 100 | 94.9 | 93.3 | 93.3 | 95.8 | 95.5 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Domagk, M.; Gu, I.Y.-H.; Meyer, J.; Schegner, P.
Automatic Identification of Different Types of Consumer Configurations by Using Harmonic Current Measurements. *Appl. Sci.* **2021**, *11*, 3598.
https://doi.org/10.3390/app11083598

**AMA Style**

Domagk M, Gu IY-H, Meyer J, Schegner P.
Automatic Identification of Different Types of Consumer Configurations by Using Harmonic Current Measurements. *Applied Sciences*. 2021; 11(8):3598.
https://doi.org/10.3390/app11083598

**Chicago/Turabian Style**

Domagk, Max, Irene Yu-Hua Gu, Jan Meyer, and Peter Schegner.
2021. "Automatic Identification of Different Types of Consumer Configurations by Using Harmonic Current Measurements" *Applied Sciences* 11, no. 8: 3598.
https://doi.org/10.3390/app11083598