Series Arc Fault Detection Method Based on Category Recognition and Artificial Neural Network
Abstract
:1. Introduction
2. Influence of Arc Fault on Current Signal under Different Loads
3. Detection Method
3.1. Basis for Dividing the Half Period
3.2. Wavelet Transform to Process Signals
3.2.1. Choosing the Wavelet Base
3.2.2. Frequency Band Distribution of the Wavelet Transform
3.3. Category Recognition
3.3.1. Primary Classification
3.3.2. Secondary Classification
3.4. Time-Frequency Indicators Selection
3.4.1. Effective Value of the Current
3.4.2. Degree of Fluctuation
3.4.3. Average Energy
3.5. Construction and Optimization of an Artificial Neural Network
3.5.1. Establishing the BP Neural Network
3.5.2. Genetic Algorithm Optimized Neural Network
4. Experimental Verification
4.1. Experimental Device and Experimental Object Selection
4.2. Category Recognition Results
4.3. Fault Arc Identification Results
5. Comparison with Previous Detection Methods
6. Conclusions
Author Contributions
Funding
Conflicts of Interest
References
- Underwriters Laboratories Inc. UL Standard for Arc-Fault Circuit-Interrupters, 2nd ed.; Underwriters Laboratories Inc.: New York, NY, USA, 2011. [Google Scholar]
- International Electrotechnical Commission. General Requirements for Arc Fault Detection Devices; IEC 62606; International Electrotechnical Commission: Geneva, Switzerland, 2017. [Google Scholar]
- General Administration of Quality Supervision, Inspection and Quarantine of China; Standardization Administration of China. Electrical Fire Monitoring System–Part 4: Arcing Fault Detectors (GB14287.4-2014); Standards Press of China: Beijing, China, 2014. [Google Scholar]
- General Administration of Quality Supervision, Inspection and Quarantine of China; Standardization Administration of China. General Requirements for Arc Fault Detection Devices (AFDD) (GB/T31143); Standards Press of China: Beijing, China, 2014. [Google Scholar]
- Artale, G.; Cataliotti, A.; Cosentino, V.; Privitera, G. Experimental Characterization of Series Arc Faults in AC and DC Electrical Circuits. In Proceedings of the 2014 IEEE International Instrumentation and Measurement Technology Conference: Instrumentation and Measurement for Sustainable Development, I2MTC 2014, Montevideo, Uruguay, 12–15 May 2014; Institute of Electrical and Electronics Engineers Inc.: Montevideo, Uruguay, 2014; pp. 1015–1020. [Google Scholar]
- Kumpulainen, L.; Hussain, G.A.; Rival, M.; Lehtonen, M.; Kauhaniemi, K. Aspects of arc-flash protection and prediction. Electr. Power Syst. Res. 2014, 116, 77–86. [Google Scholar] [CrossRef]
- Sifuzzaman, M.; Islam, M.; Ali, M. Application of Wavelet Transform and its Advantages Compared to Fourier Transform. J. Phys. Sci. 2009, 13, 121–134. [Google Scholar]
- Duan, P.; Xu, L.; Ding, X.; Ning, C.; Duan, C. An Arc Fault Diagnostic Method for Low Voltage Lines Using the Difference of Wavelet Coefficients. In Proceedings of the 2014 9th IEEE Conference on Industrial Electronics and Applications, Hangzhou, China, 9–11 June 2014; pp. 401–405. [Google Scholar]
- Zhao, Y.; Zhang, X.; Dong, Y.; Li, W. Characteristics Analysis and Detection of AC Arc Fault in SSPC based on Wavelet Transform. In Proceedings of the 2016 IEEE International Conference on Aircraft Utility Systems (AUS), Beijing, China, 10–12 October 2016; pp. 476–481. [Google Scholar]
- Yin, Z.; Wang, L.; Zhang, Y.; Gao, Y. A Novel Arc Fault Detection Method Integrated Random Forest, Improved Multi-scale Permutation Entropy and Wavelet Packet Transform. Electronics 2019, 8, 396. [Google Scholar] [CrossRef] [Green Version]
- Li, W.; He, K.; Liu, W.; Zhang, X.; Dong, Y. A fast arc fault detection method for AC solid state power controllers in MEA. Chin. J. Aeronaut. 2018, 31, 1119–1129. [Google Scholar] [CrossRef]
- She, Z.; Yang, L.; Hu, D. Detection Method of Series Arc Fault Based on Wavelet Transform. In Proceedings of the 2018 13th World Congress on Intelligent Control and Automation (WCICA), Changsha, China, 4–8 July 2018; pp. 1026–1030. [Google Scholar]
- Lu, Q.; Wang, T.; He, B.; Ru, T.; Chen, D. A new series arc fault identification method based on wavelet transform. In Proceedings of the IECON 2017-43rd Annual Conference of the IEEE Industrial Electronics Society, Beijing, China, 29 October–1 November 2017; pp. 4817–4822. [Google Scholar]
- Ilman, A. Low Voltage Series Arc Fault Detecting With Discrete Wavelet Transform. In Proceedings of the 2018 International Conference on Applied Engineering, Batam, Indonesia, 3–4 October 2018; pp. 1–5. [Google Scholar]
- Ji, H.-K.; Wang, G.; Kim, W.-H.; Kil, G.-S. Optimal Design of a Band Pass Filter and an Algorithm for Series Arc Detection. Energies 2018, 11, 992. [Google Scholar] [CrossRef] [Green Version]
- Wang, Y.; Zhang, F.; Zhang, S.; Yang, G. A novel diagnostic algorithm for AC series arcing based on correlation analysis of high-frequency component of wavelet. COMPEL-Int. J. Comput. Math. Electr. Electron. Eng. 2017, 36, 271–288. [Google Scholar] [CrossRef]
- Nan, C.; Yixing, T.; Qian, Z. Experimental Research about the Current Characteristic of the Fault Arc in Series based on Wavelet Analysis. Procedia Eng. 2014, 84, 736–741. [Google Scholar] [CrossRef]
- Kostyantyn, K.; Bei, G.; Joel, A. A low-cost power-quality meter with series arc-fault detection capability for smart grid. IEEE Trans. Power Deliv. 2013, 28, 1584–1591. [Google Scholar]
- Wang, Y.; Zhang, F.; Zhang, S. A New Methodology for Identifying Arc Fault by Sparse Representation and Neural Network. IEEE Trans. Instrum. Meas. 2018, 67, 2526–2537. [Google Scholar] [CrossRef]
- Wu, C.; Hung, C.; Liu, Y. Smart Detection Technology of Serial Arc-Fault on Low Voltage Power Lines. Adv. Mater. Res. 2013, 734, 3190–3193. [Google Scholar] [CrossRef]
- Qu, N.; Chen, J.; Zuo, J.; Liu, J. PSO–SOM Neural Network Algorithm for Series Arc Fault Detection. Adv. Math. Phys. 2020, 2020, 1–8. [Google Scholar] [CrossRef]
- Yang, K.; Chu, R.B.; Zhang, R.C.; Xiao, J.C.; Tu, R. A Novel Methodology for Series Arc Fault Detection by Temporal Domain Visualization and Convolutional Neural Network. Sensors 2020, 20, 162. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Calderón, E.; Schweitzer, P.; Weber, S. A double ended AC series arc fault location algorithm for a low-voltage indoor power line using impedance parameters and a neural network. Electr. Power Syst. Res. 2018, 165, 84–93. [Google Scholar] [CrossRef]
- Siegel, J.; Pratt, S.; Sun, Y.; Sarma, S. Real-time Deep Neural Networks for internet-enabled arc-fault detection. Eng. Appl. Artif. Intell. 2018, 74, 35–42. [Google Scholar] [CrossRef] [Green Version]
- Wang, Y.; Zhang, F.; Zhang, X.; Zhang, S. Series AC Arc Fault Detection Method Based on Hybrid Time and Frequency Analysis and Fully Connected Neural Network. IEEE Trans. Ind. Inform. 2019, 15, 6210–6219. [Google Scholar] [CrossRef]
- Lu, Q.W.; Ye, Z.Y.; Zhang, Y.L.; Wang, T.; Gao, Z.X. Analysis of the Effects of Arc Volt—Ampere Characteristics on Different Loads and Detection Methods of Series Arc Faults. Energies 2019, 12, 323. [Google Scholar] [CrossRef] [Green Version]
- Kanwaljot, S.S.; Baljeet, S.K.; Ishpreet, S.V. Medical Image Denoising in the Wavelet Domain Using Haar and DB3 Filtering. Int. Refereed J. Eng. Sci. 2012, 1, 1–8. [Google Scholar]
- Yang, H.; Zhao, D. Application of Wavelet Transform to Image Compression Based on Mallat Algorithm. In Proceedings of the 2013 Asia-Pacific Computational Intelligence and Information Technology Conference (APCIIT), Shanghai, China, 28 December 2013; pp. 193–199. [Google Scholar]
- Spyker, R.; Schweickart, D.L.; Horwath, J.C.; Walko, L.C.; Grosjean, D. An evaluation of diagnostic techniques relevant to arcing fault current interrupters for direct current power systems in future aircraft. In Proceedings of the Electrical Insulation Conference and Electrical Manufacturing Conference, Indianapolis, IN, USA, 23–26 October 2005; pp. 146–150. [Google Scholar]
- Bao, G.; Jiang, R.; Gao, X. Novel Series Arc Fault Detector Using High-frequency Coupling Analysis and Multi-indicator Algorithm. IEEE Access 2019, 7, 92161–92170. [Google Scholar] [CrossRef]
- Jin, W.; Li, Z.; Wei, L.; Zhen, H. The improvements of BP neural network learning algorithm. In Proceedings of the 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000(WCC 2000-ICSP 2000), Beijing, China, 21–25 August 2000; pp. 1647–1649. [Google Scholar]
- Ding, S.F.; Su, C.Y.; Yu, J.Z. An optimizing BP neural network algorithm based on genetic algorithm. Artif. Intell. Rev. 2011, 36, 153–162. [Google Scholar] [CrossRef]
- Han, Z.; Huang, X. GA-BP in Thermal Fatigue Failure Prediction of Microelectronic Chips. Electronics 2019, 8, 542. [Google Scholar] [CrossRef] [Green Version]
- Growe, G. Comparing Algorithms and Clustering Data: Components of the Data Mining Process. Master’s Thesis, Grand Valley State University, Allendale, Michigan, 2000. [Google Scholar]
- Looney, C.G. Advances in feedforward neural networks: Demystifying knowledge acquiring black boxes. IEEE Trans. Knowl. Data Eng. 1996, 8, 211–226. [Google Scholar] [CrossRef]
Layer Number | Approximate Signal Concentration Frequency Band (kHz) | Detail Signal Concentration Frequency Band (kHz) | ||
---|---|---|---|---|
1 | a1 | 0–50 | d1 | 50–100 |
2 | a2 | 0–25 | d2 | 25–50 |
3 | a3 | 0–12.5 | d3 | 12.5–25 |
4 | a4 | 0–6.25 | d4 | 6.25–12.5 |
5 | a5 | 0–3.125 | d5 | 3.125–6.25 |
Primary Classification | Secondary Classification |
---|---|
Normal without shoulder | Resistive category (Re) |
Resistive-inductive category (RI) | |
Normal with shoulder | Rectifying circuit with a capacitive filter category (RCCF) |
Category | Time Domain Indicator | Frequency Domain Indicator |
---|---|---|
Re | Shoulder proportion | Average energy |
RI | Degree of fluctuation | |
RCCF |
Population Size | Genetic Algebra | Mutation Probability | Crossover Probability | Generation Gap |
---|---|---|---|---|
20 | 50 | 0.01 | 0.7 | 0.95 |
Experimental Load | Category |
---|---|
Pure resistance (15 Ω) Fluorescent lamp (80 W in total) Halogen lamp (300 W) | Resistive category (Re) |
Electric drill (600 W) | Resistive-inductive category (RI) |
Air conditioner (4500 W) Computer host (two sets of 500 W in total) Tungsten filament lamp (conduction angle 60°) | Rectifying circuit with a capacitive filter category (RCCF) |
Category | Number of Samples | Classification Accuracy% |
---|---|---|
Re | 210 | 100% |
RI | 210 | 100% |
RCCF | 210 | 100% |
Total | 630 | 100% |
Load Categories | Training Samples 70% | Test Samples 30% | Comprehensive Accuracy | ||||||
---|---|---|---|---|---|---|---|---|---|
Normal Group | Arc Group | Total | Normal Group | Arc Group | Total | BP Networks | GA-BP Networks | ||
Re | Pure resistance | 75 | 75 | 450 | 30 | 30 | 180 | 90.79% 1 ~ 93.65% | 99.21% |
Fluorescent lamp | |||||||||
Halogen lamp | |||||||||
RCCF | Air conditioner | 75 | 75 | 30 | 30 | ||||
Computer host | |||||||||
Tungsten Filament lamp (with dimmer) | |||||||||
RI | Electric drill | 75 | 75 | 30 | 30 |
Reference [25] | Reference [21] | Reference [24] | Test Method in This Paper | |
---|---|---|---|---|
Method framework | Select the 1st, 3rd, and 5th harmonics to extract the frequency domain feature quantities for category recognition, combine the time domain feature values to train a fully connected neural network | Extract frequency domain indicators, train SOM neural network | Use Fourier coefficients and wavelet eigenvalues as input for deep neural networks | Two-level category recognition based on current and voltage waveforms, combine wavelet transform to extract time-frequency eigenvalues, train BP neural network |
Recognition standard | Extract feature quantity from 1.3.5 harmonic | Not mentioned | Not mentioned | The relationship of load current, voltage waveform, and current waveform |
Recognition category | RE, CI, SW | Re, RI, RCCF | ||
Neural network structure | 4-5-5-2, 2-5-5-4, 5-12-12-4 | 49-36-10 | 607-16-32-16-1 | 2-5-1 |
Neural network optimization | Not mentioned | Use particle swarm optimization to optimize initial value of neural network | Not mentioned | Use genetic algorithm to optimize initial value of neural network |
Detection accuracy | 99.00% | 95.00% | 95.61% | 99.21% |
Category of GA-BP Network | Training Time(s) | Running Time of Detection Function(s) |
---|---|---|
Re | 372.34 | 0.009 |
RI | 382.36 | 0.011 |
RCCF | 403.23 | 0.010 |
© 2020 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 (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Han, X.; Li, D.; Huang, L.; Huang, H.; Yang, J.; Zhang, Y.; Wu, X.; Lu, Q. Series Arc Fault Detection Method Based on Category Recognition and Artificial Neural Network. Electronics 2020, 9, 1367. https://doi.org/10.3390/electronics9091367
Han X, Li D, Huang L, Huang H, Yang J, Zhang Y, Wu X, Lu Q. Series Arc Fault Detection Method Based on Category Recognition and Artificial Neural Network. Electronics. 2020; 9(9):1367. https://doi.org/10.3390/electronics9091367
Chicago/Turabian StyleHan, Xiangyu, Dingkang Li, Lizong Huang, Hanqing Huang, Jin Yang, Yilei Zhang, Xuewei Wu, and Qiwei Lu. 2020. "Series Arc Fault Detection Method Based on Category Recognition and Artificial Neural Network" Electronics 9, no. 9: 1367. https://doi.org/10.3390/electronics9091367