DEEP LEARNING BASED PREDICTION OF POWER QUALITY DISTURBANCES IN DISTRIBUTION NETWORKS

Abstract

The main objective of this study is to predict harmonic distortions in a power distribution system using real-world active and reactive power data. To achieve this, a Gated Recurrent Unit (GRU)-based artificial intelligence algorithm was employed, which is particularly effective in modeling the dynamic nature of time series. Unlike conventional methods, the GRU model demonstrates successful performance by shortening training duration and increasing prediction accuracy. The prediction results yielded promising error metrics, with mean absolute error (MAE) values of 0.5200, 0.5330, and 0.5771; mean absolute percentage error (MAPE) values of 7.52%, 7.55%, and 7.72%; and root mean square error (RMSE) values of 0.7014, 0.7231, and 0.7848 for the THD_I1, THD_I2, and THD_I3 indices, respectively. These findings indicate that the proposed approach provides a reliable and practical solution for predicting harmonic distortions and can effectively support decision-making mechanisms aimed at enhancing power quality in distribution systems.

Keywords

• electric power systems
• deep learning
• power quality

References

1. Y. Li, Y. Sun, Q. Wang, K. Sun, K.-J. Li, and Y. Zhang, “Probabilistic harmonic forecasting of the distribution system considering time-varying uncertainties of the distributed energy resources and electrical loads,” Applied Energy, vol. 329, art. no. 120298, Nov. 2022.
2. Žnidarec, M.; Klaić, Z.; Šljivac, D.; Dumnić, B. “Harmonic Distortion Prediction Model of a Grid‐Tie Photovoltaic Inverter Using an Artificial Neural Network,” Energies, 12(5):790, 2019.
3. Ł. Michalec, M. Jasiński, T. Sikorski, Z. Leonowicz, Ł. Jasiński, and V. Suresh, “Impact of harmonic currents of nonlinear loads on power quality of a low voltage network—review and case study,” Energies, vol. 14, no. 12, p. 3665, 2021.
4. E. M. Kuyunani, A. N. Hasan, and T. Shongwe, “Improving voltage harmonics forecasting at a wind farm using deep learning techniques,” in Proc. IEEE Int. Symp. Ind. Electron. (ISIE), pp. 1–6, 2021.
5. H. Sharma, M. Rylander, and D. Dorr, “Grid impacts due to increased penetration of newer harmonic sources,” in Proc. IEEE Rural Electric Power Conference (REPCON), pp. B3–1–B3–7, 2013.
6. E. M. Kuyumani, A. N. Hasan, and T. Shongwe, “A Hybrid Model Based on CNN-LSTM to Detect and Forecast Harmonics: A Case Study of an Eskom Substation in South Africa,” Electric Power Components and Systems, vol. 51, no. 8, pp. 746–760, May 2023.
7. F. M. Al Hadi, H. H. Aly, “Harmonics Forecasting of Renewable Energy System Using Hybrid Model Based on LSTM and ANFIS,” IEEE Access, in press, art. no. 3386092, 2024.
8. M. Panoiu, C. Panoiu, S. Mezinescu, G. Militaru, and I. Baciu, “Machine Learning Techniques Applied to the Harmonic Analysis of Railway Power Supply,” Mathematics, vol. 11, no. 6, p. 1381, Mar. 2023.

9. E. Yiğit, U. Özkaya, Ş. Öztürk, D. Singh, and H. Gritli, “Automatic Detection of Power Quality Disturbance Using Convolutional Neural Network Structure with Gated Recurrent Unit,” Mobile Information Systems, vol. 2021, Article ID 7917500, pp. 1–13, 2021.
10. Y. Dong, F. Zhang, X. Li, L. Zhang, J. Yu, Y. Mao, and G. Jiang, “Nonlinear Load Harmonic Prediction Method Based on Power Distribution Internet of Things,” Scientific Programming, vol. 2021, Article ID 9978900, 12 pages, May 2021.
11. A. Karadeniz, “Advancing harmonic prediction for offshore wind farms using synthetic data and machine learning,” Computers and Electrical Engineering, 127: 110613. 2025.
12. A. Karadeniz, “Harmonic forecasting in offshore wind systems utilizing DFIG on bozcaada island: a hybrid machine learning and deep learning approach,” Engineering Research Express, 7.2: 025280, 2025.
13. Y. Zhao and J. V. Milanović, "Prediction of Harmonic Distortion in Sparsely Monitored Transmission Networks With Renewable Generation," IEEE Transactions on Power Delivery, vol. 39, no. 3, pp. 1710-1722, June 2024.
14. A. Iqbal, A. Rahman, K. Thomas, M. Abdushakkoor, S. Islam, U. Attique, "Predictive Modeling of Harmonic Distortion in Electrical Systems Using Machine Learning," IECON 2024 - 50th Annual Conference of the IEEE Industrial Electronics Society, Chicago, IL, USA, pp. 1-6, 2024.
15. Y. Terriche, A. Lashab, H. Çimen, J. M. Guerrero, C.-L. Su, and J. C. Vasquez, “Power Quality Assessment Using Signal Periodicity Independent Algorithms – A Shipboard Microgrid Case Study,” Applied Energy, vol. 307, art. no. 118151, 2022.
16. J. J. Flores, J. L. Garcia-Nava, J. R. Cedeno Gonzalez, V. M. Tellez, F. Calderon, and A. Medrano, “A Machine-Learning Pipeline for Large-Scale Power-Quality Forecasting in the Mexican Distribution Grid,” Applied Sciences, vol. 12, no. 17, art. no. 8423, 2022.
17. A. Satapathy, N. Nayak, and T. Parida, “Real-Time Power Quality Enhancement in a Hybrid Micro-Grid Using Nonlinear Autoregressive Exogenous Input Neural Network,” Energies, vol. 15, no. 23, art. no. 9081, Nov. 2022.
18. R. Tiwari, M. Senthil Kumar, T. D. Diwan, L. Pinjarkar, K. Mehta, H. Nayak, R. Reddy, A. Nigam, and R. Shrivastava, “Enhanced Power Quality and Forecasting for PV-Wind Microgrid Using Proactive Shunt Power Filter and Neural Network-Based Time Series Forecasting,” Electric Power Components and Systems, 2023. https://doi.org/10.1080/15325008.2023.2249894
19. P. Rodríguez-Pajarón, A. Hernández-Bayo, and J. V. Milanović, “Forecasting voltage harmonic distortion in residential distribution networks using smart meter data,” International Journal of Electrical Power & Energy Systems, vol. 136, art. no. 107653, Oct. 2021.
20. P. Kiran and A. S. Menon, “Harmonics and Neurons: a Fourier-Neural approach to energy pattern analysis,” Discover Applied Sciences, vol. 7, art. no. 133, Feb. 2025.
21. İ. Bozdağ, S. B. Efe, and İ. Özer, “Short-term prediction of power quality disturbances in electrical energy systems using LSTM and GRU networks,” Scientia Iranica, 2023. 10.24200/sci.2023.61430.730