Vol. 172
Latest Volume
All Volumes
PIERC 173 PIERC 172 PIERC 171 PIERC 170 PIERC 169 PIERC 168 PIERC 167 PIERC 166 PIERC 165 PIERC 164 PIERC 163 PIERC 162 PIERC 161 PIERC 160 PIERC 159 PIERC 158 PIERC 157 PIERC 156 PIERC 155 PIERC 154 PIERC 153 PIERC 152 PIERC 151 PIERC 150 PIERC 149 PIERC 148 PIERC 147 PIERC 146 PIERC 145 PIERC 144 PIERC 143 PIERC 142 PIERC 141 PIERC 140 PIERC 139 PIERC 138 PIERC 137 PIERC 136 PIERC 135 PIERC 134 PIERC 133 PIERC 132 PIERC 131 PIERC 130 PIERC 129 PIERC 128 PIERC 127 PIERC 126 PIERC 125 PIERC 124 PIERC 123 PIERC 122 PIERC 121 PIERC 120 PIERC 119 PIERC 118 PIERC 117 PIERC 116 PIERC 115 PIERC 114 PIERC 113 PIERC 112 PIERC 111 PIERC 110 PIERC 109 PIERC 108 PIERC 107 PIERC 106 PIERC 105 PIERC 104 PIERC 103 PIERC 102 PIERC 101 PIERC 100 PIERC 99 PIERC 98 PIERC 97 PIERC 96 PIERC 95 PIERC 94 PIERC 93 PIERC 92 PIERC 91 PIERC 90 PIERC 89 PIERC 88 PIERC 87 PIERC 86 PIERC 85 PIERC 84 PIERC 83 PIERC 82 PIERC 81 PIERC 80 PIERC 79 PIERC 78 PIERC 77 PIERC 76 PIERC 75 PIERC 74 PIERC 73 PIERC 72 PIERC 71 PIERC 70 PIERC 69 PIERC 68 PIERC 67 PIERC 66 PIERC 65 PIERC 64 PIERC 63 PIERC 62 PIERC 61 PIERC 60 PIERC 59 PIERC 58 PIERC 57 PIERC 56 PIERC 55 PIERC 54 PIERC 53 PIERC 52 PIERC 51 PIERC 50 PIERC 49 PIERC 48 PIERC 47 PIERC 46 PIERC 45 PIERC 44 PIERC 43 PIERC 42 PIERC 41 PIERC 40 PIERC 39 PIERC 38 PIERC 37 PIERC 36 PIERC 35 PIERC 34 PIERC 33 PIERC 32 PIERC 31 PIERC 30 PIERC 29 PIERC 28 PIERC 27 PIERC 26 PIERC 25 PIERC 24 PIERC 23 PIERC 22 PIERC 21 PIERC 20 PIERC 19 PIERC 18 PIERC 17 PIERC 16 PIERC 15 PIERC 14 PIERC 13 PIERC 12 PIERC 11 PIERC 10 PIERC 9 PIERC 8 PIERC 7 PIERC 6 PIERC 5 PIERC 4 PIERC 3 PIERC 2 PIERC 1
2026-08-25
Multi-Objective Cooperative Control of a Three-Level NPC Active Power Filter Based on Single-Agent DDPG
By
Progress In Electromagnetics Research C, Vol. 172, 520-532, 2026
Abstract
Conventional multi loop control of diode clamped active power filters suffers from response conflicts and slow regulation due to strong coupling among DC voltage, neutral point potential, and reactive power. This paper proposes a single agent deep deterministic policy gradient (DDPG) framework for multi objective cooperative control. Unlike generic DRL applications, our design retains the current PI inner loop for hardware safety. At the same time, a single agent replaces the three outer loops (voltage, neutral point, and reactive power) to eliminate cascade coupling. A composite reward function based on the L2 norm - rather than linear weighting - quadratically penalizes large errors, forcing the agent to prioritize the most severe deviations and resolve multi objective conflicts optimally. Simulations show that the proposed strategy reduces grid side current THD from 6.7% to 1.09%, cuts midpoint balancing and DC bus settling times by 50% and 66.67% respectively, achieves zero overshoot and zero steady state error, and exhibits strong robustness under voltage sags and load surges. This work offers a customized, low tuning effort intelligent solution for power quality control in complex grids.
Citation
Zihang Xie, Jianxun Mo, Ziqiang Liu, Shijie Fang, Zibo Wang, Wenbiao Li, and Mingrui Chen, "Multi-Objective Cooperative Control of a Three-Level NPC Active Power Filter Based on Single-Agent DDPG," Progress In Electromagnetics Research C, Vol. 172, 520-532, 2026.
doi:10.2528/PIERC26051901
References

1. Tian, Shuya, Zhibo Ning, Pengcheng Li, Yingjun Guo, Zhiqiang Li, and Hexu Sun, "Siting and capacity allocation for APFs based on optimal reactive power compensation and harmonic mitigation capacities of STATCOMs and MFGCIs," International Journal of Electrical Power & Energy Systems, Vol. 173, 111357, Dec. 2025.
doi:10.1016/j.ijepes.2025.111357        Google Scholar

2. Etanya, Tiku Fidelis, Pierre Tsafack, and Divine Khan Ngwashi, "Grid-connected distributed renewable energy generation systems: Power quality issues, and mitigation techniques --- A review," Energy Reports, Vol. 13, 3181-3203, Jun. 2025.
doi:10.1016/j.egyr.2025.02.050        Google Scholar

3. Elkholy, Mahmoud M., M. A. El-Hameed, and A. A. El-Fergany, "Harmonic analysis of hybrid renewable microgrids comprising optimal design of passive filters and uncertainties," Electric Power Systems Research, Vol. 163, 491-501, Oct. 2018.
doi:10.1016/j.epsr.2018.07.023        Google Scholar

4. Wu, Zhaokang, Guoqing Xu, Wei Zhu, and Gang Sheng, "A dispersing proportional resonance controller for selective harmonic current compensation in a shunt active power filter," IEEE Transactions on Power Electronics, Vol. 40, No. 1, 279-289, Jan. 2025.
doi:10.1109/tpel.2024.3457990        Google Scholar

5. Smaida, Elyas, Juma Saulo, Kouzou Abdellah, Mary Ahuna, and Ali Teta, "A novel multifunctional shunt active power filter based AI-driven for grid-tied PV system using Z-source inverter," 2025 4th International Conference on Computing and Information Technology (ICCIT), 656-661, Tabuk, Saudi Arabia, Apr. 2025.
doi:10.1109/ICCIT63348.2025.10989371

6. Liu, Jianben, Yuan Ni, and Jun Zhao, "An improved control method of DC voltage for series hybrid active power filter," Energies, Vol. 17, No. 14, 3390, Jul. 2024.
doi:10.3390/en17143390        Google Scholar

7. Santiprapan, Phonsit, Kongpol Areerak, and Kongpan Areerak, "An adaptive gain of proportional-resonant controller for an active power filter," IEEE Transactions on Power Electronics, Vol. 39, No. 1, 1433-1446, Jan. 2024.
doi:10.1109/tpel.2023.3319476        Google Scholar

8. Gök, Rukiye, Alperen M. Çolak, Erdal Irmak, and Nihat Öztürk, "Comparative analysis of fuzzy logic and PI control for harmonic suppression in shunt active power filter," 2024 13th International Conference on Renewable Energy Research and Applications (ICRERA), 1651-1656, Nagasaki, Japan, Nov. 2024.
doi:10.1109/ICRERA62673.2024.10815186

9. Zakaria, Laala and Benaissa Amar, "Neural network controller for three phase shunt active power filter using self tuning filter," 2022 19th International Multi-Conference on Systems, Signals & Devices (SSD), 333-338, Sétif, Algeria, May 2022.
doi:10.1109/SSD54932.2022.9955663

10. Daou, Nora, Mohammed Saci Chabani, Betka Achour, Salah Eddine Zouzou, Vicenç Puig, Mahmoud A. Mossa, and Mishari Metab Almalki, "Robust second-order sliding-mode backstepping control of the parallel active filter for harmonics and reactive power compensation-experimental validation," Energy Reports, Vol. 15, 109117, Jun. 2026.
doi:10.1016/j.egyr.2026.109117        Google Scholar

11. Balouji, Ebrahim, Karl Bäckstrüm, and Tomas McKelvey, "Deep reinforcement learning based grid-forming inverter," 2023 IEEE Industry Applications Society Annual Meeting (IAS), 1-9, Nashville, TN, USA, Oct. 2023.
doi:10.1109/IAS54024.2023.10406664

12. Li, Baoliang, Qiuwei Wu, Yongji Cao, Wenshu Jiao, and Changgang Li, "Physically informed multi-agent deep reinforcement learning for distributed voltage control in distribution networks," International Journal of Electrical Power & Energy Systems, Vol. 174, 111451, Jan. 2026.
doi:10.1016/j.ijepes.2025.111451        Google Scholar

13. Wang, He, Jinling Li, Shiqiang Li, Xiao Liu, Jing Bian, and Huanan Yu, "Optimal inverter-based droop control in active distribution network with evolutionary strategy-embedded deep reinforcement learning," Electric Power Systems Research, Vol. 247, 111789, Oct. 2025.
doi:10.1016/j.epsr.2025.111789        Google Scholar

14. Dash, Dipak Kumar, Pradip Kumar Sadhu, and Alok Kumar Shrivastav, "Dynamic optimization of active power filters in grid-tied photovoltaic systems using partial reinforcement learning," Neural Computing and Applications, Vol. 37, No. 20, 15605-15634, May 2025.
doi:10.1007/s00521-025-11186-6        Google Scholar

15. Li, Peishuai, Jiawei Shen, Zaijun Wu, Minghui Yin, Yanhao Dong, and Jing Han, "Optimal real-time Voltage/Var control for distribution network: Droop-control based multi-agent deep reinforcement learning," International Journal of Electrical Power & Energy Systems, Vol. 153, 109370, Nov. 2023.
doi:10.1016/j.ijepes.2023.109370        Google Scholar

16. Wang, Jianing, Renhai Yang, and Zhanghao Yao, "Efficiency optimization design of three-level active neutral point clamped inverter based on deep reinforcement learning," 2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2), 605-610, Chengdu, China, Nov. 2022.
doi:10.1109/EI256261.2022.10117037

17. Qashqai, Pouria, Mohammad Babaie, Rawad Zgheib, and Kamal Al-Haddad, "A model-free switching and control method for three-level neutral point clamped converter using deep reinforcement learning," IEEE Access, Vol. 11, 105394-105409, Sep. 2023.
doi:10.1109/access.2023.3318264        Google Scholar

18. Liang, Zhenyan, Jiang Wang, Yitong Wu, and Zhen Zhang, "Weight-adaptable disturbance observer for continuous-control-set model predictive control of NPC-3L-Fed PMSMs," Energies, Vol. 18, No. 21, 5864, Nov. 2025.
doi:10.3390/en18215864        Google Scholar

19. Xu, Dongli and Liuling Zhang, "Research on PVPC system based on an improved ip-iq algorithm," 2021 IEEE Sustainable Power and Energy Conference (iSPEC), 1395-1399, Nanjing, China, Dec. 2021.
doi:10.1109/iSPEC53008.2021.9735705

20. Hrgović, Ivana and Ivica Pavić, "Reward design for intelligent deep reinforcement learning based power flow control using topology optimization," Sustainable Energy, Grids and Networks, Vol. 41, 101580, Mar. 2025.
doi:10.1016/j.segan.2024.101580        Google Scholar