This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.
Journal Menu
By: Taj Prakash Verma and Rohit Kumar.
1.Assistant Professor, Department of Electrical Engineering, B.I.E.T. Lucknow, Uttar pradesh, India
2.Student, Department of Electrical Engineering, B.I.E.T. Lucknow, Uttar pradesh, India
Partial shade and quickly changing environmental conditions make Maximum Power Point Tracking (MPPT) in photovoltaic (PV) arrays particularly difficult. In these situations, traditional algorithms frequently fail to find the global optimum or experience sluggish convergence and steady-state oscillations. To obtain reliable, real-time MPPT under non-uniform irradiance, this research suggests a hybrid control architecture that combines a deep recurrent reinforcement learning (DRL) agent with metaheuristic-based parameter optimization and a digital-twin training environment. While a metaheuristic optimizer (such as evolutionary/Dandelion-inspired search) adjusts learning and control hyper parameters to speed convergence and prevent local maxima, the DRRL agent uses sequence modelling (LSTM) and an actor-critic architecture to learn temporally consistent control policies from real-world and synthetic irradiance profiles. This research suggests a hybrid intelligent MPPT architecture that combines a high- fidelity digital twin environment, metaheuristic-based hyperparameter optimization, and a deep recurrent reinforcement learning (DRL) controller in order to address these issues. To provide adaptive and reliable control decisions, the DRL agent models temporal dependencies in irradiance, temperature, and voltage–current dynamics using an actor–critic architecture improved by Long Short-Term Memory (LSTM). To improve tracking accuracy and convergence speed while avoiding local optima, a population- based metaheuristic optimizer that incorporates evolutionary and Dandelion-inspired techniques is used to automatically adjust learning rates, exploration strategies, and reward parameters. For safe offline training and transfer learning to hardware-in-the-loop (HIL) configurations, a digital twin of the PV string and power-electronics interface is utilized. In comparison to traditional and modern ML-based MPPT techniques, extensive simulation studies and HIL experiments under standardized partial-shading situations show that the suggested framework enhances tracking efficiency, decreases transient settling time, and mitigates steady-state oscillations. Lastly, we highlight future work directions and address sample efficiency, safe exploration, and on-board implementation limits.
Keywords – Deep reinforcement learning, Partial shading, MPPT, photovoltaic systems, and metaheuristic optimization.
Citation:
Refrences:
- Wadehra A, Bhalla S, Jaiswal V, Rana KP, Kumar V. A deep recurrent reinforcement learning approach for enhanced MPPT in PV systems. Applied Soft Computing. 2024 Sep 1; 162:111728.
- Phan BC, Lai YC, Lin CE. A deep reinforcement learning-based MPPT control for PV systems under partial shading condition. Sensors. 2020 May 27;20(11):3039.
- Leelavathi M, Suresh KV. Enhancing MPPT in partially shaded PV modules: a novel approach using adaptive reinforcement learning with neural network architecture. Bulletin of the Polish Academy of Sciences. Technical Sciences. 2024;72(4).
- Wei C, Zhang Z, Qiao W, Qu L. An adaptive network-based reinforcement learning method for MPPT control of PMSG wind energy conversion systems. IEEE Transactions on Power Electronics. 2016 Jan 4;31(11):7837-48.
- Ghazi GA, Al-Ammar EA, Hasanien HM, Ko W, Park J, Kim D, Ullah Z. Dandelion optimizer-based reinforcement learning techniques for MPPT of grid-connected photovoltaic systems. IEEE Access. 2024 Mar 18; 12:42932-48.
- Nwachukwu S. Deep Reinforcement Learning-Based Approaches for the MPPT Control of Standalone Solar PV Systems . Cape Town (ZA): University of Cape Town; 2024
- Nwachukwu SE, Folly KA, Awodele KO, Chen Q. Reinforcement Learning Approaches to PV System Maximum Power Point Tracking Control Under Partial Shading Conditions: A Review. IEEJ Transactions on Power and Energy. 2025 Aug 1;145(8):508-21.
- Trimboli M, Avila L, Rahmani-Andebili M. Reinforcement learning techniques for MPPT control of PV system under climatic changes. InApplications of Artificial Intelligence in Planning and Operation of Smart Grids 2022 Mar 27 (pp. 31-73). Cham: Springer International Publishing.
- Roy B, Adhikari S, Datta S, Devi KJ, Devi AD, Ustun TS. Harnessing deep learning for enhanced MPPT in solar PV systems: an LSTM approach using Real-World data. Electricity. 2024 Nov 4;5(4):843- 60.
- Salman D, Elmi YK, Isak AM, Sheikh-Muse A. Evaluation of MPPT Algorithms for Solar PV Systems with Machine Learning and Metaheuristic Techniques. Mathematical Modelling of Engineering Problems. 2025 Jan 1;12(1).
- Artetxe E, Uralde J, Barambones O, Calvo I, Martin I. Maximum power point tracker controller for solar photovoltaic based on reinforcement learning agent with a digital twin. Mathematics. 2023 May 5;11(9):2166.
- Albert JR, Kaliannan T, Singaram G, Sehar FI, Periasamy M, Kuppusamy S. A remote diagnosis using variable fractional order with reinforcement controller for solar-MPPT intelligent system. InPhotovoltaic Systems 2022 Mar 7 (pp. 45-64). CRC press.
- Shuvo SS, Gebremariam H, Yilmaz Y. Deep reinforcement learning based optimal perturbation for mppt in photovoltaics. In2021 North American Power Symposium (NAPS) 2021 Nov 14 (pp. 1-6). IEEE.
- Bavarinos K, Dounis A, Kofinas P. Maximum power point tracking based on reinforcement learning using evolutionary optimization algorithms. Energies. 2021 Jan 9;14(2):335. 14. Yap KY, Sarimuthu CR, Lim JM. Artificial intelligence based MPPT techniques for solar power system: A review. Journal of Modern Power Systems and Clean Energy. 2020 Oct 22;8(6):1043-59.
- Boubaker O. MPPT techniques for photovoltaic systems: a systematic review in current trends and recent advances in artificial intelligence. Discover Energy. 2023 Dec 6;3(1):9.
- Bag A, Puhan PS, Kumar TA. An adaptive Reinforcement Learning MPPT technique applied to standalone solar PV system. InElectric Vehicle Propulsion Drives and Charging Systems 2024 Jun 18 (pp. 236-254). CRC Press.
- El Telbany ME, Youssef A, Zekry AA. Intelligent techniques for MPPT control in photovoltaic systems: A comprehensive review. In2014 4th International Conference on Artificial Intelligence with Applications in Engineering and Technology 2014 Dec 3 (pp. 17-22). IEEE.
- Pallekonda AK, Muthukrishnan S, Meenakshi B, Saravanan R. Intelligent MPPT control for wind energy conversion systems based on reinforcement learning. InJournal of Physics: Conference Series 2021 Jul (Vol. 1964, No. 4, p. 042076). IOP Publishing.
- Bollipo RB, Mikkili S, Bonthagorla PK. Hybrid, optimal, intelligent and classical PV MPPT techniques: A review. CSEE Journal of Power and Energy Systems. 2020 Aug 19;7(1):9-33.
- Wei C, Zhang Z, Qiao W, Qu L. An adaptive network-based reinforcement learning method for MPPT control of PMSG wind energy conversion systems. IEEE Transactions on Power Electronics. 2016 Jan 4;31(11):7837-48.
- Ram JP, Babu TS, Rajasekar N. A comprehensive review on solar PV maximum power point tracking techniques. Renewable and Sustainable Energy Reviews. 2017 Jan 1; 67:826-47.
- Bendib B, Belmili H, Krim F. A survey of the most used MPPT methods: Conventional and advanced algorithms applied for photovoltaic systems. Renewable and Sustainable Energy Reviews. 2015 May 1; 45:637-48.
- Mirza AF, Ling Q, Javed MY, Mansoor M. Novel MPPT techniques for photovoltaic systems under uniform irradiance and Partial shading. Solar Energy. 2019 May 15; 184:628-48.