Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed 
Buck–Boost Converter

Olusesan Orebanjo; Stephen Nwachukwu Ukagu; Dennis Okwuchukwu Okonkwo; Osamudiamen Nelson Amasowomwan; Ikenna Augustine Onyegbadue; Dedacus Nnadozie Ohaegbuchi; Lucky Olatunde Odejimi

doi:doi:10.11648/j.ajece.20261002.12

Research Article |

| Peer-Reviewed

Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed Buck–Boost Converter

Olusesan Orebanjo^*

, Stephen Nwachukwu Ukagu

, Dennis Okwuchukwu Okonkwo, Osamudiamen Nelson Amasowomwan, Ikenna Augustine Onyegbadue, Dedacus Nnadozie Ohaegbuchi, Lucky Olatunde Odejimi

Published in American Journal of Electrical and Computer Engineering (Volume 10, Issue 2)

Received: 24 September 2026 Accepted: 8 October 2026 Published: 28 October 2026

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Maximum power point tracking (MPPT) is essential for efficient photovoltaic (PV) operation under varying irradiance and temperature. Conventional methods such as Incremental Conductance (INC) remain attractive for their simplicity, yet fixed proportional–integral (PI) gains limit dynamic performance when environmental conditions change rapidly. Under partial shading the power–voltage characteristic can also contain multiple local maxima, making reliable global tracking more difficult. This paper proposes a hybrid MPPT architecture in which a tabular Q-learning agent adaptively tunes the PI gains of a PV-fed buck–boost converter while INC retains responsibility for generating the operating-point reference. The learning state is defined from PV voltage, power and their increments; actions are bounded gain steps constrained by small-signal stability analysis; and the reward balances power improvement against excessive gain movement. Because the learning agent only supervises the inner-loop gains rather than replacing the entire control law, the design preserves the interpretability of classical INC while introducing operating-condition-dependent adaptation. The controller is evaluated in simulation against fixed-gain INC and direct-duty-cycle Q-learning under six irradiance profiles, with additional temperature, partial-shading, ablation and hyperparameter-sensitivity tests. Results demonstrate that the hybrid architecture achieves competitive tracking performance while maintaining adaptive PI gains within stability bounds. Ablation confirms that joint adaptation of both gains outperforms single-gain cases, and the sensitivity study identifies a robust hyper-parameter region. The study confirms the feasibility of lightweight, hierarchical reinforcement learning for adaptive PI regulation in PV converters and provides a practical framework for online simulation-based gain tuning under dynamic environmental conditions.

Published in	American Journal of Electrical and Computer Engineering (Volume 10, Issue 2)
DOI	10.11648/j.ajece.20261002.12
Page(s)	47-64
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2026. Published by Science Publishing Group

Keywords

Maximum Power Point Tracking, Q-Learning, Adaptive Control, Controller Gain Tuning, Hybrid Intelligent Control

References

[1]	O. Boubaker, “MPPT techniques for PV systems: A systematic review in current trends and recent advances in AI,” Discover Energy, vol. 3, Art. no. 9, 2023. https://doi.org/10.1007/s43937-023-00024-2
[2]	D. Mazumdar, C. Sain, P. K. Biswas, P. Sanjeevikumar, and B. Khan, “Overview of solar photovoltaic MPPT methods: A state of the art on conventional and artificial intelligence control techniques,” International Transactions on Electrical Energy Systems, vol. 2024, Art. no. 8363342, 2024. https://doi.org/10.1155/2024/8363342
[3]	G. Lyu, A. M. Soomar, S. H. H. Shah, S. Shaikh, and P. Musznicki, “Maximum power point tracking strategies for solar PV systems: A review of current methods and future innovations,” Results in Engineering, vol. 28, Art. no. 107227, 2025. https://doi.org/10.1016/j.rineng.2025.107227S
[4]	S. J. Yaqoob, H. Arnoos, N. T. Alwan, M. Bajaj, I. M. Alwan, M. H. A. Aljanabi, B. A. Zneid, and M. S. Geremew, “Advanced maximum power point tracking in photovoltaic systems: A comprehensive review of classical, AI-based, and metaheuristic optimization techniques,” Engineering Reports, vol. 7, no. 9, Art. no. e70404, 2025. https://doi.org/10.1002/eng2.70404
[5]	M. Sameera, M. Tariq, M. Rihan, and M. Ayan, “A comprehensive review on the application of recently introduced optimization techniques obtaining maximum power in the solar PV system,” Renewable Energy Focus, vol. 49, Art. no. 100564, 2024. https://doi.org/10.1016/j.ref.2024.100564
[6]	B. O. Oyebola and B. S. Emmanuel, “Artificial intelligence-based maximum power point tracking in smart photovoltaic systems: A systematic review,” FEDM Journal of Sustainable Design & Management, vol. 1, no. 1, pp. 20–54, 2025.
[7]	K.-Y. Chou, S.-T. Yang, and Y.-P. Chen, “Maximum power point tracking of photovoltaic system based on reinforcement learning,” Sensors, vol. 19, no. 22, Art. no. 5054, 2019. https://doi.org/10.3390/s19225054
[8]	Y. Chen, D. Lin, F. Xu, X. Li, W. Wang, and S. Ding, “Research on Q-table design for maximum power point tracking-based reinforcement learning in PV systems,” Energies, vol. 16, no. 21, Art. no. 7286, 2023. https://doi.org/10.3390/en16217286
[9]	L. F. Giraldo, J. F. Gaviria, M. I. Torres, C. Alonso, and M. Bressan, “Deep reinforcement learning using deep-Q-network for global maximum power point tracking: Design and experiments in real photovoltaic systems,” Heliyon, vol. 10, no. 21, Art. no. e37974, 2024. https://doi.org/10.1016/j.heliyon.2024.e37974
[10]	E. Artetxe, J. Uralde, O. Barambones, I. Calvo, and I. Martin, “Maximum power point tracker controller for solar photovoltaic based on reinforcement learning agent with a digital twin,” Mathematics, vol. 11, no. 9, Art. no. 2166, 2023. https://doi.org/10.3390/math11092166
[11]	S. E. Nwachukwu, K. A. Folly, K. O. Awodele, and Q. Chen, “Reinforcement learning approaches to PV system maximum power point tracking control under partial shading conditions: A review,” IEEJ Transactions on Power and Energy, vol. 145, no. 8, pp. 508–521, 2025. https://doi.org/10.1541/ieejpes.145.508
[12]	W. Jeungthanasirigool, T. Sirimaskasem, T. Boonraksa, and P. Boonraksa, “Comparison of PPO-DRL and A2C-DRL algorithms for MPPT in photovoltaic systems via buck-boost converter,” International Journal of Innovative Research and Scientific Studies, vol. 8, no. 3, pp. 2438–2453, 2025. https://doi.org/10.53894/ijirss.v8i3.7022
[13]	M. E. Girgis and N. A. Elkhateeb, “Enhancing photovoltaic MPPT with P&O algorithm performance based on adaptive PID control using exponential forgetting recursive least squares method,” Renewable Energy, vol. 237, Art. no. 121801, 2024. https://doi.org/10.1016/j.renene.2024.121801

Cite This Article

Plain Text BibTeX RIS

APA Style

Orebanjo, O., Ukagu, S. N., Okonkwo, D. O., Amasowomwan, O. N., Onyegbadue, I. A., et al. (2026). Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed Buck–Boost Converter. American Journal of Electrical and Computer Engineering, 10(2), 47-64. https://doi.org/10.11648/j.ajece.20261002.12

Copy | Download

ACS Style

Orebanjo, O.; Ukagu, S. N.; Okonkwo, D. O.; Amasowomwan, O. N.; Onyegbadue, I. A., et al. Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed Buck–Boost Converter. Am. J. Electr. Comput. Eng. 2026, 10(2), 47-64. doi: 10.11648/j.ajece.20261002.12

Copy | Download

AMA Style

Orebanjo O, Ukagu SN, Okonkwo DO, Amasowomwan ON, Onyegbadue IA, et al. Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed Buck–Boost Converter. Am J Electr Comput Eng. 2026;10(2):47-64. doi: 10.11648/j.ajece.20261002.12

Copy | Download

@article{10.11648/j.ajece.20261002.12,
  author = {Olusesan Orebanjo and Stephen Nwachukwu Ukagu and Dennis Okwuchukwu Okonkwo and Osamudiamen Nelson Amasowomwan and Ikenna Augustine Onyegbadue and Dedacus Nnadozie Ohaegbuchi and Lucky Olatunde Odejimi},
  title = {Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed 
Buck–Boost Converter},
  journal = {American Journal of Electrical and Computer Engineering},
  volume = {10},
  number = {2},
  pages = {47-64},
  doi = {10.11648/j.ajece.20261002.12},
  url = {https://doi.org/10.11648/j.ajece.20261002.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajece.20261002.12},
  abstract = {Maximum power point tracking (MPPT) is essential for efficient photovoltaic (PV) operation under varying irradiance and temperature. Conventional methods such as Incremental Conductance (INC) remain attractive for their simplicity, yet fixed proportional–integral (PI) gains limit dynamic performance when environmental conditions change rapidly. Under partial shading the power–voltage characteristic can also contain multiple local maxima, making reliable global tracking more difficult. This paper proposes a hybrid MPPT architecture in which a tabular Q-learning agent adaptively tunes the PI gains of a PV-fed buck–boost converter while INC retains responsibility for generating the operating-point reference. The learning state is defined from PV voltage, power and their increments; actions are bounded gain steps constrained by small-signal stability analysis; and the reward balances power improvement against excessive gain movement. Because the learning agent only supervises the inner-loop gains rather than replacing the entire control law, the design preserves the interpretability of classical INC while introducing operating-condition-dependent adaptation. The controller is evaluated in simulation against fixed-gain INC and direct-duty-cycle Q-learning under six irradiance profiles, with additional temperature, partial-shading, ablation and hyperparameter-sensitivity tests. Results demonstrate that the hybrid architecture achieves competitive tracking performance while maintaining adaptive PI gains within stability bounds. Ablation confirms that joint adaptation of both gains outperforms single-gain cases, and the sensitivity study identifies a robust hyper-parameter region. The study confirms the feasibility of lightweight, hierarchical reinforcement learning for adaptive PI regulation in PV converters and provides a practical framework for online simulation-based gain tuning under dynamic environmental conditions.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed 
Buck–Boost Converter
AU  - Olusesan Orebanjo
AU  - Stephen Nwachukwu Ukagu
AU  - Dennis Okwuchukwu Okonkwo
AU  - Osamudiamen Nelson Amasowomwan
AU  - Ikenna Augustine Onyegbadue
AU  - Dedacus Nnadozie Ohaegbuchi
AU  - Lucky Olatunde Odejimi
Y1  - 2026/10/28
PY  - 2026
N1  - https://doi.org/10.11648/j.ajece.20261002.12
DO  - 10.11648/j.ajece.20261002.12
T2  - American Journal of Electrical and Computer Engineering
JF  - American Journal of Electrical and Computer Engineering
JO  - American Journal of Electrical and Computer Engineering
SP  - 47
EP  - 64
PB  - Science Publishing Group
SN  - 2640-0502
UR  - https://doi.org/10.11648/j.ajece.20261002.12
AB  - Maximum power point tracking (MPPT) is essential for efficient photovoltaic (PV) operation under varying irradiance and temperature. Conventional methods such as Incremental Conductance (INC) remain attractive for their simplicity, yet fixed proportional–integral (PI) gains limit dynamic performance when environmental conditions change rapidly. Under partial shading the power–voltage characteristic can also contain multiple local maxima, making reliable global tracking more difficult. This paper proposes a hybrid MPPT architecture in which a tabular Q-learning agent adaptively tunes the PI gains of a PV-fed buck–boost converter while INC retains responsibility for generating the operating-point reference. The learning state is defined from PV voltage, power and their increments; actions are bounded gain steps constrained by small-signal stability analysis; and the reward balances power improvement against excessive gain movement. Because the learning agent only supervises the inner-loop gains rather than replacing the entire control law, the design preserves the interpretability of classical INC while introducing operating-condition-dependent adaptation. The controller is evaluated in simulation against fixed-gain INC and direct-duty-cycle Q-learning under six irradiance profiles, with additional temperature, partial-shading, ablation and hyperparameter-sensitivity tests. Results demonstrate that the hybrid architecture achieves competitive tracking performance while maintaining adaptive PI gains within stability bounds. Ablation confirms that joint adaptation of both gains outperforms single-gain cases, and the sensitivity study identifies a robust hyper-parameter region. The study confirms the feasibility of lightweight, hierarchical reinforcement learning for adaptive PI regulation in PV converters and provides a practical framework for online simulation-based gain tuning under dynamic environmental conditions.
VL  - 10
IS  - 2
ER  -

Copy | Download

Author Information

Olusesan Orebanjo

Department of Electrical and Computer Engineering, Igbinedion University, Okada, Nigeria

Contact Email

http://orcid.org/0009-0000-9736-9346
Stephen Nwachukwu Ukagu

Department of Electrical and Computer Engineering, Igbinedion University, Okada, Nigeria

http://orcid.org/0009-0008-7645-9930
Dennis Okwuchukwu Okonkwo

Department of Electrical and Computer Engineering, Igbinedion University, Okada, Nigeria
Osamudiamen Nelson Amasowomwan

Department of Electrical and Computer Engineering, Igbinedion University, Okada, Nigeria
Ikenna Augustine Onyegbadue

Department of Electrical and Computer Engineering, Igbinedion University, Okada, Nigeria
Dedacus Nnadozie Ohaegbuchi

Department of Electrical and Electronics Engineering Science, University of Johannesburg, Johannesburg, South Africa
Lucky Olatunde Odejimi

Department of Mechanical Engineering, Igbinedion University, Okada, Nigeria

Download PDF

Submit an Article

Sections

Plain Text BibTeX RIS

APA Style

Orebanjo, O., Ukagu, S. N., Okonkwo, D. O., Amasowomwan, O. N., Onyegbadue, I. A., et al. (2026). Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed Buck–Boost Converter. American Journal of Electrical and Computer Engineering, 10(2), 47-64. https://doi.org/10.11648/j.ajece.20261002.12

Copy | Download

ACS Style

Orebanjo, O.; Ukagu, S. N.; Okonkwo, D. O.; Amasowomwan, O. N.; Onyegbadue, I. A., et al. Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed Buck–Boost Converter. Am. J. Electr. Comput. Eng. 2026, 10(2), 47-64. doi: 10.11648/j.ajece.20261002.12

Copy | Download

AMA Style

Orebanjo O, Ukagu SN, Okonkwo DO, Amasowomwan ON, Onyegbadue IA, et al. Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed Buck–Boost Converter. Am J Electr Comput Eng. 2026;10(2):47-64. doi: 10.11648/j.ajece.20261002.12

Copy | Download

@article{10.11648/j.ajece.20261002.12,
  author = {Olusesan Orebanjo and Stephen Nwachukwu Ukagu and Dennis Okwuchukwu Okonkwo and Osamudiamen Nelson Amasowomwan and Ikenna Augustine Onyegbadue and Dedacus Nnadozie Ohaegbuchi and Lucky Olatunde Odejimi},
  title = {Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed 
Buck–Boost Converter},
  journal = {American Journal of Electrical and Computer Engineering},
  volume = {10},
  number = {2},
  pages = {47-64},
  doi = {10.11648/j.ajece.20261002.12},
  url = {https://doi.org/10.11648/j.ajece.20261002.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajece.20261002.12},
  abstract = {Maximum power point tracking (MPPT) is essential for efficient photovoltaic (PV) operation under varying irradiance and temperature. Conventional methods such as Incremental Conductance (INC) remain attractive for their simplicity, yet fixed proportional–integral (PI) gains limit dynamic performance when environmental conditions change rapidly. Under partial shading the power–voltage characteristic can also contain multiple local maxima, making reliable global tracking more difficult. This paper proposes a hybrid MPPT architecture in which a tabular Q-learning agent adaptively tunes the PI gains of a PV-fed buck–boost converter while INC retains responsibility for generating the operating-point reference. The learning state is defined from PV voltage, power and their increments; actions are bounded gain steps constrained by small-signal stability analysis; and the reward balances power improvement against excessive gain movement. Because the learning agent only supervises the inner-loop gains rather than replacing the entire control law, the design preserves the interpretability of classical INC while introducing operating-condition-dependent adaptation. The controller is evaluated in simulation against fixed-gain INC and direct-duty-cycle Q-learning under six irradiance profiles, with additional temperature, partial-shading, ablation and hyperparameter-sensitivity tests. Results demonstrate that the hybrid architecture achieves competitive tracking performance while maintaining adaptive PI gains within stability bounds. Ablation confirms that joint adaptation of both gains outperforms single-gain cases, and the sensitivity study identifies a robust hyper-parameter region. The study confirms the feasibility of lightweight, hierarchical reinforcement learning for adaptive PI regulation in PV converters and provides a practical framework for online simulation-based gain tuning under dynamic environmental conditions.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Adaptive Hybrid MPPT with Tabular Reinforcement Learning for Adaptive PI Gain Tuning in a PV-Fed 
Buck–Boost Converter
AU  - Olusesan Orebanjo
AU  - Stephen Nwachukwu Ukagu
AU  - Dennis Okwuchukwu Okonkwo
AU  - Osamudiamen Nelson Amasowomwan
AU  - Ikenna Augustine Onyegbadue
AU  - Dedacus Nnadozie Ohaegbuchi
AU  - Lucky Olatunde Odejimi
Y1  - 2026/10/28
PY  - 2026
N1  - https://doi.org/10.11648/j.ajece.20261002.12
DO  - 10.11648/j.ajece.20261002.12
T2  - American Journal of Electrical and Computer Engineering
JF  - American Journal of Electrical and Computer Engineering
JO  - American Journal of Electrical and Computer Engineering
SP  - 47
EP  - 64
PB  - Science Publishing Group
SN  - 2640-0502
UR  - https://doi.org/10.11648/j.ajece.20261002.12
AB  - Maximum power point tracking (MPPT) is essential for efficient photovoltaic (PV) operation under varying irradiance and temperature. Conventional methods such as Incremental Conductance (INC) remain attractive for their simplicity, yet fixed proportional–integral (PI) gains limit dynamic performance when environmental conditions change rapidly. Under partial shading the power–voltage characteristic can also contain multiple local maxima, making reliable global tracking more difficult. This paper proposes a hybrid MPPT architecture in which a tabular Q-learning agent adaptively tunes the PI gains of a PV-fed buck–boost converter while INC retains responsibility for generating the operating-point reference. The learning state is defined from PV voltage, power and their increments; actions are bounded gain steps constrained by small-signal stability analysis; and the reward balances power improvement against excessive gain movement. Because the learning agent only supervises the inner-loop gains rather than replacing the entire control law, the design preserves the interpretability of classical INC while introducing operating-condition-dependent adaptation. The controller is evaluated in simulation against fixed-gain INC and direct-duty-cycle Q-learning under six irradiance profiles, with additional temperature, partial-shading, ablation and hyperparameter-sensitivity tests. Results demonstrate that the hybrid architecture achieves competitive tracking performance while maintaining adaptive PI gains within stability bounds. Ablation confirms that joint adaptation of both gains outperforms single-gain cases, and the sensitivity study identifies a robust hyper-parameter region. The study confirms the feasibility of lightweight, hierarchical reinforcement learning for adaptive PI regulation in PV converters and provides a practical framework for online simulation-based gain tuning under dynamic environmental conditions.
VL  - 10
IS  - 2
ER  -

Copy | Download