Laboratoire de Génie Informatique et d’Automatique de l’Artois

This PhD project is situated within the context of the energy transition, in which photovoltaic systems play a key role due to their sustainability and low environmental impact. However, their efficiency strongly depends on the ability to continuously track the maximum power point (MPPT), a task made challenging by real operating conditions such as rapid irradiance variations, temperature fluctuations, and especially partial shading. The latter leads to multimodal power–voltage characteristics with multiple local maxima, which significantly complicates the identification of the global maximum.

Conventional MPPT methods [1,2,3,4], although simple and widely used, exhibit significant limitations in dynamic environments [5,6,7,8], including oscillations, slow convergence, and a high risk of being trapped in local maxima. Intelligent approaches, such as fuzzy logic and artificial neural networks, have provided some improvements, but they remain highly dependent on training data and lack robustness when confronted with previously unseen conditions.

In light of these limitations, recent advances in artificial intelligence, particularly in reinforcement learning and optimization techniques, offer promising perspectives for the design of more robust and adaptive MPPT controllers. The main objective of this thesis is to develop hybrid approaches that combine these techniques in order to optimize power extraction from photovoltaic systems, even under partial shading and highly variable climatic conditions.

The proposed methodology is structured around four phases: a comprehensive state-of-the-art review, advanced modeling of the photovoltaic system, the development of intelligent AI- and optimization-based controllers, and rigorous experimental validation. Performance will be evaluated in terms of energy efficiency, tracking speed, stability, and robustness, and compared with conventional methods. The expected outcomes aim at the design of a new generation of intelligent MPPT controllers.

References

[1] N. Femia, G. Petrone, G. Spagnuolo, M. Vitelli, "Optimization of perturb and observe maximum power point tracking method," IEEE Transactions on power electronics, n°20, pp. 963-973, 2005.

[2] M. Sarvi, S. Ahmadi, S. Abdi, "A PSO-based maximum power point tracking for photovoltaic systems under environmental and partially shaded conditions," Progress in Photovoltaics: Research and Applications, vol. 23, n°2, pp. 201–214, 2015.

[3] O. Wasynezuk, "Dynamic behavior of a class of photovoltaic power systems," IEEE Transactions on power apparatus and systems, PAS-102, pp. 3031-3037, 1983.

[4] K. H. Hussein, I. Muta, T. Hoshino, and M. Osakada, "Maximum photovoltaic power tracking: an algorithm for rapidly changing atmospheric conditions," IEE Proc. Gen., Transm. & Distr., vol. 142, no. 1, pp. 59–64, 1995.

[5] M. S. Endiz, G. Gökkuş, A. E. Coşgun, H. Demir, "A Review of Traditional and Advanced MPPT Approaches for PV Systems Under Uniformly Insolation and Partially Shaded Conditions," Applied Sciences, vol. 15, n°3:1031, 2025.

[6] J. Ahmed, Z. Salam, "An improved perturb and observe (P&O) maximum power point tracking (MPPT) algorithm for higher efficiency," Applied Energy, n°150, pp. 97–108, 2015.

[7] R. Alik, A. Jusoh, "An enhanced P&O checking algorithm MPPT for high tracking efficiency of partially shaded PV module," Sol. Energy, n°163, pp. 570–580, 2018.

[8] H. Shahid, M. Kamran, Z. Mehmood, M. Y. Saleem, M. Mudassar, K. Haider, "Implementation of the novel temperature controller and incremental conductance MPPT algorithm for indoor photovoltaic system," Sol. Energy, n°163, pp. 235–242, 2018.