Experimental validation of irradiance-sensorless self-adaptive GMPPT algorithm for PV systems under partial shading

Abstract Photovoltaic (PV) systems are among the most promising renewable energy sources, yet their efficiency is critically compromised under partial shading conditions (PSC). Under PSC, the power-voltage (P-V) characteristic exhibits multiple peaks, causing conventional maximum power point tracking (MPPT) algorithms to fail, which leads to significant power losses that undermine economic viability. Addressing this challenge, this work experimentally validates a deterministic, Irradiance Sensorless Self-adaptive Global MPPT (ISS-GMPPT) algorithm, bridging the critical gap between theoretical simulation and practical hardware deployment. The controller was implemented on a dSPACE DS1104 real-time platform interfaced with a custom DC-DC boost converter, a dynamic resistive load, and a physical two-panel PV array. Experimental results demonstrate that the ISS-GMPPT successfully identifies the true GMPP across standard shading profiles without complex parameter tuning. It achieves a high tracking efficiency of up to 99.3% alongside a rapid operational convergence speed of $$2.3\text { s}$$ . Furthermore, a comprehensive analytical loss model validates the hardware design, yielding an electrical converter efficiency peaking at 98.4%. Under dynamic load disturbances, the control loop exhibits highly effective rejection, restoring the optimal operating point in under $$0.5\text { s}$$ without triggering unnecessary global rescans. Ultimately, this validation establishes the practical implementability, low computational complexity, and real-world effectiveness of the ISS-GMPPT algorithm for grid-tied and microgrid PV applications.