Reliability-aware global maximum power point tracking for partially shaded photovoltaic systems

Under partial shading conditions (PSCs), conventional global maximum power point tracking (GMPPT) methods for photovoltaic (PV) systems mainly pursue instantaneous power maximization, while often neglecting thermal stress and lifetime degradation induced by frequent bypass-diode conduction. To address this problem, this paper proposes an improved optimized voltage search (IOVS) strategy integrating analytical P-V reconstruction with electro-thermal-lifetime cooperative decision-making. In the startup stage, a few feature-voltage samples are used to reconstruct the global operating state of the PV array, compress the effective search interval, and identify bypass-diode conduction states. An adaptive voltage search is then performed to rapidly locate the GMPP. Furthermore, a hierarchical hard/soft constraint decision module coordinates output power, junction temperature, and cumulative lifetime damage. Simulation results show that, under four static PSC cases, IOVS achieves 99.92%-99.99% tracking efficiency, with power loss below 0.1% and tracking time of 0.25–0.29 s. Under step-changing shading transitions, its tracking efficiency remains 99.8%-100%, while the minimum efficiencies of some compared algorithms drop to 36.7%-55.1%. For reliability-oriented operation, the proposed module can block bypass-diode conduction with only about 3% power sacrifice, eliminating 2.58 W conduction loss and the associated junction-temperature rise. Under continuous moving-cloud shading, although energy tracking efficiency decreases from 95.73% to 89.97% compared with OVS, cumulative lifetime damage is reduced by 58.12%, equivalent diode lifetime increases to 2.39 times, and the maximum junction temperature is limited to 28.84 °C. These results demonstrate that IOVS achieves accurate GMPP tracking while mitigating thermal aging and improving long-term reliability.