Efficient and accurate estimation parameters of the PV model from real experimental data is crucial for modeling, increasing investments, and determining the actual performance before installing the PV panels. Recently, several stochastic methods have been proposed to extract the parameters of the PV module/cells optimally. However, some of these approaches have restrictions in terms of exploration and exploitation capacities individually or combined due to their stochastic searching strategies. In this research work, Marine Predator Algorithm (MPA) is combined with Lambert W function to tackle the parameters extraction optimization problem for the single diode and double diode PV models and called (MPALW). The performance of the MPALW algorithm is tested by using real experimental data at seven sunlight and temperature settings. The MPALW shows an excellent agreement with obtained experimental data compared with other well-published algorithms. The findings of this research show that the MPALW algorithm can be utilized for real engineering applications such as smart grids, energy sector, and fault error detection.