Abstract: In photovoltaic power generation systems, partial shading effects often cause significant changes in power output characteristics, presenting itself as a multi-peak and complex curve. Introducing the ICSA algorithm to adjust the step size Pa can better adapt to the characteristics of the PV system; using a triangular walking strategy instead of the traditional random walking strategy allows the population of the intelligent optimization algorithm to search finely in the surrounding area while approaching the optimal position, significantly enhancing the algorithm"s local exploration capability; incorporating the Levy flight function strengthens the algorithm"s ability in local power search, ensuring that the algorithm can ultimately stably converge to the global maximum power point.