To optimize PI parameters of the constant power controller of a microgrid inverter, an online optimization algorithm of PI parameters based on the multi-strategy adaptive fruit fly (ppmaFOA) optimization, which is itself based on population partition is proposed. According to the fitness value, the fruit fly population is divided into three areas, and a multi-strategy update mode is proposed because of the difference in the performance of the fruit flies in each area: local fine search in area I ensures that the population does not degenerate, area II adjusts the diversity and convergence of the algorithm adaptively according to the environment in which the algorithm is located, and area III guides fruit flies to accelerate convergence. All this improves the development and exploration capabilities of the algorithm. An experiment uses five comparison algorithms and uses the micro-grid constant power control system to perform experiments to verify the performance of the proposed algorithm. The inverter response speed after optimization by the ppmaFOA algorithm is faster, the error smaller, and the output more stable. This work is supported by the National Natural Science Foundation of China (No. U1804143) and the Science and Technology Key Project of Henan Province (No. 212102210146).