[Objective] With the continuous development and improvement of microgrid technology

the use of renewable energy is increasingly valued. The impact of source-load uncertainty on microgrid security as well as economic operation cannot be ignored

and in order to solve the problem of optimal economic operation of microgrid

a hybrid Harris Hawk optimization (HHHO) algorithm is proposed. [Methods] Firstly

an optimal microgrid scheduling model considering source-load uncertainty was developed. Secondly

a HHHO algorithm combining elite strategy and differential learning mechanism was proposed. The elite strategy improved the convergence speed of the algorithm by retaining the elite individuals in the population. The differential learning mechanism optimized the accuracy of the algorithm’s search by enhancing the dissemination of valid information. Then

a hybrid scene reduction method based on Latin hypercubic sampling-probabilistic distance was proposed

and the reduced equivalent scene was used as the basis for the economic operation of the microgrid. Finally

the superiority of the proposed HHHO algorithm was verified by test functions and scene simulations. [Results] The test function experimental results showed that the proposed HHHO algorithm had high solution accuracy and fast convergence speed. Simulation results in scenarios such as wind and photovoltaic fluctuations showed that the HHHO algorithm was highly superior and significantly reduced the overall operating cost of the microgrid. [Conclusion] The proposed HHHO algorithm enables optimal economic operation of microgrids and provides theoretical support for long-term economic operation of microgrid.