Abstract: It is difficult for traditional proportional integral derivative(PID) control and model theory control methods to deal with the complex and changeable operation scenarios of microgrid under the background of new power system, therefore, a fuzzy neural network based intelligent control method for load storage coordination of microgrid is proposed. Firstly, the input and output variables of fuzzy control of the microgrid are determined. The control experience of the microgrid is summarized into a fuzzy rule table for the purpose of stabilizing the fluctuation of net load and reducing the frequency of charging and discharging of energy storage. The membership function center, width and output weight of the fuzzy control model are modified by using the deep learning algorithm of neural network, so as to improve the adaptive ability of the model. Thus, the power control factors of adjustable load and energy storage are formulated. Then, to solve the problem that the load regulation demand output of fuzzy neural network control is distributed among various adjustable loads, a load control priority selection method based on the flexible supply index ranking is proposed. Finally, the energy storage unit and the control strategy of adjustable loads are formulated. The simulation results of a typical microgrid system demonstrate that the adjustable load and energy storage control strategies, developed by the proposed method, can not only prevent the frequent and excessive charge or discharge of energy storage, but also effectively reduce the random disturbance of grid-connected power to the superior grid in the grid-connected state, and effectively stabilize the power fluctuations of the system in the isolated island state, thus improving the operational stability of the system.