APPLICATION OF DISTRIBUTED ALGORITHM IN A MICROGRID FOR OPTIMAL ENERGY MANAGEMENT AND RAPID CONVERGENCE

An intelligent energy management system is used as a powerful tool for energy management on the demand side and production units. Optimal energy management in microgrids is usually formulated as a nonlinear optimization problem. Due to the nonlinear and discrete nature of the problem, solving it centrally requires a high volume of computations in the central microgrid controller. This paper proposes the distributed energy management strategy in the microgrid with two distributed methods, the alternating direction method of multiplier and the predictor-corrector proximal multiplier so that the central controller and local controllers jointly optimize a single program. The proposed distributed algorithms on the sample microgrid are investigated and the performance of the algorithms is compared through a case study. The results show that the proposed distributed methods reduce operating costs. The simulation results show better efficiency and faster convergence of the distributed methods than the centralized method. Also, the alternating direction method of the multiplier method with less repetition and lower operating cost than the predictor-corrector proximal multiplier method has optimized the main problem.