A day-ahead joint energy management and battery sizing framework based on θ-modified krill herd algorithm for a renewable energy-integrated microgrid

The penetration level of intermittent power generation into power systems has been substantial during recent years, which in turn highlights the need for installing storage systems. Renewable energy sources have been widely integrated into distribution systems and microgrids. One effective solution would be utilizing battery energy storage systems, which can provide the system with various merits like ancillary services and enhanced power quality, mainly due to their high power density and fast response. Accordingly, the problem of resource scheduling of microgrids with volatile power generation and storage systems needs to be further studied, and an effective model should be presented. In this respect, this paper investigates the problem of day-ahead operation of a grid-connected MG, integrated with distributed generation units and storage systems. The problem has been modeled an optimization problem while the objective has been assigned to the model as the total cost minimization, subject to different constraints, both system constraints and assets' constraints. Such constraints further complicate the original problem and an efficient solution method is required to tackle the problem as a large-scale optimization one. Thus, theta-modified krill herd approach is employed to solve the problem and provide the decision maker with an efficient solution. The simulation has also been conducted using a test MG and the results, obtained have been validated by comparing the results obtained from the presented method and those ones, derived from some well-known optimization algorithms. (C) 2020 Elsevier Ltd. All rights reserved.

Automated summary

This paper investigates the day-ahead operation of a grid-connected microgrid with distributed generation units and storage systems, and utilizes the theta-modified krill herd approach to provide an efficient solution. The simulation results were validated by comparing with results from well-known optimization algorithms.