Research Area:  Machine Learning
Grid-connected mode of microgrid (MG) operation is the most economical and reliable mode of service because the grid participates in the buying and selling of power with the MG, which reduces the generation cost of the MG system. Various types of electricity market pricing strategies are available in the literature with which the grid buys and sells power to the MG system. These strategies are implemented in turns on two different low voltage (LV) MG systems, and the generation cost of the systems are recorded to distinguish the cheapest strategy among them. A novel hybrid of recently developed superior optimization algorithms viz. grey wolf optimizer (GWO), sine cosine algorithm (SCA), and crow search algorithm (CSA) is used to minimize the overall generation cost of the MG system. It was evident from the results that generation cost was minimum when electricity market pricing was done using Strategy 2. Also, active participation of the grid incurred 15% savings in the system generation cost for the same strategy compared to the case when the grid was operating passively. Furthermore, statistical analysis corroborates to the superiority of the proposed hybrid MGWO-SCA-CSA over other algorithms used.
Keywords:  
Author(s) Name:  Bishwajit Dey, Biplab Bhattacharyya, Ramesh Devarapalli
Journal name:  International Journal of Intelligent Systems
Conferrence name:  
Publisher name:  Wiley
DOI:  10.1002/int.22326
Volume Information:  Volume36, Issue2 February 2021
Paper Link:   https://onlinelibrary.wiley.com/doi/abs/10.1002/int.22326