Data-driven optimization for microgrid control under distributed energy resource variability

Akhilesh Mathur; Ruchi Kumari; V P Meena; V P Singh; Ahmad Taher Azar; Ibrahim A Hameed

doi:10.1038/s41598-024-58767-4

Data-driven optimization for microgrid control under distributed energy resource variability

Sci Rep. 2024 May 11;14(1):10806. doi: 10.1038/s41598-024-58767-4.

Authors

Akhilesh Mathur¹, Ruchi Kumari¹, V P Meena^{2

3}, V P Singh¹, Ahmad Taher Azar^{4

5

6}, Ibrahim A Hameed⁷

Affiliations

¹ Department of Electrical Engineering, Malaviya National Institute of Technology, Jaipur, Rajasthan, 302017, India.
² Department of Electrical Engineering, Malaviya National Institute of Technology, Jaipur, Rajasthan, 302017, India. vmeena1@ee.iitr.ac.in.
³ Department of Electrical and Electronics Engineering, Amrita School of Engineering, Amrita Vishwa Vidyapeetham, Bengaluru, India. vmeena1@ee.iitr.ac.in.
⁴ College of Computer and Information Sciences, Prince Sultan University, 11586, Riyadh, Saudi Arabia. aazar@psu.edu.sa.
⁵ Automated Systems and Soft Computing Lab (ASSCL), Prince Sultan University, Riyadh, Saudi Arabia. aazar@psu.edu.sa.
⁶ Faculty of Computers and Artificial Intelligence, Benha University, Benha, 13518, Egypt. aazar@psu.edu.sa.
⁷ Department of ICT and Natural Sciences, Norwegian University of Science and Technology, Larsgardsvegen, 2, 6009, Alesund, Norway. ibib@ntnu.no.

Abstract

The integration of renewable energy resources into the smart grids improves the system resilience, provide sustainable demand-generation balance, and produces clean electricity with minimal leakage currents. However, the renewable sources are intermittent in nature. Therefore, it is necessary to develop scheduling strategy to optimise hybrid PV-wind-controllable distributed generator based Microgrids in grid-connected and stand-alone modes of operation. In this manuscript, a priority-based cost optimization function is developed to show the relative significance of one cost component over another for the optimal operation of the Microgrid. The uncertainties associated with various intermittent parameters in Microgrid have also been introduced in the proposed scheduling methodology. The objective function includes the operating cost of CDGs, the emission cost associated with CDGs, the battery cost, the cost of grid energy exchange, and the cost associated with load shedding. A penalty function is also incorporated in the cost function for violations of any constraints. Multiple scenarios are generated using Monte Carlo simulation to model uncertain parameters of Microgrid (MG). These scenarios consist of the worst as well as the best possible cases, reflecting the microgrid's real-time operation. Furthermore, these scenarios are reduced by using a k-means clustering algorithm. The reduced procedures for uncertain parameters will be used to obtain the minimum cost of MG with the help of an optimisation algorithm. In this work, a meta-heuristic approach, grey wolf optimisation (GWO), is used to minimize the developed cost optimisation function of MG. The standard LV Microgrid CIGRE test network is used to validate the proposed methodology. Results are obtained for different cases by considering different priorities to the sub-objectives using GWO algorithm. The obtained results are compared with the results of Jaya and PSO (particle swarm optimization) algorithms to validate the efficacy of the GWO method for the proposed optimization problem.

Keywords: Grey-Wolf optimization; Jaya algorithm; K-mean clustering; Microgrids; Monte Carlo simulation; Optimal scheduling; Probability distribution function; Stochastic process.