View in EDS

Implementation of a novel enhanced hybrid multi-objective osprey optimization algorithm for off-grid hybrid system sizing.

Saved in:
Bibliographic Details
Title: Implementation of a novel enhanced hybrid multi-objective osprey optimization algorithm for off-grid hybrid system sizing.
Authors: Bandopadhyay, Joy, Roy, Provas Kumar
Source: Evolutionary Intelligence; Oct2025, Vol. 18 Issue 5, p1-39, 39p
Abstract: A novel enhanced hybrid multi-objective osprey optimization algorithm (EHMOOOA) is proposed in this research work. It is used here for the optimum sizing of an off-grid composite renewable energy sources-based framework comprising a photovoltaic (PV) system, wind turbine generators (WTs), and battery energy storage system (BESS). The system is modelled to supply the administrative block of Kalyani Government Engineering College, Kalyani, India. The novelties include the integration of the quasi-oppositional-based learning mechanism, composite solution-oriented differential evolution algorithm (CSODEA), and Brownian motion concept into the osprey optimization algorithm (OOA). The contribution lies in the development of two novel mutation mechanisms, three new scaling coefficients and advanced extraction strategies for the core vector and the primary predecessor solutions of the difference vectors. The final stage of the exploitation process presents a novel scheme for the location upgrade of the solutions employing Brownian motion. A mathematical configuration has been built to minimize three primary fitness functions: loss of power supply probability (LPSP), levelized cost of electrical energy (LCOE), and net present cost (NPC). For validation purposes, the performances of the proposed hybrid meta-heuristic algorithm are compared with five other powerful optimization algorithms. Eight standard CEC benchmark functions and three CEC2020 actual scenario-constrained optimization problem functions are employed here for the evaluation of the proposed algorithm. The suggested composite methodology yielded the lowest and optimum values of LCOE (0.3029 $/kWh), NPC (0.9184e+05 $), and LPSP (0.000541). The optimum contributions made by PV, WT, and BESS are 80.9%, 10%, and 9.1%, respectively. Validation of the suggested technique has been further carried out by conducting statistical performance analysis through standard deviations. The proposed technique yielded the lowest standard deviation values: 7.4330e 04 (LCOE), 376.8249 (NPC), and 1.9901e 06 (LPSP), respectively. Further, to validate the outcomes, the Shapiro-Wilk test is conducted, followed by the Welch-like ANOVA and Games-Howell post-hoc tests. Moreover, a sensitivity test is performed to evaluate the robustness of EHMOOOA. [ABSTRACT FROM AUTHOR]
Copyright of Evolutionary Intelligence is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Complementary Index
Description
Abstract:A novel enhanced hybrid multi-objective osprey optimization algorithm (EHMOOOA) is proposed in this research work. It is used here for the optimum sizing of an off-grid composite renewable energy sources-based framework comprising a photovoltaic (PV) system, wind turbine generators (WTs), and battery energy storage system (BESS). The system is modelled to supply the administrative block of Kalyani Government Engineering College, Kalyani, India. The novelties include the integration of the quasi-oppositional-based learning mechanism, composite solution-oriented differential evolution algorithm (CSODEA), and Brownian motion concept into the osprey optimization algorithm (OOA). The contribution lies in the development of two novel mutation mechanisms, three new scaling coefficients and advanced extraction strategies for the core vector and the primary predecessor solutions of the difference vectors. The final stage of the exploitation process presents a novel scheme for the location upgrade of the solutions employing Brownian motion. A mathematical configuration has been built to minimize three primary fitness functions: loss of power supply probability (LPSP), levelized cost of electrical energy (LCOE), and net present cost (NPC). For validation purposes, the performances of the proposed hybrid meta-heuristic algorithm are compared with five other powerful optimization algorithms. Eight standard CEC benchmark functions and three CEC2020 actual scenario-constrained optimization problem functions are employed here for the evaluation of the proposed algorithm. The suggested composite methodology yielded the lowest and optimum values of LCOE (0.3029 $/kWh), NPC (0.9184e+05 $), and LPSP (0.000541). The optimum contributions made by PV, WT, and BESS are 80.9%, 10%, and 9.1%, respectively. Validation of the suggested technique has been further carried out by conducting statistical performance analysis through standard deviations. The proposed technique yielded the lowest standard deviation values: 7.4330e 04 (LCOE), 376.8249 (NPC), and 1.9901e 06 (LPSP), respectively. Further, to validate the outcomes, the Shapiro-Wilk test is conducted, followed by the Welch-like ANOVA and Games-Howell post-hoc tests. Moreover, a sensitivity test is performed to evaluate the robustness of EHMOOOA. [ABSTRACT FROM AUTHOR]
ISSN:18645909
DOI:10.1007/s12065-025-01083-1