An Integrated Optimization and AI-Based Framework for Performance Enhancement of Weak Distribution Grids
Authors
Department of Electrical and Electronics Engineering, University of Maiduguri, Borno Sate, Nigeria (Nigeria)
Department of Electrical and Electronics Engineering, University of Maiduguri, Borno Sate, Nigeria (Nigeria)
Department of Electrical and Electronics Engineering, University of Maiduguri, Borno Sate, Nigeria (Nigeria)
Department of Electrical and Electronics Engineering, University of Maiduguri, Borno Sate, Nigeria (Nigeria)
Article Information
DOI: 10.51583/IJLTEMAS.2026.151000147
Subject Category: Engineering and Technology
Volume/Issue: 15/10 | Page No: 1811-1826
Publication Timeline
Submitted: 2026-10-25
Accepted: 2026-10-30
Published: 2026-11-20
Abstract
Weak electric distribution networks in many developing countries continue to experience problems such as voltage instability, high technical losses, feeder overloading, and limited operational flexibility. The increasing use of renewable energy sources, particularly solar photovoltaic and wind systems, has added further challenges because of the variability and uncertainty associated with their power generation. This study proposes an integrated optimization and intelligent control framework for improving the performance of weak distribution networks. The framework involves three main stages: optimal placement and sizing of energy storage systems (ESS) using the Non-dominated Sorting Genetic Algorithm II (NSGA-II), distribution network reconfiguration using a hybrid Particle Swarm Optimization–Genetic Algorithm (PSO-GA), and an Artificial Intelligence (AI)-based predictive model for rapid identification of suitable feeder switch states under different operating conditions.
The proposed framework was evaluated using the standard IEEE 33-bus radial distribution system under different load levels and renewable generation conditions. The simulation results showed noticeable improvements in network performance. Active power losses were reduced by 41.8%, while the minimum bus voltage increased from 0.903 p.u. under the base condition to 0.957 p.u. after applying the proposed framework. This improvement indicates better voltage regulation and overall network operating conditions. The AI-based prediction model also achieved an accuracy of 92.5% in identifying the appropriate switching configurations, with an inference time of less than 0.1 seconds.
The findings indicate that the integrated framework can effectively support loss reduction, voltage improvement, and faster operational decision-making in weak distribution networks. The approach could be particularly useful for rural and underserved distribution systems in developing countries, where improving network reliability and accommodating increasing renewable energy penetration remain important challenges.
Keywords
Distribution Networks; NSGA-II; Energy Storage Systems; PSO-GA; Artificial Intelligence; Voltage Stability; Loss Minimization; Nigeria
This article is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
Downloads
References
1. Abubakar, M., Aliyu, U.O., and Sambo, A.S. (2020). Assessment of voltage instability in weak distribution networks in Nigeria. Nigerian Journal of Power Systems Engineering, 12(2), 44–53. [Google Scholar] [Crossref]
2. Akinyele, D.O., and Rayudu, R. (2019). Techno-economic analysis of energy storage systems in microgrids. Renewable and Sustainable Energy Reviews, 101, 353–370. [Google Scholar] [Crossref]
3. Akpan, E.O., and Okoro, O.I. (2021). Renewable energy integration challenges in developing countries: A case study of Nigeria. Journal of Electrical Engineering Research, 18(3), 112–125. [Google Scholar] [Crossref]
4. Bala, E.J., and Ibeh, F.C. (2019). Impact of distributed renewable energy on voltage stability of distribution feeders. International Journal of Power and Energy Studies, 7(4), 88–96. [Google Scholar] [Crossref]
5. Electric Power Sector Reform Act (EPSRA). (2005). Federal Republic of Nigeria. [Google Scholar] [Crossref]
6. Fadare, D.A. (2018). Energy policies in Nigeria: Evolution, challenges, and future directions. Energy Policy Review, 6(1), 21–36. [Google Scholar] [Crossref]
7. Ibrahim, Y.Y., Mohammed, A.A., and Yau, M. (2022). Multi-objective particle swarm optimization for optimal ESS placement in weak grids. Energy Engineering Research, 15(1), 50–66. [Google Scholar] [Crossref]
8. Li, X., Zhang, Y., and Chen, H. (2017). Reinforcement learning-based voltage regulation in distribution networks. IEEE Transactions on Smart Grid, 8(4), 1880–1891. [Google Scholar] [Crossref]
9. Liu, S., Luo, M., and Wang, J. (2020). Storage allocation for loss minimization in distribution feeders. Journal of Energy Storage, 28, 101–123. [Google Scholar] [Crossref]
10. National Renewable Energy and Energy Efficiency Policy (NREEEP). (2015). Federal Ministry of Power, Nigeria. [Google Scholar] [Crossref]
11. Nwaigwe, K.N., and Dike, H.O. (2020). Optimal planning of hybrid renewable systems for rural electrification in Sub-Saharan Africa. African Energy Journal, 9(2), 99–118. [Google Scholar] [Crossref]
12. Ogunjuyigbe, A.S.O., Ayodele, T.R., and Alao, M.A. (2016). Evaluation of energy storage systems for peak shaving in weak African grids. Energy Conversion and Management, 111, 44–53. [Google Scholar] [Crossref]
13. Ojo, J.S., and Olarinoye, O.M. (2021). Voltage stability improvement in Nigerian feeders using battery storage. Journal of Electrical Power Systems and Automation, 14(1), 65–77. [Google Scholar] [Crossref]
14. Olaniyi, A., and Popoola, A. (2023). Machine learning for distribution network optimization in Africa: A review. Journal of Energy Informatics, 6(1), 55–72. [Google Scholar] [Crossref]
15. Rahman, M.M., and Khan, M.A. (2019). Comparative analysis of multi-objective optimization methods for ESS sizing. Applied Energy Systems Journal, 5(3), 87–102. [Google Scholar] [Crossref]
16. Yunusa, M.B., and Ahmed, T. (2022). Load flow and voltage performance analysis of rural Nigerian feeders. Journal of Power Distribution Studies, 11(2), 76–89. [Google Scholar] [Crossref]
This work is licensed under CC BY 4.0