Improving Operational Resilience and Peak-Load Management in the Al-Ruwais Network Through Coordinated PV, Battery Storage, and Demand Response
DOI:
https://doi.org/10.65405/hcna7g39Keywords:
Hybrid Neural Forecasting-Resilience-Oriented Load Management (HNF-ROLM), General Electricity Company of Libya (GECOL), forward Neural Network (FNN), photovoltaic (PV), Battery Energy Storage System (BESS), Demand Response (DR), Energy Not Supplied (ENS), System Average Interruption Duration Index (SAIDI), System Average Interruption Frequency Index (SAIFI), Resilience Index (RI),Abstract
High-voltage networks in western Libya operate under a demanding combination of pronounced daily and seasonal load variation, incomplete station-level visibility, and narrow security margins whenever a transformer or feeder is unavailable. The engineering challenge is therefore broader than forecasting accuracy. A statistically acceptable forecast can still lead to overload, voltage stress, uneven feeder loading, or emergency shedding when it is not translated into a feasible operating action. The same principle applies to photovoltaic generation (PV) can reduce daytime imports, yet an uncoordinated plant may leave the evening peak almost unchanged and can simply relocate stress within the network for these reasons, we conducted a study of a mini-loop represented by the Al-Ruwais loop in the western Libya network.
This research is considered an extension of the doctoral dissertation. introduces the Hybrid Neural Forecasting Resilience Oriented Load Management (HNF-ROLM) framework for Libya's Al-Ruwais network, using (GECOL) demand data. It integrates an (FNN) forecast with a decision algorithm coordinating (PV, BESS, DR), and reinforcement under normal and (N-1) constraints. Performance is evaluated via (ENS, SAIDI, SAIFI), and a Resilience Index (RI), moving beyond mere forecasting accuracy MATLAB software was used. Results show (PV) cuts daytime imports, while (BESS) and (DR) provide critical evening-peak and contingency support, their coordinated use significantly reduces unserved energy, outage duration/frequency, and network violations while improving (RI). The planning layer identifies when demand growth requires structural reinforcement confirming that operational flexibility and firm upgrades are complementary. HNF-ROLM strengthens resilience via PV, BESS, and DR, delivering peak relief while guiding staged reinforcement planning for future security.
Downloads
References
[1] M. K. A. Al-Mashakwi, “Enhancing Electrical Network Resilience through Advanced Load Demand Management, Including Load Forecasting with Renewable Energy Integration in West Libya,” Ph.D. research proposal, School of Applied Sciences and Engineering, Libyan Academy for Postgraduate Studies, Libya, 2025, unpublished.
[2] Y. Zhou, D. Qin, and Y. Wang, Shall We Evaluate Load Forecasts,” Power and Energy Future, vol. 1, no. 1, Art. no. 9650004, 2026, doi: 10.26599/PEF.2026.9650004.
[3] S. Haykin, Neural Networks and Learning Machines, 3rd ed. Upper Saddle River, NJ, USA: Pearson, 2009.
[4] F. Alhaj Omar and N. Pamuk, “Comparative Forecasting of Electricity Load and Generation in Türkiye Using Prophet, XGBoost, and Deep Neural Networks,” Sustainability, vol. 18, no. 6, Art. no. 2838, 2026, doi: 10.3390/su18062838.
[5] J. Eckhoff, S. Wadhwa, M. Fette, J. P. Wulfsberg, and C. Wanigasekara, “Electrical Load Forecasting in the Industrial Sector: A Literature Review of Machine Learning Models and Architectures for Grid Planning,” Energies, vol. 19, no. 2, Art. no. 538, 2026, doi: 10.3390/en19020538.
[6] D. Rabie, M. Moradi, W. Xuan, and H. Farzaneh, “Impact of Accurate Load Forecasting on Electricity Market Stability in Japan Using Classical Time-Series and Deep-Learning Methods,” Scientific Reports, vol. 16, Art. no. 11781, 2026, doi: 10.1038/s41598-026-46859-2.
[7] H. Liu, Y. Wei, H. Zhang, Y. Wang, J. Zhang, and Z. Luo, “A Hybrid LSTM-Transformer Model for Short-Term Urban Electricity Load Forecasting: A Two-City Case Study,” Scientific Reports, vol. 16, Art. no. 16964, 2026, doi: 10.1038/s41598-026-47905-9.
[8] M. Haggag, K. Abdelhady, M. Guirguis, M. Saudy, and W. El Dakhakhni, “Machine Learning Long-Term Electricity Demand Forecasting System for Strategic Energy Investments,” Scientific Reports, vol. 16, Art. no. 12471, 2026, doi: 10.1038/s41598-026-45123-x.
[9] S. Luo, X. Chen, X. Pang, B. Wang, and Z. Zheng, “Mid-Term Power Load Forecasting Using an Ensemble Deep Learning Model with BKA and CWGAN-GP Enhancements,” Scientific Reports, vol. 16, Art. no. 18612, 2026, doi: 10.1038/s41598-026-49674-x.
[10] K. L. Yandar, O. Revelo Sánchez, and M. Bolaños Gonzales, “Joint Forecasting of Energy Consumption and Generation in P2P Networks Using LSTM-CNN and Transformers,” Energies, vol. 19, no. 3, Art. no. 760, 2026, doi: 10.3390/en19030760.
[11] General Electricity Company of Libya (GECOL), General Control Center/Western Region, “Daily Generation, Demand, Frequency, and Load-Shedding Operational Reports,” 19 macro-enabled workbooks, unpublished utility operational data, 2023.
[12] M. K. Al-Mashakwi, M. A. Nasrst, and O. M. Abdulrahman, “Analysis of Load Growth and Renewable Energy-Based Optimization of a Libyan Power Network Segment Using ETAP,” research manuscript, Az-Zawiya University, Libya, 2026, unpublished.
[13] F. E. Abrahamsen, I. Norheim, and K. O. Uhlen, “Flexibility and Controllability in Low-Voltage Distribution Grids Under High PV Penetration,” Energies, vol. 19, no. 9, Art. no. 2072, 2026, doi: 10.3390/en19092072.
[14] R. Mu, Y. Zhang, Y. Wu, X. Wan, X. Song, D. Wang, L. Sun, and B. Yang, “A Critical Review of State-of-the-Art Stability Control of PV Systems: Methodologies, Challenges, and Perspectives,” Energies, vol. 19, no. 2, Art. no. 507, 2026, doi: 10.3390/en19020507.
[15] Y. Xu, Z. You, Y. Shi, G. Wang, Y. Wang, and B. Yang, “Optimal Dispatch of Energy Storage Systems in Flexible Distribution Networks Considering Demand Response,” Energies, vol. 19, no. 2, Art. no. 407, 2026, doi: 10.3390/en19020407.
[16] H. Saadat, Power System Analysis. Boston, MA, USA: WCB/McGraw-Hill, 1999.
[17] Y. Yang, R. Ou, W. Liu, D. Zheng, K. Hou, C. Wu, Y. Tuo, J. Wang, and J. Tang, “Optimal Operation of Coordinated Transmission and Distribution Networks Considering Multi-Type Flexible Resource Aggregation,” Frontiers in Energy Research, vol. 14, Art. no. 1656716, 2026, doi: 10.3389/fenrg.2026.1656716.
[18] MathWorks, “MATLAB Documentation: Language Fundamentals, Optimization, and Neural-Network Implementation Resources,” Natick, MA, USA, accessed Jul. 19, 2026.
[19] Ramadan, S. I. (2026). gRPC (HTTP/2) vs REST (HTTP/1.1) in Distributed Systems: Performance, Scalability, and the Cryptographic Cost of TLS 1.3. Al-Farooq Journal of Sciences, 2(3), 1031-1039.
[20] Sowan, E., Elbuaishi, A., & Areir, M. (2026). Enhancing Face Recognition Performance: A Comparative Evaluation of ArcFace and FaceNet using Image Preprocessing and ROC–EER Metrics. Al-Farooq Journal of Sciences, 2(3), 973-990.
[21] Alnnale, T. (2026). Predictive Governance in Digital Enterprises: An LSTM-Enhanced Deep Learning Framework for Economic Optimization of IT Incident Management Using Enriched Process Logs. Al-Farooq Journal of Sciences, 2(3), 86-113.
