Research Work

1. Mati Ullah Ahsan, Bahrom, Ihsan Khalil, Zainab “Comparative analysis of mathematical and simulation models for electric field generated beneath 400 kV overhead transmission lines.” Measurement, Elsevier, 2025. (IF 5.6)

2. Umais Khan, Sarim Farqaleet, Ihsan Ullah Khalil, Matiullah Ahsan, “Deep Learning-Based Vision Transformer Approach for Detecting Overlapping PV Faults Using Multi-Labeling”. Measurement, Elsevier, 2025. (IF 5.6)

3. Mati Ullah Ahsan, Bahrom, Ihsan Khalil, Zainab “ Analysis of high-ampacity and low-sag conductors of 27 5 kV overhead transmission lines using reconductoring techniques”. Electrical Power Systems Research, Elsevier, 2025. (IF 4.2)

4. Mati Ullah Ahsan, Bahrom, Ihsan Khalil, Zainab “ Simulation Based Analysis of Electric Field Characterisatics under High-Voltage Double Circuit and Quadrapole Overhead Transmisison lines”. Electrical Power Systems Research, Elsevier, 2025. (IF 4.2)

5. Ihsan Khalil, Marium Jalal, Azhar ul Haq, et.al, “A Fuzzy Reconfiguration Approach for PV losses mitigation” Result in Engineering, “Elsevier, 2025. (IF 6.9)

6. Marium Jalal, Ihsan Khalil, et.al. "Advancements in PV Array Reconfiguration Techniques: Review Article”, IEEE ACCESS 2024. (IF 3.4)

7. Mati Ullah Ahsan, Bahrom, Ihsan Khalil, “Measuring and Simulation of Magnetic Field Generated by High Voltage Overhead Transmission Lines”. Results in Engineering, Elsevier, 2024. (IF 6.0)

8. Marium Jalal, Ihsan Ullah Khalil, Azhar ul Haq, “Deep Learning Approaches for Visual Faults Diagnosis of Photovoltaic Systems: State-of-the-Art Review”. Results in Engineering, Elsevier, 2024. (IF 6.0)

9. Mati Ullah Ahsan, Bahrom, Ihsan Khalil, “Improving Electric Field Stress using Grading Ring Devices for 132kV Insulated Cross-arm”. Results in Engineering, Elsevier, 2024. (IF 6.0)

10. Mati Ullah Ahsan, Bahrom, Ihsan Khalil, “simulation based comparative analysis of electric field stress on insulated cross-arm”. Results in Engineering, Elsevier, 2024. (IF 6.0)

11. Aqdas, Imran Shafi, Ihsan Khalil, Shahzor, et.al, “Enhancing Photovoltaic Systems using Gaussian Process Regression for Parameter Identification and Fault Detection,” Energy Reports, Elsevier, 2024. (IF 5.2)

12. Ihsan Khalil, Azhar ul Haq, “A modified Chess Knight PV array Reconfiguration,” Energy Reports, Elsevier, 2024. (IF 5.2)

13. Ihsan Khalil, Azhar ul Haq,” A deep learning-based transformer model for photovoltaic fault forecasting and classification”, Electrical Power Systems Research, Elsevier, 2024. (IF 3.9)

14. Haris, Ihsan Khalil, “High Gain Coupled Inductor SEPIC based Boost Inverter using Extended SPWM”, Energy Reports, Energy Reports, Elsevier, 2023. (IF 5.2)

15. Ihsan Khalil, Azhar ul Haq,” A Novel procedure for PV fault forecasting”, Electrical Power Systems Research, Elsevier, 2023. (IF 3.9)

16. Zeeshan, Ihsan Khalil, “A Novel Row Index Mathematical Procedure for Mitigation of PV Output Power Losses during partial Shading Conditions”, Symmetry 2023. (IF 2.7)

17. Haris, Ihsan Khalil, “Analysis of the Dual Active Bridge-Based DC-DC Converter Topologies, High-Frequency Transformer, and Control Techniques,” Energies, 2022. (IF 3.2)

18. Haris, Ihsan Khalil,” Analysis and Verification of Leakage Inductance Calculation in DAB Converters Based on High-Frequency Toroidal Transformers under Different Design Scenarios”, Energies, 2022. (IF 3.2)

19. Imran, Ihsan Khalil,” Unified Fuzzy Logic Based Approach for Detection and Classification of PV Faults Using I-V Trend Line”, Energies 2022. (IF 3.2)

20. Kaleem, A.; Khalil, I.U. “Feedback PID Controller-Based Closed-Loop Fast Charging of Lithium-Ion Batteries Using Constant-Temperature–Constant-Voltage Method”. Electronics 2021. (IF 2.9)

21. Usman, Ihsan Khalil; “Novel approach for overtaking maneuvering” PLoS ONE, 2021. (IF 3.7)

22. Jaswinder Kumar, I.U khalil et.al. “Solver-Based Mixed Integer Linear Programming (MILP) Based Novel Approach for Hydroelectric Power Generation Optimization” IEEE ACCESS Vol (8), 2020. (IF 3.9)

23. Saba gul, Azhar ul Haq, I.U Khalil. “A Unified Approach for Analysis of Faults in Different Configurations of PV Arrays and their Impact on Power Grid” Energies, 2020. (IF 3.2)

24. Shahmeer, Azhar ul Haq, I.U. Khalil,” Cross-Border Power Trade and Grid Interconnection in SAARC Region: Technical Standardization and Power Pool Model” IEEE ACCESS vol 7(1),2019. (IF 3.9)

25. I.U Khalil, Azhar ul Haq, Yousaf Mahmoud,” Comparative analysis of photovoltaic faults and performance evaluation of its detection technique”. IEEE ACCESS 2020. (IF 3.9)

For more information: https://orcid.org/0000-0001-5725-0420

Thesis Supervision (MS/PhD)

1) A Deep Learning based Framework for Infrared Small Object Segmentation in Naval Defense Systems 

Lt. Cdr. Syed Mubashir Batch 2024 (Master in AI) in progress

Description

In naval military operations, reliable detection of small infrared (IR) objects, such as UAVs and drones, is vital for early warning and defense, yet remains highly challenging due to their low contrast, minimal signatures, and the presence of complex maritime backgrounds. Traditional detection methods fail under these conditions, while traditional detection and segmentation approaches often struggle with information loss caused by the tiny target size. Naval applications demand real-time, computationally efficient solutions that can withstand varying sea states, weather, and noise. This creates a critical research gap that calls for the development of robust and efficient infrared small object detection algorithms using SIRST datasets as a benchmark for performance evaluation.

2) Benchmarking Vision Transformer Variants Against CNNs for Multi-Label Photovoltaic Fault Detection and Cross-Domain Adaptation

Ms Fiza Karim Palijo Batch 2024 (Master in AI) in progress

Description

As solar photovoltaic installations expand globally, efficient and accurate fault detection is critical to ensure reliable power generation and minimize operational losses. Vision Transformers (ViTs) have recently shown strong potential in detecting overlapping PV faults, but further exploration of newer variants such as Swin Transformer, DeiT, PVT, and BEiT is needed to assess their ability to balance accuracy with computational efficiency. While pretrained ViTs on the PVEL-AD dataset demonstrate promising results, their adaptability to broader PV fault scenarios has not been systematically investigated, particularly under cross-domain settings such as training on electroluminescence (EL) images and testing on infrared (IR) images without additional labeling. This gap highlights the need for research that benchmarks state-of-the-art ViT variants on PV datasets and explores transfer learning and domain adaptation strategies to improve robustness and generalization. The novelty of this work lies in combining these directions, evaluating diverse ViT architectures for multi-label PV fault detection and extending their applicability through transfer learning and cross-domain adaptation, ultimately advancing toward models that are both accurate and deployment-ready for real-world PV monitoring.

3) Unified Approach for Enhancing Fidelity of Simulated Human Feedback in LLMs through Multi-Turn Conversations and Robustness Evaluation

Mr Shah Hussain Brohi Batch 2024 (Master in AI) in progress
Description

Large language models show considerable promise as simulators of human feedback, yet current approaches remain constrained in scope and fidelity. Most validation efforts have relied on limited annotator pools and focused primarily on single-turn instruction-following tasks, leaving open questions about their effectiveness in more complex interaction settings such as multi-turn dialogue or subjective and creative domains. Moreover, simulated feedback tends to exhibit less variability than real human feedback, which can distort training dynamics and reduce generalizability. Another overlooked dimension is robustness, as existing methods have not been systematically stress-tested under adversarial or noisy perturbations that frequently arise in real-world deployments. These limitations highlight the need for research that extends feedback simulation to multi-turn conversational contexts, rigorously evaluates robustness, and explores hybrid strategies that combine synthetic with small-scale real human feedback to improve both fidelity and reliability.

4) Physics-Informed Evaluation of Large Language Models for Autonomous Systems (Co-Supervision)

Lt. Cdr. Adeel Batch 2024 (Phd in AI) in progress

Thesis Supervision Completed

a) Overlapping Fault detecting using Visual Transformer (Completed- South Korea)

b) Hotspot and Shading Fault Mitigation Using Fuzzy Reconfiguration (completed) Abasyn University

c) Solar PV faults detection using Fuzzy logic (completed) Abasyn University

d) Impacts of different Solar PV faults on PV systems (Completed) Abasyn University