Causal Neuromorphic Graph Learning with Physics-Informed Micro-Twins for Explainable Wind Turbine Fault Diagnosis

International Journal of Electrical and Electronics Engineering

© 2025 by SSRG - IJEEE Journal

Volume 12 Issue 12

Year of Publication : 2025

Authors : Surendar Aravindhan, M. Shyamalagowri, J.Karthika, Rajan.C

10.14445/23488379/IJEEE-V12I12P111

How to Cite?

Surendar Aravindhan, M. Shyamalagowri, J.Karthika, Rajan.C, "Causal Neuromorphic Graph Learning with Physics-Informed Micro-Twins for Explainable Wind Turbine Fault Diagnosis," SSRG International Journal of Electrical and Electronics Engineering, vol. 12,  no. 12, pp. 140-150, 2025. Crossref, https://doi.org/10.14445/23488379/IJEEE-V12I12P111

Abstract:

Wind Turbine Generators (WTGs) are considered one of the essential elements in the renewable energy system. Despite this, due to the complex, non-linear, coupled nature of fault dynamics, fault detection of WTG poses enough challenge. Conventional signal processing, model-based estimation, and machine learning fault diagnosis methods provide valuable insights, but have limitations in prediction, scalability, comprehensibility, and uncertainty. New hybrid frameworks relying on techniques like deep learning fused with fuzzy decision support help enhance the detection of harmonics and condition-based optimization of control devices. Unfortunately, these various advanced methods are currently missing under-utilized capabilities that focus on explainable, reasonable, and proactive prediction of the fault’s progressive harm—a paper framework for advanced fault diagnosis and predictive maintenance of Wind Turbine Generators (WTGs). To begin, the turbine’s dynamics are embedded into adaptive digital twins using physics-informed neural networks that develop self-evolving micro-digital twins at the component level (rotor, gearbox, bearings, stator). Secondly, the neuromorphic spiking Graph Neural Networks (S-GNNs) will extract temporal transients from multiple vibration, current, and rotor speed signals while integrating causal inference for root-cause analysis and counterfactual reasoning. The system combines micro-twin predictions with causal learning of neuromorphic to provide estimates of fault intensity, RUL, and interpretable decision support. This technique will provide improved accuracy of fault diagnosis, real-time deployment on edge devices using voltage-efficient neuromorphic computing, uncertainty calibration of RUL prediction, and operator trust using transparent causal explanations. Through the unification of physics-informed modeling, neuromorphic efficiency, and causal AI, this work sets the groundwork for the next generation of explainable and sustainable fault diagnosis in wind turbine generators.

Keywords:

Wind Turbine Fault Diagnosis, Neuromorphic Computing, Causal AI, Physics-Informed Micro-Twins, Predictive Maintenance.

References:

[1] Bo Li et al., “Research, Application, and Challenges of Causal Inference in Industrial Fault Diagnosis: A Survey,” Engineering Applications of Artificial Intelligence, vol. 158, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Ruonan Liu et al., “Causal Intervention Graph Neural Network for Fault Diagnosis of Complex Industrial Processes,” Reliability Engineering & System Safety, vol. 251, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Agus Ristono, and Pratikto Budi, “Next-Gen Power Systems in Electrical Engineering,” Innovative Reviews in Engineering and Science, vol. 2, no. 1, pp. 34-44, 2025.
[Google Scholar] [Publisher Link]
[4] Zehui Mao et al., “Graph Convolutional Neural Network for Intelligent Fault Diagnosis of Machines via Knowledge Graph,” IEEE Transactions on Industrial Informatics, vol. 20, no. 5, pp. 7862-7870, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Madhanraj, “AI-Powered Energy Forecasting Models for Smart Grid-Integrated Solar and Wind Systems,” National Journal of Renewable Energy Systems and Innovation, vol. 1, no. 1, pp. 1-7, 2025.
[Publisher Link]
[6] Florian Stadtmann, and Adil Rasheed, “Diagnostic Digital Twin for Anomaly Detection in Floating Offshore Wind Energy,” ASME 2024 43rd International Conference on Ocean, Offshore and Arctic Engineering, Singapore, vol. 7, pp. 1-10, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Jincheng Zhang, and Xiaowei Zhao, “Digital Twin of Wind Farms Via Physics-Informed Deep Learning,” Energy Conversion and Management, vol. 293, pp. 1-12, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[8] P. Joshua Reginald, “Hybrid AC/DC Microgrid Power Management using Intelligent Power Electronics Interfaces,” Transactions on Power Electronics and Renewable Energy Systems, vol. 1, no. 1, pp. 21-29, 2025.
[Google Scholar] [Publisher Link]
[9] Davide Astolfi, Fabrizio De Caro, and Alfredo Vaccaro, “Recent Advances in the use of Explainable Artificial Intelligence Techniques for Wind Turbine Systems Condition Monitoring,” Electronics, vol. 12, no. 16, pp. 1-4, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Kanru Cheng et al., “Research on Gas Turbine Health Assessment method based on Physical Prior Knowledge and Spatial-Temporal Graph Neural Network,” Applied Energy, vol. 367, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Simi Job et al., “Exploring Causal Learning through Graph Neural Networks: An In-Depth Review,” Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, vol. 15, no. 2, pp. 1-34, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Simon Letzgus, and Klaus-Robert Müller, “An Explainable AI Framework for Robust and Transparent Data-Driven Wind Turbine Power Curve Models,” Energy and AI, vol. 15, pp. 1-16, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Robbi Rahim, “Spectrum-Adaptive Neuromorphic Learning for Wireless Wind Turbine Condition Monitoring,” Journal of Wireless Intelligence and Spectrum Engineering, vol. 1, no. 1, pp. 10-18, 2024.
[Google Scholar] [Publisher Link]
[14] Amirashkan Haghshenas et al., “Predictive Digital Twin for Offshore Wind Farms,” Energy Informatics, vol. 6, no. 1, pp. 1-26, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[15] C. Arun Prasath, “Digital Twin-Driven Predictive Maintenance in Intelligent Power Systems,” National Journal of Intelligent Power Systems and Technology, vol. 1, no. 1, pp. 29-37, 2025.
[Google Scholar] [Publisher Link]
[16] Muhammad Irfan et al., “Revolutionizing Wind Turbine Fault Diagnosis on Supervisory Control and Data Acquisition Systems with Transparent Artificial Intelligence,” International Journal of Green Energy, vol. 22, no. 10, pp. 2029-2045, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Saravanakumar Veerappan, “Digital Twin Modeling for Predictive Maintenance in Large-Scale Power Transformers,” National Journal of Electrical Machines and Power Conversion, vol. 1, no. 1, pp. 39-44, 2025.
[Google Scholar] [Publisher Link]
[18] S. Sindhu, “Model Predictive Control-based Bidirectional EV Charging Strategy for Grid Load Shaping and Cost Optimization in Smart Grids,” Journal of Reconfigurable Hardware Architectures and Embedded Systems, vol. 2, no. 1, pp. 1-6, 2025.
[Google Scholar] [Publisher Link]
[19] Jiaying Chen et al., “Bearing Fault Diagnosis using Graph Sampling and Aggregation Networks,” arXiv Preprint, pp. 1-11, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Dahlan Abdullah, “Redox Flow Batteries for Long-Duration Energy Storage: Challenges and Emerging Solutions,” Transactions on Energy Storage Systems and Innovation, vol. 1, no. 1, pp. 9-16, 2025.
[Google Scholar] [Publisher Link]
[21] Emmanuel Branlard et al., “A Digital Twin Solution for Floating Offshore Wind Turbines Validated using A Full-Scale Prototype,” Wind Energy Science, vol. 9, no. 1, pp. 1-24, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Davide Astolfi, Fabrizio De Caro, and Alfredo Vaccaro, “Condition Monitoring of Wind Turbine Systems by Explainable Artificial Intelligence Techniques,” Sensors, vol. 23, no. 12, pp. 1-24, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[23] K.P. Uvarajan, “Vibration Analysis of Smart Structures Integrated with Embedded Piezoelectric Sensor Networks: A Comprehensive Review,” Journal of Reconfigurable Hardware Architectures and Embedded Systems, vol. 1, no. 1, pp. 18-29, 2024.
[Google Scholar] [Publisher Link]
[24] Achmad Widodo, and Bo-Suk Yang, “Support Vector Machine in Machine Condition Monitoring and Fault Diagnosis,” Mechanical Systems and Signal Processing, vol. 21, no. 6, pp. 2560-2574, 2007.
[CrossRef] [Google Scholar] [Publisher Link]
[25] K.P. Uvarajan, “Automated Distributed Neuromorphic Pipelines for Wind Turbine Fault Analytics at Scale,” SECITS Journal of Scalable Distributed Computing and Pipeline Automation, vol. 1, no. 1, pp. 1-6, 2024.
[Google Scholar] [Publisher Link]
[26] Saravanakumar Veerappan, “Scalable Graph-Driven Neuromorphic Analytics for Large-Scale Wind Farm Fault Detection,” Journal of Scalable Data Engineering and Intelligent Computing, vol. 1, no. 1, pp. 1-7, 2024.
[Google Scholar] [Publisher Link]
[27] S. Sindhu, “Secure Neuromorphic Graph Intelligence for Fault-Resilient Wind Turbine Sensor Networks,” Transactions on Secure Communication Networks and Protocol Engineering, vol. 1, no. 1, pp. 1-10, 2024.
[Google Scholar] [Publisher Link]
[28] S. Poornimadarshini, “Runtime-Reconfigurable Neuromorphic Graph Accelerators for Energy-Efficient Real-Time Wind Turbine Fault Inference,” Journal of Reconfigurable Hardware Architectures and Embedded Systems, vol. 1, no. 1, pp. 52-58, 2024.
[Google Scholar] [Publisher Link]