Citation :

Amruta Vaibhav Adwant, Manpreet Singh, Suhas Deshmukh, "AI-Based Correlation Analysis of Torque and Vibration for Misalignment Fault Detection in Rotating Machinery," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 5, pp. 577-597, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I5P136

Abstract

Shaft misalignment is a prevalent issue in rotating machinery. The majority of individuals initially overlook this issue due to the relatively low vibration amplitudes, and conventional vibration-based diagnostics are not very adept at detecting them. This study investigates the relationship between torque and vibration data to enable precise diagnosis of misalignment issues in real operational conditions. A test rig was built to gather real-time data on torque and vibration in both aligned (0°) and misaligned (0.5°–2°) states. It had a shaft made of A36 steel, torque estimates based on encoders, and piezoelectric accelerometers. Principal Component Analysis was performed to get rid of and reduce features in the time, frequency, and cross-domains. After that, Support Vector Machine, Random Forest, and Convolutional Neural Network models were used to group the errors. The results show that misalignment causes a lot of harmonic amplification and less torque-vibration coherence. This is shown by the fact that correlation values drop from about 0.95 in aligned conditions to 0.65 in misaligned conditions. The combination of deep learning and multisensory fusion was proven to be better by the classification accuracies of 91.3%, 93.2%, and 95.7% reached using SVM, RF, and CNN models, respectively. The primary innovative feature of this study is how it combines torque and vibration characteristics from an encoder with knowledge of physics and geometry. Because of its integration, low-angle misalignment (0.5°) can be precisely identified, even though vibration-only approaches usually miss it. A scalable and easy way to forecast when rotating gear needs maintenance is provided by this approach to incorporating torque and vibration. It also improves the longevity of models across different shaft shapes and facilitates early issue identification.

Keywords

Torque-Vibration Correlation, Rotating Machinery, Support Vector Machines, Random Forest, Convolutional Neural Networks, Vibration Signatures, Time-Frequency Spectrogram, Fast Fourier Transform.

References

[1] Robert Bond Randall, Vibration-based Condition Monitoring: Industrial, Automotive and Aerospace Applications, 2^nd ed., John Wiley and Sons, 2021.
[Google Scholar] [Publisher Link]

[2] Jérôme Antoni, “The Spectral Kurtosis: A Useful tool for Characterizing Non-Stationary Signals,” Mechanical Systems and Signal Processing, vol. 20, no. 2, pp. 282-307, 2006.
[CrossRef] [Google Scholar] [Publisher Link]

[3] Yaguo Lei et al., “Machinery Health Prognostics: A Systematic Review from Data Acquisition to RUL Prediction,” Mechanical Systems and Signal Processing, vol. 104, pp. 799-834, 2018.
[CrossRef] [Google Scholar] [Publisher Link]

[4] Linfeng Deng, Aihua Zhang, and Rongzhen Zhao, “Intelligent Identification of Incipient Rolling Bearing Faults based on VMD and PCA-SVM,” Advances in Mechanical Engineering, vol. 14, no. 1, pp. 1-18, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[5] Lijun Zhang, Yuejian Zhang, and Guangfeng Li, “Fault-Diagnosis Method for Rotating Machinery based on SVMD Entropy and Machine Learning,” Algorithms, vol. 16, no. 6, pp. 1-18, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[6] Ziyuan Jiang, Qinkai Han, and Xueping Xu, “Fault Diagnosis of Planetary Gearbox based on Motor Current Signal Analysis,” Shock and Vibration, vol. 2020, no. 1, pp. 1-13, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[7] Pauline Ong et al., “A Deep Convolutional Neural Network for Vibration-Based Health-Monitoring of Rotating Machinery,” Decision Analytics Journal, vol. 7, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[8] Chao Fu, Qing Lv, and Hsiung-Cheng Lin, “Development of Deep Convolutional Neural Network with Adaptive Batch Normalization Algorithm for Bearing Fault Diagnosis,” Shock and Vibration, vol. 2020, no. 1, pp. 1-10, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[9] Siyu Zhang et al., “Rotating Machinery Fault Detection and Diagnosis based on Deep Learning: A Survey,” Chinese Journal of Aeronautics, vol. 36, no. 1, pp. 45-74, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[10] Mahmoud Elsamanty, Abdelkader Ibrahim, and Wael Saady Salman, “Principal Component Analysis Approach for Detecting Faults in Rotary Machines based on Vibrational and Electrical Fused Data,” Mechanical Systems and Signal Processing, vol. 200, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[11] Clayton Eduardo Rodrigues, Cairo Lúcio Nascimento Júnior, and Domingos Alves Rade, “Application of Machine Learning Techniques and Spectrum Images of Vibration Orbits for Fault Classification of Rotating Machines,” Journal of Control, Automation and Electrical Systems, vol. 33, no. 1, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[12] P.A. Pérez, F.C. Gómez, and F. Marín, “Measurement Techniques of Torsional Vibration in Rotating Shafts,” CMC-Tech Science Press, vol. 44, no. 2, pp. 85-104, 2014.
[CrossRef] [Google Scholar] [Publisher Link]

[13] Amruta V. Adwant et al., “Development of Portable Dynamic Torque Power Measurement System,” International Journal of Advanced Mechatronic Systems, vol. 11, no. 2, pp. 95-101, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[14] 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]

[15] Leo Breiman, “Random Forests,” Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.
[CrossRef] [Google Scholar] [Publisher Link]

[16] Masoud Jalayer et al., “Fault Detection and Diagnosis with Imbalanced and Noisy Data: A Hybrid Framework for Rotating Machinery,” Machines, vol. 10, no. 4, pp. 1-22, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[17] Xiaotian Zhang et al., “Feature Engineering and Artificial Intelligence-Supported Approaches used for Electric Powertrain Fault Diagnosis: A Review,” IEEE Access, vol. 10, pp. 29069-29088, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[18] A. Glowacz, “Recognition of Acoustic Signals of Induction Motors with the Use of MSAF10 and Bayes Classifier,” Archives of Metallurgy and Materials, vol. 61, pp. 153-158, 2016.
[CrossRef] [Google Scholar] [Publisher Link]

[19] Fasikaw Kibrete, Dereje Engida Woldemichael, and Hailu Shimels Gebremedhen, “Multi-Sensor Data Fusion in Intelligent Fault Diagnosis of Rotating Machines: A Comprehensive Review,” Measurement, vol. 232, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[20] Hao Chen et al., “Review of Intelligent Fault Diagnosis for Rotating Machinery under Imperfect Data Conditions,” Expert Systems with Applications, vol. 285, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[21]  Xiao Yu et al., “Intelligent Fault Diagnosis of Rotating Machinery under Variable Working Conditions based on Deep Transfer Learning with Fusion of Local and Global Time-Frequency Features,” Structural Health Monitoring, vol. 23, pp. 2238-2254, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[22] Shengnan Tang et al., “Deep Transfer Learning Strategy in Intelligent Fault Diagnosis of Rotating Machinery,” Engineering Applications of Artificial Intelligence, vol. 134, 2024.
       [CrossRef] [Google Scholar] [Publisher Link]