Mitigating the Effects of Faulty Squirrel-cage-rotor Bars of Three-phase Induction Motor through Convergence Analysis of a Hybrid ANN-PSO Control System

References

[1] D. Kim, H. Park, and J. Lee, “Influence of load conditions on motor current signature analysis-based fault detection,” IET Electric Power Applications, vol. 14, no. 8, pp. 1105–1112, 2020.

[2] R. Kumar, A. Gupta, and P. Sharma, “Optimization of induction motor fault diagnosis using particle swarm optimization,” Eng. Appl. Artif. Intell., vol. 115, p. 105159, 2022.

[3] R. Kumar, A. Patel, and S. Verma, “Identifying broken rotor bars in three-phase induction motors using current signature analysis,” Electr. Power Compon. Syst., vol. 46, no. 12, pp. 1240–1255, 2018.

[4] R. Kumar, K. Sharma, and M. Gupta, “Evaluating rotor bar defects using FFT and wavelet transforms,” J. Vib. Acoust., vol. 143, no. 2, p. 021003, 2021.

[5] R. Kumar, S. Verma, and P. Singh, “Advanced spectral analysis for diagnosing rotor faults in induction motors,” IEEE Trans. Power Electron., vol. 35, no. 5, pp. 4489–4502, 2020.

[6] S. Kumar, R. Patel, and A. Sharma, “Machine learning-based approaches for rotor bar fault detection in induction motors,” IEEE Access, vol. 11, pp. 10432–10450, 2023.

[7] J. Li, W. Zhang, and X. Chen, “Predictive maintenance strategies for induction motors using real-time monitoring,” Reliab. Eng. Syst. Saf., vol. 195, p. 106719, 2020.

[8] X. Li, J. Wang, and R. Patel, “Predictive maintenance methodologies for induction motors,” Mech. Syst. Signal Process., vol. 62, no. 4, pp. 321–337, 2017.

[9] X. Li, J. Wang, and Y. Zhang, “Particle swarm optimization for PID controller tuning in industrial automation,” J. Control Eng., vol. 32, no. 1, pp. 55–67, 2024.

[10] Y. Li, N. Gupta, and P. Singh, “Condition monitoring techniques for induction motors: A review,” IEEE Trans. Ind. Appl., vol. 57, no. 8, pp. 2104–2118, 2021.

[11] C. H. Lin, “Altered grey wolf optimization and Taguchi method with FEA for six-phase copper squirrel cage rotor induction motor design,” Energies, vol. 13, no. 9, p. 2282, 2020.

[12] Y. Liu, H. Chen, and L. Wang, “Enhancing electromagnetic field optimization using hybrid particle swarm and deep learning methods,” IEEE Trans. Magn., vol. 58, no. 6, pp. 1–8, 2022.

[13] G. Lodewijks, Y. Cao, N. Zhao, and H. Zhang, “Reducing CO₂ emissions of an airport baggage handling transport system using a particle swarm optimization algorithm,” IEEE Access, vol. 9, pp. 121894–121905, 2021.

[14] M. Y. Mohamed, M. Fawzi, S. A. A. Maksoud, and A. E. Kalas, “Finite element analysis of multi-phase squirrel cage induction motor to develop the optimum torque,” in Proc. IEEE Conf. Power Electron. Renew. Energy (CPERE), Aswan, Egypt, Oct. 2019, pp. 504–510.

[15] X. Ning, T. Zhang, and Z. Wang, “Advanced optimization techniques in induction motor design: A review,” IEEE Trans. Ind. Electron., vol. 71, no. 3, pp. 2251–2263, 2024.

@article{1716211,
      author = {Ogbu Ifeanyi Nnanwezi, E. C. Obuah, C. O. Ahiakwo, H. N. Amadi},
      title = {Mitigating the Effects of Faulty Squirrel-cage-rotor Bars of Three-phase Induction Motor through Convergence Analysis of a Hybrid ANN-PSO Control System},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {10},
      pages = {1692-1701},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1716211.pdf},
      abstract = {Induction motors play a vital role in industrial systems, yet their reliability is threatened by rotor bar defects that cause current imbalance, torque pulsations, and efficiency losses. Traditional diagnostic approaches such as vibration monitoring and thermal imaging often fail to detect these faults at early stages, resulting to unplanned downtime and economic losses. To address these challenges, this study established a hybrid Artificial Neural Network–Particle Swarm Optimization (ANN-PSO) framework for fault detection and performance optimization in a 30 kW squirrel-cage induction motor with broken rotor bars. The ANN is trained to identify faults and their severity from current and vibration signals, while PSO optimizes motor control parameters by minimizing fitness functions incapacitations that integrate stator current imbalance, torque ripple, and power loss. Simulations with 20 swarm particles over 50 iterations demonstrated convergence to optimal solutions. Quantitatively, torque ripple was reduced from 0.15pu to 0.08pu, representing a 46.7% improvement, while stator current imbalance dropped from 0.12pu to 0.05pu, a 58.3% reduction. Furthermore, Total Harmonic Distortion (THD) was decreased by approximately 40% across harmonics, with the 7th harmonic reduced from 49% to 29.4%. The ANN achieved a mean squared error of less than 1×10⁻⁴ after 200 training epochs, confirming its predictive accuracy. These results demonstrated that the ANN-PSO model enhances efficiency and extends operational life while maintaining rated power output. Policy implications highlighted the importance of adopting AI-driven predictive maintenance strategies in Industry 4.0 to reduce downtime, save energy, and promote sustainable motor operations. The study therefore contributes to industrial reliability and cost-effectiveness through intelligent, adaptive rotor bars’ fault management.},
      keywords = {ANN, PSO, THD, Induction Motor, Vibration},
      month = {April},
      doi = {https://doi.org/10.64388/IREV9I10-1716211}
  }