Verilog HDL-Based Simulation Framework for Multi-Parameter Industrial Motor Fault Classification

This work presented a Verilog HDL-based simulation framework for multi-parameter industrial motor fault classification. The framework models motor operating conditions using simulated motor current, vibration, temperature, rotor flux, stator flux, and supply voltage. These parameters are evaluated using deterministic decision rules to generate a compact 3-bit fault code representing predefined motor operating and fault conditions. The proposed design was organized as a modular HDL system comprising motor signal generation, feature processing, classification logic, and a simulation testbench. The design was compiled and simulated using Icarus Verilog, while Value Change Dump (VCD) generation and GTKWave were used for waveform-level observation. The documented simulation demonstrated the classification of healthy, rotor, stator, bearing, and voltage-fault conditions using the corresponding fault codes 000, 001, 010, 011, and 100. The overload condition was also defined in the classification architecture using code 101, although it was not included in the directly documented live simulation sequence. The simulation results provide a waveform-level view of the relationship between the predefined motor parameters and the classification output. These results demonstrate the functional behavior of the classification logic under the specified simulation conditions and provide a basis for further development toward physical implementation. The present work is limited to HDL simulation and does not include validation using a physical motor, sensor-acquisition system, or FPGA platform. Consequently, the results represent simulation-level functional verification rather than experimental motor-fault detection performance. Future development can incorporate measured sensor inputs, noise, varying load conditions, transient behavior, additional fault combinations, and FPGA implementation. More advanced feature-processing and machine-learning-based classification methods can also be investigated as extensions to the present deterministic approach.

[1] P. Gangsar and R. Tiwari, “Signal based condition monitoring techniques for fault detection and diagnosis of induction motors: A state-of-the-art review,” Mechanical Systems and Signal Processing, vol. 144, Art. no. 106908, 2020, doi: 10.1016/j.ymssp.2020.106908. [2] S. K. Gundewar and P. V. Kane, “Condition monitoring and fault diagnosis of induction motor,” Journal of Vibration Engineering & Technologies, vol. 9, pp. 643–674, 2021, doi: 10.1007/s42417-020-00253-y. [3] X. Yang, R. Yan, and R. X. Gao, “Induction motor fault diagnosis based on ensemble classifiers,” in Proc. IEEE Int. Instrum. Meas. Technol. Conf. (I2MTC), 2016, doi: 10.1109/I2MTC.2016.7520470. [4] I. A. Juarez-Trujillo, S. De León Aldaco, J. G. Velásquez-Aguilar, and J. Aguayo-Alquicira, “Fault analysis in induction motors through signal acquisition with Hilbert transform and FPGA,” IEEE Embedded Systems Letters, vol. 17, no. 6, pp. 390–393, 2025, doi: 10.1109/LES.2025.3546847. [5] C.-Y. Lee and W.-C. Lin, “Induction motor fault classification based on ROC curve and t-SNE,” IEEE Access, vol. 9, pp. 56330–56343, 2021, doi: 10.1109/ACCESS.2021.3072646. [6] J. R. Millan-Almaraz, R. D. J. Romero-Troncoso, L. M. Contreras-Medina, and A. Garcia-Perez, “Embedded FPGA based induction motor monitoring system with speed drive fed using multiple wavelet analysis,” in Proc. IEEE 3rd Int. Symp. Industrial Embedded Systems (SIES), 2008, doi: 10.1109/SIES.2008.4577702.