Current-Based Fault Diagnosis in Electromechanical Systems

Current-based fault diagnosis exploits measurements of electrical currents in motors and drivetrains to detect incipient mechanical and electrical faults. By analysing variations, modulations and spectral content of stator currents or residual currents, this approach bypasses the need for direct vibration or acoustic sensors and is particularly suited to harsh or inaccessible environments. Common implementations include motor current signature analysis (MCSA), where characteristic frequencies linked to gear meshing, bearing defects or rotor asymmetries produce sidebands around the supply frequency. Signal-processing techniques such as demodulation, envelope analysis, advanced time–frequency transformations and statistical residual evaluation enhance sensitivity to weak fault signatures under variable load and speed conditions. Model-based schemes use state-space representations and parameter estimation algorithms to generate residual currents whose variance or clustering behaviour indicates emerging faults. Machine learning and deep-learning frameworks increasingly augment feature extraction, enabling automated classification of multi-fault conditions in complex electromechanical assemblies. The global significance of current-based diagnosis spans industrial manufacturing, renewable energy systems and transportation, offering cost-effective, non-invasive monitoring and early warning of gearboxes, belt drives, pumps, conveyors and wind-turbine generators.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Current-Based Fault Diagnosis in Electromechanical Systems publication trend

The graph below shows the total number of articles in current-based fault diagnosis in electromechanical systems across all publications each year (not limited to Nature Index journals).

Technical terms

Motor Current Signature Analysis (MCSA): A diagnostic technique analysing motor current waveforms to identify mechanical and electrical faults via characteristic spectral components.

Variational Mode Decomposition (VMD): A signal-processing method that adaptively decomposes a signal into band-limited intrinsic mode functions.

Hilbert Spectrum: A time–frequency representation obtained by applying the Hilbert transform to intrinsic mode functions, revealing instantaneous energy distributions.

Permanent Magnet Synchronous Motor (PMSM): An electric machine featuring permanent magnets on the rotor, commonly used in high-performance drives.

Drive-Tolerant Current Residual Variance (DTCRV): A fault indicator computed as the variance of the residual current after removal of drive-related components, invariant to speed and load changes.

References

1. Intelligent Fault Diagnosis Method Based on VMD-Hilbert Spectrum and ShuffleNet-V2: Application to the Gears in a Mine Scraper Conveyor Gearbox. Sensors (2023).
2. Permanent Magnet Synchronous Motor Driving Mechanical Transmission Fault Detection and Identification: A Model-Based Diagnosis Approach. Electronics (2022).
3. Drive-Tolerant Current Residual Variance (DTCRV) for Fault Detection of a Permanent Magnet Synchronous Motor Under Operational Speed and Load Torque Conditions. IEEE Access (2021).