A fault diagnosis method for induction motor rotors combining correlation analysis with phase difference correction

Fault Diagnosis of Induction Machines With Rotor Asymmetries Using the Backward-Rotating Transient Current Component

Induction machine fault diagnosis using current spectral analysis is a well-established diagnostic technique based on the identification of the characteristic harmonic components generated in the machine current by each type of fault. However, one of the main problems with the application of this technique to the diagnosis of rotor asymmetry faults in induction machines is that the fault components have much lower amplitudes than the fundamental component and can be very close to it, making their detection difficult, especially in transient regimes. To improve the detection of fault harmonics, this work proposes a new diagnostic current signal, the backward-rotating transient current signal, which is generated in the time domain using the Hilbert transform of the stator currents and is free of the strong influence of the fundamental component. The key novelty of this proposal is the combination of the analytical current signals and the symmetrical components method, which produces a purely backward-rotating transient current signal that cannot be obtained using the raw phase current signals. This proposal is presented theoretically and validated in transient regime using a commercial induction motor with rotor asymmetries.

Phase Voltage-Based Diagnosis of Inter-Turn Short Circuits in Permanent Magnet Synchronous Motor Stator Windings

The problem of fault diagnosis in electrical motors has an important impact on the supervision of dynamic systems, and model-based methods are efficient tools for this purpose. In this regard, this work presents a novel method for detecting inter-turn short circuits (ITSCs) in the stator windings of permanent magnet synchronous motors (PMSMs). The approach is based on algebraic identification to process the motor voltage signals, estimating the offsets, amplitudes, and phases of the fundamental and third-harmonic components. Fault detection is performed in two steps: first, a voltage imbalance index is evaluated to determine the presence of abnormal operating conditions. Subsequently, characteristic patterns in the estimated parameters are analyzed to identify both the fault type and the affected phase(s). The experimental results show that single-phase ITSC faults produce a reduction in the offset of the faulted phase together with an increase in its third-harmonic amplitude, whereas phase-to-phase ITSC faults lead to an increase in the offsets of the affected phases and nearly identical third-harmonic amplitudes between them. In both cases, only minor variations are observed in the estimated phase angles. The effectiveness of the proposed methodology is supported through theoretical analysis and validated experimentally using voltage measurements acquired from a PMSM test bench. The results demonstrate that the proposed technique can accurately identify fault conditions through voltage imbalance and harmonic-pattern analysis, providing a practical and computationally efficient methodology for PMSM stator winding fault diagnosis.

Stator Fault Classifier in a Doubly-Fed Induction Generator DFIG

Failure Modes Analysis of High-Speed Permanent Magnet Generator Rotor

This paper proposes a reliability analysis method and process for the rotor of a high-speed permanent magnet generator. Firstly, the typical structure of the magnetic bearing rotor of the high-speed permanent magnet generator is analyzed, and the rotor reliability block diagram is established. Based on the reliability block diagram, a rotor fault tree is constructed, and fault description, cause analysis, and qualitative analysis are carried out on the fault tree. On the basis of the fault tree, failure mode, effects, and criticality analysis (FMECA) is performed to obtain the rotor FMECA table. Through the FMECA table, risk analysis and criticality analysis are conducted to identify high-risk items, and suggestions for design and process improvements are proposed.

Simulation-Driven Diagnosis of Stator Phase-Current Imbalance in Wind Turbine Induction Generators Using Multi-Signal Features and Explainable Machine Learning

Wind turbine reliability depends on timely identification of electromechanical faults, especially in generator-related subsystems under variable mechanical loads. This study presents a simulation-based, multi-signal, and physically interpretable diagnostic workflow for wind turbine electrical systems. It combines multiphysics simulation, FFT feature extraction, and explainable machine learning, emphasising the integration of existing methods rather than developing new AI models. A COMSOL Multiphysics (2D electromagnetic with 3D multibody dynamics) model of an induction machine simulated both healthy and imbalanced operating conditions with increasing stator phase-A current imbalance (parameter ε). The verified fault mechanism was incorporated into the model. From these simulations, a multisignal dataset was built using electromagnetic torque, rotor speed, electromagnetic force, and foundation force responses across 9 configurations, resulting in 54 samples (each with 100 features) classified into healthy, minor, and major imbalance groups. We tested Support Vector Machine, Multilayer Perceptron, and Random Forest algorithms. Repeated stratified cross-validation showed Random Forest performed best, with an average accuracy of 92.3% (±4.1%) and macro-F1 of 0.764 (±0.146). A leave-one-configuration-out test, where no data from the same configuration appear in both training and testing, produced more conservative results: 46.3% accuracy and 0.317 macro-F1, with no healthy-condition samples correctly classified, because only one independent healthy configuration was available. SHAP analysis identified foundation-force spectral energy in the 50-150 Hz range as the most important predictor, suggesting imbalance severity at the configuration level. Since foundation-force features are fixed within each configuration, this indicator should be considered a configuration-level marker rather than an individual sample marker. Overall, the sample-level results are promising, indicating that the multi-signal, physics-based feature set and interpretability are useful. However, the configuration-level results suggest that the current 9-configuration simulation setup is not yet sufficient for definitive diagnostic accuracy.

Analysis of Open-Circuit Fault Current of Wind Power Converters under Fluctuating Wind Speed

With the depletion of fossil fuels and the worsening of environmental pollution, wind energy has garnered widespread attention as a renewable energy source. Direct-drive permanent magnet wind turbines offer advantages such as high efficiency and a gearbox-free design; however, their power converters are prone to failure under fluctuating wind speed conditions, making research into fault diagnosis particularly significant. This paper focuses on IGBT open-circuit faults in wind power converters. A simulation model of a direct-drive permanent magnet wind power system was established, employing grid voltage-oriented vector control to simulate three types of faults: single-tube, double-tube, and out-of-phase double-tube open circuits. Three-phase currents were used as feature signals to analyze waveform distortion patterns. By extracting quantitative indicators such as RMS value, THD, and three-phase asymmetry to compare fault characteristics, and using the Pearson correlation coefficient to analyze the correlation between wind speed and the amplitude of these characteristics, the results indicate that faults significantly increase current distortion and asymmetry, with distinct differences among the various fault types. Wind speed shows a weak correlation with the amplitude of these fault characteristics, and the characteristics demonstrate good stability and robustness. This study provides theoretical and data support for diagnosing converter open-circuit faults under fluctuating wind speeds.