Jérémi Regnier

On-Line Monitoring of Mechanical Faults in Variable-Speed Induction Motor Drives Using the Wigner Distribution

This paper deals with the detection of mechanical load faults in induction motors during speed transients. The detection strategy is based on stator current analysis. Mechanical load faults generally lead to load torque oscillations at specific frequencies related to the mechanical rotor speed. The torque oscillations produce a characteristic sinusoidal phase modulation of the stator current. Speed transients result in time-varying supply frequencies that prevent the use of classical, Fourier transform-based spectral estimation. This paper proposes the use of a time-frequency distribution, the Wigner Distribution, for stator current analysis. Fault indicators are extracted from the distribution for on-line condition monitoring. The proposed methods are implemented on a low-cost digital signal processor. Experimental results in a steady-state and during transients with load torque oscillations and load imbalance are presented.

Distinguishing Load Torque Oscillations and Eccentricity Faults in Induction Motors Using Stator Current Wigner Distributions

This paper proposes a novel diagnosis method for detection and discrimination of two typical mechanical failures in induction motors by stator current analysis: load torque oscillations and dynamic rotor eccentricity. A theoretical analysis shows that each fault modulates the stator current in a different way: torque oscillations lead to stator current phase modulation, whereas rotor eccentricities produce stator current amplitude modulation. The use of traditional current spectrum analysis involves identical frequency signatures with the two fault types. A time-frequency analysis of the stator current with the Wigner distribution leads to different fault signatures that can be used for a more accurate diagnosis. The theoretical considerations and the proposed diagnosis techniques are validated on experimental signals.

Comparison Between Stator Current and Estimated Mechanical Speed for the Detection of Bearing Wear in Asynchronous Drives

This paper deals with the detection of worn rolling bearings in asynchronous machines using electrical measurements and estimated mechanical variables. These two approaches are based on the use of the available electrical quantities, e.g., the machine stator currents, which are often already measured for control and protection purposes. Considering that bearing faults induce load-torque oscillations, a theoretical stator-current model, in case of load-torque oscillations, is recalled. Then, a theoretical estimated rotor flux and estimated speed model demonstrates the presence of harmonics related to load-torque oscillations. Phase-modulation components on stator current and harmonics on estimated speed can be used for detection purposes. The frequency behavior of monitored quantities with regard to the load-torque oscillation frequency is particularly investigated. Fault detectors are then proposed on monitored variables. The efficiency of the indicators is studied for different operating points considering the frequency behavior of the system. Finally, the use of detectors is discussed regarding the supply-frequency range usable for the detection.

Mechanical Fault Detection in Induction Motor Drives through Stator Current Monitoring - Theory and Application Examples

In a wide variety of industrial applications, an increasing demand exists to improve the reliability and availability of induction motor drives. Common failures occurring in such drives can be classified into electrical and mechanical faults (rotor eccentricity, bearing faults, shaft misalignment, load unbalance, gearbox fault or general failure in the load part of the drive). Mechanical faults are most commonly detected through vibration or noise monitoring, but stator current monitoring is an interesting alternative. Indeed, current sensors are cost-effective, easy to implement, and most drives already contain such sensors for protection and control purposes. However, the effects of mechanical faults on the stator currents are more indirect compared to vibration or noise analysis. This work focuses on various aspects of mechanical fault detection through stator current monitoring, starting from a general theoretical analysis to signal processing methods for fault detection and several application examples.

Recombination and self-adaptation in Multi-Objective Genetic Algorithms

This paper investigates the influence of recombination and self-adaptation in real-encoded Multi-Objective Genetic Algorithms (MOGAs). NSGA-II and SPEA2 are used as example to characterize the efficiency of MOGAs in relation to various recombination operators. The blend crossover, the simulated binary crossover and the breeder genetic crossover are compared for both MOGAs on multi-objective problems of the literature. Finally, a self-adaptive recombination scheme is proposed to improve the robustness of MOGAs.

Current-Based Detection of Mechanical Unbalance in an Induction Machine Using Spectral Kurtosis With Reference

This paper explores the design, online, of an electrical machines healthy reference by means of statistical tools. The definition of a healthy reference enables the computation of normalized fault indicators whose value is independent of the systems characteristics. This is a great advantage when diagnosing a broad range of systems with different power, coupling, inertia, load, etc. In this paper, an original method called spectral kurtosis with reference is presented to design a systems healthy reference. Its principle is first explained on a synthetic signal. This approach is then evaluated for mechanical unbalance detection in an induction machine using the stator current instantaneous frequency. The normalized behavior of the proposed indicator is then confirmed for different operating conditions, and its robustness with respect to load variations is demonstrated. Finally, the advantages of using a statistical indicator based on a healthy reference compared with a raw fault signature are discussed.

Indicator for bearing fault detection in asynchronous motors using stator current spectral analysis

This paper deals with the application of motor current spectral analysis for the detection of artificially damaged rolling bearings in asynchronous machine. Vibration monitoring of mechanical characteristic frequencies related to the bearings is widely used to detect faulty operations. However, vibration measurement is expensive and can not always be performed. An alternative is to base the monitoring on the available electrical quantities e.g. the machine stator current which is often already measured for control and protection purposes. The bearing faults reveal the presence of mechanical load torque oscillations. A theoretical stator current model in case of load torque oscillations demonstrates the presence of phase modulation. Related sideband components appear in the current spectrum and can be used for detection. Experimental measurements show that their amplitudes are linked to the fault frequency by a transfer function including resonance. This singularity will be used to improve the detection efficiency. Fault detectors using the energy of stator current in specific frequency ranges are then proposed. The efficiency of indicators is studied on long and short data records of experimental current for different bearing faults. The most significant of the investigated indicators is finally improved to guarantee a higher reliability of the detection.

System optimization by multiobjective genetic algorithms and analysis of the coupling between variables, constraints and objectives

Purpose – This paper presents a methodology based on Multiobjective Genetic Algorithms (MOGAs) for the design of electrical engineering systems. MOGAs allow one to optimize multiple heterogeneous criteria in complex systems, but also simplify couplings and sensitivity analysis by determining the evolution of design variables along the Pareto‐optimal front.Design/methodology/approach – To illustrate the use of MOGAs in electrical engineering, the optimal design of an electromechanical system has been investigated. A rather simplified case study dealing with the optimal dimensioning of an inverter – permanent magnet motor – reducer – load association is carried out to demonstrate the interest of the approach. The purpose is to simultaneously minimize two objectives: the global losses and the mass of the system. The system model is described by analytical model and we use the MOGA called NSGA‐II.Findings – From the extraction of Pareto‐optimal solutions, MOGAs facilitate the investigation of parametric sensi...

Detection of turn short-circuit faults in stator of PMSM by on-line parameter estimation

In this paper, a new technique of on-line detection of inter-turn short circuits in stator of permanent magnet synchronous motor (PMSM) is proposed. The study is focused on a PMSM with non sinusoidal electromotive force (emf). The technique is formulated by combining extended Parkpsilas vector approach and classical recursive least squares algorithm (RLS). The faults are detected by analysing the abnormal variation of electrical estimated parameters of the 2-phases model in the Extended Park frame.

Detection of mechanical load faults in induction motors at variable speed using stator current time-frequency analysis

This paper examines the detection of mechanical load faults in induction motors during speed transients by stator current analysis. Mechanical load faults generally lead to load torque oscillations at specific frequencies. these frequencies are related to the mechanical rotor speed. The torque oscillations produce a characteristic sinusoidal phase modulation of the stator current. Speed transients result in time-varying supply frequencies that prevent the classical, Fourier transform based spectral estimation. This problem can be overcome using time-frequency signal analysis. The methods applied in this paper are instantaneous frequency estimation and the Wigner Distribution. Furthermore, an adaptive demodulation method is proposed. The theoretical considerations are validated on signals obtained from an experimental setup.