Müslüm Arkan

A new method to separate broken rotor failures and low frequency load oscillations in three-phase induction motor

For a time-varying loads, the presence of load fluctuation may sometimes have the same effect as broken rotor failure in a stator current of induction motors. In recent years, many methods, which are used to distinguish these two effects, have been published. In this study, a new method, which is based on Analytical Signal Angular Fluctuation (ASAF) signal, is developed to separate effects of load oscillation from broken rotor bar, especially when load oscillation is close to twice slip frequency. The simulation results are presented for the proposed method to show that, discerning broken rotor bar faults from low frequency load oscillation is possible. The developed method is independent of motor parameters.

Detection of Stator Winding Inter-Turn Short Circuit Faults in Permanent Magnet Synchronous Motors and Automatic Classification of Fault Severity via a Pattern Recognition System

In this study, automatic detection of stator winding inter-turn short circuit fault (SWISCFs) in surface-mounted permanent magnet synchronous motors (SPMSMs) and automatic classification of fault severity via a pattern recognition system (PRS) are presented. In the case of a stator short circuit fault, performance losses become an important issue for SPMSMs. To detect stator winding short circuit faults automatically and to estimate the severity of the fault, an artificial neural network (ANN)-based PRS was used. It was found that the amplitude of the third harmonic of the current was the most distinctive characteristic for detecting the short circuit fault ratio of the SPMSM. To validate the proposed method, both simulation results and experimental results are presented.

A new approach to detect stator fault in permanent magnet synchronous motors

In this paper, detection of the stator winding inter-turn short circuit fault (SWISCF) in surface-mounted permanent magnet synchronous motors (SPMSMs) and classification of the fault severity via pattern recognition system (PRS) are presented. In order to automatically detect stator winding short circuit fault and to estimate severity of this fault, artificial neural network (ANN) based PRS has been used. It has been observed that the amplitude of the 3rd harmonics of the current is the most distinctive characteristic for detecting the short circuit fault ratio of the SPMSM. To increase the fault clasification accuracy of PRS both fundamental (1st) and 3rd harmonics are used. In order to validate proposed method experimental results are presented.

Separation harmonics for detecting broken bar fault in case of load torque oscillation

This paper presents separation harmonics to discriminate rotor failure from low frequency load torque oscillations in three phase induction motors. The most common method for detecting broken rotor bar faults is to analyze the corresponding sidebands through motor current signature analysis (MCSA). If a motor is subjected to load fluctuation, then the oscillation related sidebands exhibit similar behaviors as well. Particularly, when the load fluctuation frequency is close or equal to that of broken bars, the stator current spectrum analysis can be misleading. In this study, torque and motor phase voltage waveforms are exhaustively analyzed to discriminate broken rotor bar fault from low frequency load torque oscillation in three phase induction motors. In order to extract and justify the separation patterns, 2-D Time Stepping Finite Element Method (TSFEM) is used. The simulation and experimental results show that the proposed approach can successfully be applied to fault separation process in star connected motors.

Broken rotor bar fault monitoring based on fluxgate sensor measurement of leakage flux

Broken rotor bar fault in induction motors significantly affects the motor dynamic performance and increases the mechanical oscillations in torque and speed. Motor current signature analysis has widely been used to detect such fault, yet it has some shortcomings due to motor topology, stator winding and load type dependencies. In this paper, radial leakage flux which contains most critical fault related information is analyzed using a fluxgate sensor to detect broken bar fault in induction motors (IMs). The 2D-Time Stepping Finite Element Method (2D-TSFEM) is used to analyze fault patterns in leakage flux. A 2-Dimensional (2D) finite element analysis and experimental results show that using leakage flux can provide superior and more reliable results than classical motor stator current analysis in IMs.

Diagnosis of broken rotor fault in inverter-fed IM by using analytical signal angular fluctuation

The aim of this paper is to detect broken rotor bar fault at the presence of low frequency load torque oscillation in inverter-fed induction motors. The low frequency load torque oscillation in induction motor may sometimes have the same effect as broken rotor bar fault on the stator current. Especially, when load torque oscillation frequency is close to twice the slip frequency, additional processing need to be done to separate these two effects from each other. To discern these two effects, Analytical Signal Angular Fluctuation (ASAF) spectrum is used. Experimental results are presented for separating broken rotor bar fault from low frequency load torque oscillation.

Discerning broken rotor bar failure from low-frequency load torque oscillation in DTC induction motor drives

This paper proposes a method for separation of broken rotor bar failures from low-frequency load torque oscillation in direct torque control (DTC) induction motor drives by using vq voltage and iq current components’ spectra. The effect of load torque oscillation should be considered in induction motor drives for reliable broken bar fault detection. Induction machine drivers are run in DTC mode to control its torque and speed. In practice, the presence of load torque fluctuation may sometimes cause false positive alarms on stator current spectrum. However, discerning of broken rotor bar failure from low-frequency load variation for DTC drives remains unexplored. Experimental results show that by using the proposed method broken rotor bar failure can be reliably detected in the presence of low-frequency load torque oscillation in DTC induction motor drives.

Separation of induction motor rotor faults and low frequency load oscillations through the radial leakage flux

This paper presents a separation method to discern broken rotor bar fault from low frequency load torque oscillation thorough radial leakage flux spectrums in induction motors (IMs). Broken rotor bar fault can usually be detected using classical motor current based analysis (MCSA), but it may not provide reliable results since its performance depend on motor topology, stator winding and load type. Particularly, if a motor is subjected to load fluctuation, then oscillation related signatures exhibit similar behavior that of broken bar which leads misleading signatures. In this paper, radial leakage flux spectrum is exhaustively analyzed thorough a fluxgate sensor to discern these two effects in IMs. It is shown that there are some additional characteristics broken bar signatures such as 3sfs and (fs-fr)±2sfs in radial leakage flux which do not appear in low frequency load torque oscillation case. A 2-Dimensional (2D) finite element analysis and experiments are carried to show that using leakage flux can provide a separation method and more reliable results than classical MCSA in IMs.

Fault location determination for transmission lines with different series-compensation levels using transient frequencies

In this paper, based on the theory of traveling waves, the fault distances on long transmission lines with various series-compensation levels are determined using transient current and voltage frequencies. Transmission lines with series compensation are modeled using Alternative Transients Program software with frequency-dependent effects on the line included in the simulation. The transient current and voltage signals are obtained from the model. A fast Fourier transform is used for frequency-domain conversion and fault location is estimated from the frequencies of fault-generated harmonics in the transient spectrum. The algorithm is implemented in MATLAB. To investigate the effect of compensation on accuracy, the results are obtained for different series-compensation levels. The undesirable source-inductance effect is removed and estimation accuracy is further improved using a waveform-relaxation method. The method is found to be successful in determining fault location on series-compensated transmission lines. The effects of the compensation level, fault resistance, and phase angle are investigated.

Analysis of stator inter-turn short-circuit fault signatures for inverter-fed permanent magnet synchronous motors

It is quite important to detect stator short-circuit fault, which is the most common fault type, at incipient stage. It is possible to carry out fault detection using Motor Current Signature Analysis (MCSA) method. In this study, stator current and voltage space vectors of PMSMs were analyzed with MCSA under various load torque, speed and fault percentages conditions. The Negative & positive harmonics were obtained by applying Fast Fourier Transform (FFT) to space vectors of stator current and voltage. It is suggested that by using the obtained fault signatures, stator inter-turn fault estimation can be achieved accurately. The results of comprehensive analysis carried out under various load torque and speed conditions show that characteristics fault signatures are both present in the current and the voltage space vectors spectra.