Induction Motor Fault Detection and Diagnosis (Simulink)
💥1 Overview
Induction motors are the most widely used devices in industrial production processes, and their operating conditions have a great impact on the safe operation of industrial production. The occurrence of faults not only leads to economic losses, but also brings huge casualties and social impacts. Therefore, how to conduct online status monitoring of induction motors to timely detect early abnormalities of motors and avoid motor failures and failures has become a hot topic in the research of induction motor fault diagnosis in recent years. The induction motor system can be divided into circuit system, insulation system, magnetic circuit system, mechanical system, ventilation and cooling system, etc. Any system that works poorly or does not work properly will cause motor failure. According to statistical surveys [1], winding faults and bearing faults of induction motors account for 46% and 40% of motor failures respectively, while other faults account for 14%. Therefore, the fault detection of induction motors mainly focuses on three parts: stator faults, rotor faults and bearing faults. Among them, the winding faults of the stator coils and rotor coils of induction motors are one of the main causes of motor failure. Therefore, the design algorithm proposed in this paper is mainly used for the fault detection of stator winding and rotor winding inter-turn short circuits.
To detect faults, we should have a clear understanding of the nature of the faults. The following is an analysis of the fault mechanism of the stator and rotor windings. The main causes of stator winding failures are insulation failure caused by wear, pollution, cracks, corrosion, etc. and displacement of slot wire bars caused by mechanical vibration. The main cause of rotor winding failures is insulation damage caused by the influence of comprehensive factors such as electrical, thermal, mechanical stress and external environment during operation. Among the above faults, the short-circuit fault between winding turns is one of the most common and dangerous faults. Assume that the stator winding and rotor winding of the three-phase induction motor have N turns, and the stator winding has Ni turns with turn-to-turn short circuit. The rotor winding in each phase is connected in parallel, and n turns of each phase are short-circuited. In the case of a stator winding fault, the stator resistance, stator self-inductance and even the mutual inductance between the stator and the rotor will change. For simplicity, only the change of stator resistance is considered. Generally speaking, in the case of a short-circuit between the rotor winding turns, the stator self-inductance, rotor self-inductance and the mutual inductance between them will change.
will not change, but the rotor resistance will change［18］.
Research on Fault Detection and Diagnosis of Induction Motors
As one of the most widely used equipment in industrial production, the operating status of induction motors has an important impact on the safety and efficiency of production lines. Therefore, fault detection and diagnosis of induction motors and timely detection and treatment of potential problems are the key to ensuring smooth production. The following is a detailed study of induction motor fault detection and diagnosis.
1. Types of induction motor faults
There are many types of induction motor failures, but they can generally be divided into two categories: electrical failures and mechanical failures. Electrical failures mainly include stator winding failures, rotor failures, bearing failures, etc.; mechanical failures involve bearing damage, excessive mechanical vibration, etc. Among them, stator winding and rotor winding failures are the most common types of induction motor failures, accounting for the majority of motor failures.
2. Fault Detection and Diagnosis Methods
1. Offline detection and diagnosis
Offline detection refers to fault detection performed when the motor is stopped. This method usually requires the motor to be removed from the system for detailed inspection and testing. The advantage of offline detection is that the motor can be fully inspected, but the disadvantage is that it requires shutdown, which affects production.
2. Online monitoring technology
Online monitoring technology is a fault detection method that is performed when the motor is in normal operation. This method measures the motor’s operating parameters such as voltage, current, power, magnetic flux, torque, speed, temperature, etc., and extracts fault characteristics by combining signal analysis and processing technology. The advantage of online monitoring technology is that it can detect faults without affecting production, detect early faults as early as possible, and take measures to avoid serious faults.
3. Application of blind deconvolution method in fault detection
Blind deconvolution is a signal processing technique that performs a deconvolution operation on the observed signal to restore the original signal or system parameters. In induction motor fault detection, blind deconvolution can be used to process vibration signals, current signals, etc. during motor operation to extract fault features.
1. Minimum Entropy Deconvolution (MED)
Minimum entropy deconvolution is a deconvolution method based on signal sparsity. It estimates the impulse response of the system by minimizing the entropy of the output signal, thereby extracting fault features. In induction motor fault detection, MED can be used to process vibration signals, identify abnormal shocks or vibrations in the motor, and then determine the fault type and location.
2. Maximum Correlation Kurtosis Deconvolution (MCKD)
Maximum correlation kurtosis deconvolution is a method that uses the correlation and kurtosis characteristics of a signal for deconvolution. It estimates the impulse response of the system by maximizing the correlation kurtosis of the output signal. In induction motor fault detection, MCKD can be used to process current signals or vibration signals and identify fault features by extracting the periodic impulse components in the signal.
3. Maximum Second-Order Ring-Stationary Blind Deconvolution (SOBSS)
Maximum second-order ring stationary blind deconvolution is a blind deconvolution method for non-stationary signals. It uses the second-order statistical characteristics of the signal (such as the covariance matrix) for deconvolution operations. In induction motor fault detection, SOBSS can be used to process vibration signals or current signals under complex working conditions, and identify the fault type and location by extracting non-stationary features in the signal.
IV. Conclusion and Outlook
Induction motor fault detection and diagnosis is an important part of ensuring the smooth progress of industrial production. With the continuous development of signal processing technology, the application prospects of blind deconvolution methods in induction motor fault detection are broad. In the future, we can further study the advantages and disadvantages and applicable scope of different blind deconvolution methods, as well as their specific application effects in different types of induction motor fault detection. At the same time, combining advanced technologies such as artificial intelligence and big data to build a more intelligent and efficient induction motor fault detection and diagnosis system will be the future development direction.
📚 2 Operation results
🎉3 References
Some theories come from the Internet. If there is any infringement, please contact us to delete them.
[1] Xia Li, Fei Qi. Research on the detection method of induction motor bearing fault[J]. Vibration, Testing and Diagnosis, 2005(04):307-310+321.DOI:10.16450/j.cnki.issn.1004-6801.2005.04.015.
[2] Zhang Changfan, Huang Yishan, Shao Rui. Induction motor fault detection method and application based on observer[J]. Chinese Journal of Scientific Instrument, 2011, 32(06): 1337-1343. DOI: 10.19650/j.cnki.cjsi.2011.06.021.
[3]hariharan mahalingam (2023). induction motor fault detection
