Motor fault diagnosis method based on broadband traveling wave reflection technology
By combining broadband traveling wave reflection technology with fault database and hardware system integration, the challenges of location and pattern recognition in motor fault diagnosis have been solved, enabling accurate detection and identification of motor coil faults and improving the reliability of equipment operation.
Patent Information
- Application Number
- CN202511739875.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-25
- Publication Date
- 2026-02-17
AI Technical Summary
Existing motor fault diagnosis methods cannot accurately locate and identify the fault modes of motor coils, and are easily affected by external electrical noise, resulting in inaccurate test results.
By employing broadband traveling wave reflection technology, combined with fault database development, hardware system integration, and software model development, motor fault diagnosis is performed using high-frequency pulse signals. A fault database is established using simulation and experimental simulation. Power supply module, high-frequency signal generation module, measurement module, control module, and display module are integrated to perform fault location and pattern recognition.
It enables precise location and pattern recognition of motor coil faults, improving equipment reliability and reducing economic losses.
Smart Images
Figure CN121541049A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of motor fault diagnosis, and particularly relates to a motor fault diagnosis method based on wideband traveling wave reflection technology. BACKGROUND
[0002] As a key power source in modern industrial manufacturing and life production, the stable operation of a motor is crucial for ensuring production safety and efficiency. However, various faults such as turn-to-turn short circuit, phase-to-phase creepage, slot wedge loosening, and local overheating inevitably occur during long-term operation. If these faults are not discovered and addressed in a timely manner, they may cause equipment damage, affect operational efficiency, and result in significant economic losses.
[0003] Traditional motor testing and diagnosis methods, such as insulation resistance testing and power frequency dielectric loss testing, can only assess whether the insulation performance is deteriorating, but cannot perform fault diagnosis. Although partial discharge testing is sensitive to partial defects, it is easily disturbed by external electrical noise during testing. In addition, due to the high inductance characteristics of motor coils, partial discharge pulse signals are prone to distortion and attenuation during propagation, making it difficult to accurately measure internal defects.
[0004] As a new motor fault diagnosis method, wideband traveling wave reflection technology provides a new solution for motor fault detection with its low voltage, non-invasive, and non-destructive characteristics. This technology was initially applied to cable fault diagnosis. In recent years, with the advancement of precision measurement technology, the measurement accuracy of testing instruments has been significantly improved, enabling the technology to be applied to the measurement of unshielded test objects. This technology can diagnose various common faults of motor coils by transmitting wideband pulse signals and acquiring their reflected waves. Although wideband traveling wave reflection technology has shown great potential in the field of motor fault diagnosis, there are still some challenges and room for improvement. For example, how to achieve accurate motor fault location and fault pattern recognition. Therefore, developing a more accurate and effective motor fault diagnosis method based on wideband traveling wave reflection technology is of great significance for improving motor operation reliability and reducing unnecessary economic losses. SUMMARY
[0005] The purpose of the present application is to provide a motor fault diagnosis method based on wideband traveling wave reflection technology, which can solve the problem of quantitative evaluation of motor coil faults.
[0006] The technical scheme of the present application is as follows: a motor fault diagnosis method based on wideband traveling wave reflection technology, which comprises three parts of fault library development, hardware system integration and software model development, the fault library is used for storing and managing motor wideband reflection test data, supporting label processing of the data, providing data basis for subsequent feature extraction and pattern recognition, the fault library development is realized by combining simulation simulation and test simulation, the simulation simulation obtains high-frequency pulse reflection signal spectrum by simulating the normal operation state of the motor and various fault conditions, the test simulation verifies the effectiveness of the simulation simulation by setting normal state and different fault states in the motor coil and collecting high-frequency pulse reflection signals.
[0007] The hardware system integration side comprises a power module, a high-frequency signal generation module, a measurement module, a control module and a display module.
[0008] The power module is used for providing stable AC power supply to ensure continuous and stable power supply of the system in complex working environment.
[0009] The high-frequency signal generation module generates high-precision high-frequency pulse signals, and the high-frequency pulse signals will produce reflection when transmitted in the motor coil, and the reflection signal is the output response of the measured system.
[0010] The measurement module is equipped with high-sensitivity sensors and precise measurement circuits to detect subtle changes in pulse reflection signals, thereby accurately evaluating the fault state of the motor.
[0011] The control module is built-in microprocessor, which analyzes and processes the data collected by the measurement module.
[0012] The display module adopts high-resolution OLED display screen to display the motor fault diagnosis results, including fault position, fault type and related statistical data.
[0013] The software model comprises four functions of test, analysis, calibration and simulation simulation, the test function tests by software control to emit high-frequency pulse signals to obtain the reflection waveform data of the measured motor coil, the analysis function processes the data obtained by testing, and the state characteristics of the motor coil loop are displayed through the feature extraction method, the calibration function is mainly used for positioning the full length of the coil and the internal circuit detection of the system, to ensure that the pulse waveform can cover the full length of the coil and the internal measurement circuit can work normally, and the simulation simulation function is used for simulating multiple categories of motor coil operating states, through inputting known parameters, different operating states of the motor can be simulated and simulated, and the output data of the simulation model are evaluated and judged.
[0014] The method can be based on high-frequency pulse reflection signals, combined with the established fault diagnosis database, to automatically locate faults and identify fault modes of typical faults such as motor coil turn-to-turn short circuit, phase-to-phase creepage, slot wedge loosening, and local overheating, so that field operation personnel can timely perceive the equipment state and improve the equipment operation reliability. BRIEF DESCRIPTION OF DRAWINGS
[0015] Figure 1 A motor fault diagnosis method development diagram based on wideband traveling wave reflection technology is provided.
[0016] Figure 2 A hardware system integration diagram is provided.
[0017] Figure 3 A motor wideband traveling wave reflection test principle diagram is provided.
[0018] Figure 4 A motor fault detection method result display example diagram is provided.
[0019] Figure 5 A turn-to-turn circuit test embodiment result diagram is provided.
[0020] Figure 6 A phase-to-phase creepage test embodiment result diagram is provided.
[0021] Figure 7 A slot wedge loosening test embodiment result diagram is provided.
[0022] Figure 8 A local overheating test embodiment result diagram is provided. DETAILED DESCRIPTION
[0023] The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0024] The application provides a motor fault diagnosis method based on wideband traveling wave reflection technology, which has two functions of fault positioning and fault mode identification. It includes three parts: fault library development, hardware system integration and software model development. The fault library is used to store and manage motor wideband reflection test data, supports data labeling, and provides data basis for subsequent feature extraction and pattern recognition. The fault library development is mainly through the combination of simulation and test simulation. Simulation simulation obtains high-frequency pulse reflection signal atlas by simulating the normal operation state and various fault conditions of the motor through accurate mathematical models and algorithms. The accuracy of the simulation model can be verified by comparing the test data. Test simulation mainly sets normal state and different fault states in the motor coil, acquires high-frequency pulse reflection signals to verify the effectiveness of simulation simulation, and then completes the fault library development.
[0025] In terms of hardware system integration, the motor fault diagnosis method designed by the application integrates multiple functional modules, including a power module, a high-frequency signal generation module, a measurement module, a control module, and a display module. The power module is used to provide stable AC power to ensure continuous and stable power supply in complex working environments. The high-frequency signal generation module uses a main oscillator stage technology to generate high-precision high-frequency pulse signals. When the high-frequency pulse signals are transmitted inside the motor coil, they will produce reflections, and the reflected signals are the output responses of the measured system. The measurement module is equipped with high-sensitivity sensors and precise measurement circuits, which can detect subtle changes in pulse reflection signals, thereby accurately evaluating the fault state of the motor. The control module, as the core of the system, is built-in with a microprocessor that analyzes and processes the data collected by the measurement module through algorithm calculation. The display module uses a high-resolution OLED display screen to display the motor fault diagnosis results in a visual and clear manner, including fault location, fault type, and related statistical data.
[0026] The software model proposed by the application integrates four functions of testing, analysis, calibration, and simulation. The testing function tests by controlling the emission of high-frequency pulse signals to obtain the reflected waveform data of the measured motor coil. The analysis function processes the test data and displays the state characteristics of the motor coil loop in a more intuitive way through feature extraction methods such as step response, pulse response, and pulse amplitude. The calibration function is mainly used for positioning the full length of the coil and detecting the internal circuit of the system to ensure that the pulse waveform can cover the full length of the coil and the internal measurement circuit can work normally, thereby improving the testing accuracy. The simulation function is used to simulate multiple categories of motor coil operating states. By inputting known parameters, the system can simulate different operating states of the motor and make evaluation judgments based on the output data of the simulation model. Simulation mainly simulates working conditions to achieve equivalent replacement of the actual physical model. The fault diagnosis method of the application has two functions of motor coil fault positioning and fault pattern recognition. Fault positioning is based on transmission line theory, which determines the abnormal point position in the coil loop by analyzing the distortion point of the impedance wave. Fault recognition is based on the fault diagnosis database built by iterative calculation and comparison, and the specific fault type is determined by the fitting optimization method.
[0027] Embodiment:
[0028] A motor fault diagnosis method based on wideband traveling wave reflection technology, comprising:
[0029] 1. Fault library development:
[0030] The development of the fault library mainly includes two parts of simulation model development and test platform building. The simulation modeling is mainly based on the COMSOL Multiphysics finite element simulation theory to build a motor structure simulation model. By setting different parameters, the normal and fault running states of the motor are simulated, and the fault simulation mainly includes turn-to-turn short circuit of the stator coil and the rotor coil, phase-to-phase creepage, slot wedge loosening, local overheating and other typical faults. Based on the simulation and analysis under the normal and fault conditions, the reference output atlas is obtained and the characteristic markers are marked. Under the laboratory conditions, by pre-embedding defects in the coil and injecting high-frequency pulse signals, the output atlas of the motor stator coil and rotor coil under the normal and fault (typical faults such as turn-to-turn short circuit, phase-to-phase creepage, slot wedge loosening, local overheating and the like) two states is obtained, and is compared with the output atlas in the simulation and analysis, so as to verify the accuracy of the simulation model, and then the development of the fault library is completed.
[0031] 2. Hardware system integration:
[0032] The hardware system integration diagram is shown as Figure 2 The part is composed of a power module, a high-frequency voltage signal generation module, a measurement module, a control module and a display module.
[0033] The power module is used to provide stable AC power output to ensure continuous and stable power supply to each module in a complex working environment. The high-frequency voltage signal generation module adopts a main oscillator level technology, and generates high-precision high-frequency pulse signals for the subsequent measurement module through frequency modulators, oscillators and internal modulation oscillators and other devices. The measurement module is equipped with high-sensitivity sensors and precise measurement circuits, which can accurately detect the high-frequency pulse signals reflected by the motor coil, and then collect the signals through a data acquisition card to accurately evaluate the working state of the motor. The control module is built-in microprocessor, which can execute complex algorithms to analyze and process the data collected by the measurement module. The display module displays the fault diagnosis results analyzed and processed by the control module in a clear and intuitive way through a high-resolution OLED display screen, including fault location, fault type and related statistical data.
[0034] 3. Software model development:
[0035] The software model of the application is developed based on C language, which integrates four function modules of test, analysis, calibration and simulation. It has the functions of controlling the operation of the hardware system, storing and calling the fault library, analyzing and comparing the test results, fault positioning and fault mode recognition, engineering management and the like.
[0036] The testing function generates high-frequency pulse signals from the aforementioned hardware system to test the coil circuit of the motor under test, obtaining the high-frequency pulse reflection signal of the motor coil. The analysis function automatically iteratively calculates and optimizes the features of the obtained high-frequency pulse reflection signal, comparing it with a fault database. The system can automatically determine whether a fault has occurred in the motor coil through step response, pulse response, and pulse amplitude. If a fault is found, the system automatically displays the fault location and fault type. The calibration function is used to locate the entire length of the coil and detect internal circuitry, ensuring that the pulse waveform covers the entire length of the coil and that the internal measurement circuitry is functioning correctly, thereby ensuring the effectiveness and accuracy of the test. The simulation function obtains high-frequency pulse reflection signals of the motor coil under various operating states by inputting initial parameters. The simulation process includes adaptive convergence calculation and mirror analysis to achieve virtual-real interaction with the physical model.
[0037] This invention utilizes broadband traveling wave reflection technology for motor fault diagnosis. Its core principle involves injecting a high-frequency pulse signal. When a fault occurs in the motor coil, the high-frequency pulse signal is distorted at that point. By analyzing the reflected signal and using Fourier transform to convert it into a time-domain signal, fault location and fault mode identification can be performed. The specific testing principle is as follows... Figure 3 As shown, the positive and negative terminals of the high-frequency signal source are connected to one phase winding port and the outer shield of the test object, respectively. The high-frequency pulse reflection signal is collected by the signal receiver. When the motor coil fails, the signal will be distorted at the fault point. The system calculates the wave impedance and phase of the tested motor coil based on the amplitude and phase of the collected high-frequency pulse reflection signal. Finally, the frequency-wave impedance spectrum is converted into a time-reflection amplitude curve through inverse Fourier transform.
[0038] After performing an inverse Fourier transform on the high-frequency pulse reflection signal, we can obtain Figure 4 The test results are shown. Figure 4 The step response shown reflects the overall and local aging and degradation of the coil by calculating the wave impedance information. Figure 4 The pulse response and pulse amplitude display methods shown mainly highlight the changes at the boundary between the two characteristics. When a coil has a fault, the curve will be distorted at the fault point. Both the pulse response and pulse amplitude curves will reflect different fault problems of the motor coil. The pulse amplitude is the logarithmic form of the pulse response, which can magnify tiny details.
[0039] Figures 5 to 8The diagram illustrates the results of an embodiment of fault detection for a motor coil. The test subject was a standard coil of a 10kV motor. This coil has 6 turns, each consisting of two copper strands. The slotted area of the coil has a semiconductor anti-corona layer, and during the test, an external copper shielding layer replaces the function of the iron core. The slotted area is unshielded. An equalizing layer is present at the boundary between the slotted and slotted areas. The length of each conductor turn in the slotted area is 2.8 meters, and the length in the slotted area is 2 meters, for a total conductor length of approximately 28.8 meters. The two ends are port 1 and port 2. The specific defect simulation method and test results are as follows:
[0040] Inter-turn short circuit: Two turns of conductor located 2.4 meters and 7.2 meters away from port 1 are used to form a low-resistance path, and then tests are performed at port 1 and port 2 respectively. FDR tests are performed at port 1 and port 2 respectively, and the results are as follows. Figure 5 As shown, the fault distance obtained from the test results of port 1 is 2.5 meters, and the fault distance obtained from the test results of port 2 is 21 meters. Therefore, this method can be used for fault diagnosis of inter-turn circuit breaks.
[0041] Phase-to-phase creepage: Two coils are connected in series and form a low-resistance path with the shielding layer at a midpoint of 37.3m to simulate phase-to-phase creepage faults. The test results are as follows: Figure 6 As shown in the fault diagram, the phase-to-phase creepage fault is located at 37m. Therefore, this method can be used to diagnose phase-to-phase creepage faults.
[0042] Slot wedge loosening: The shielding layer, located 4.5 meters from port 1, was loosened by 20 cm to simulate slot wedge loosening. Test results are as follows. Figure 7 As shown in the figure, the location of the loosening fault of the slot wedge is at 4.2m. Therefore, this method can be used to diagnose the loosening of the slot wedge.
[0043] Localized overheating: A high-temperature spray gun was used to heat a location 3.5 meters away from port 1 for 2 minutes to simulate localized overheating. Test results are as follows: Figure 8 As shown in the figure, the distance-amplitude curve shows a difference at 3.3m. Therefore, this method can be used to diagnose local overheating.
[0044] Obviously, the embodiments described above are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.
Claims
1. A method for motor fault diagnosis based on broadband traveling wave reflection technology, characterized in that: The three parts include fault library development, hardware system integration and software model development, the fault library is used for storing and managing motor broadband reflection test data, supporting data labeling, providing data basis for subsequent feature extraction and pattern recognition, fault library development is realized by combining simulation and test simulation, simulation obtains high-frequency pulse reflection signal atlas by simulating normal operation state and various fault conditions of the motor, test simulation verifies the effectiveness of simulation by setting normal state and different fault states in the motor coil and collecting high-frequency pulse reflection signals.
2. The motor fault diagnostic method based on broadband traveling wave reflection technology according to claim 1, characterized in that: The hardware system integration includes power module, high-frequency signal generation module, measurement module, control module and display module.
3. A motor fault diagnostic method based on broadband traveling wave reflection technology as claimed in claim 2, characterized in that: The power module is used to provide stable AC power supply to ensure continuous and stable power supply in complex working environment.
4. The motor fault diagnostic method based on broadband traveling wave reflection technology according to claim 2, characterized in that: The high-frequency signal generation module generates high-precision high-frequency pulse signals, which will produce reflection when transmitted in the motor coil, and the reflected signal is the output response of the measured system.
5. The motor fault diagnostic method based on broadband traveling wave reflection technology as claimed in claim 2, wherein: The measurement module is equipped with high-sensitivity sensors and precise measurement circuits to detect subtle changes in pulse reflection signals and accurately assess the fault state of the motor.
6. A motor fault diagnostic method based on broadband traveling wave reflection technology as claimed in claim 2, characterized in that: The control module is built-in microprocessor, which analyzes and processes the data collected by the measurement module.
7. The motor fault diagnostic method based on broadband traveling wave reflection technology as claimed in claim 2, wherein: The display module uses high-resolution OLED display screen to display motor fault diagnosis results, including fault location, fault type and related statistical data.
8. The motor fault diagnostic method based on broadband traveling wave reflection technology as claimed in claim 1, wherein: The software model includes four functions of test, analysis, calibration and simulation, the test function tests by emitting high-frequency pulse signals controlled by software to obtain the reflection waveform data of the measured motor coil, the analysis function processes the test data, and the state characteristics of the motor coil loop are displayed through feature extraction method; The calibration function is mainly used for positioning the full length of the coil and detecting the internal circuit of the system to ensure that the pulse waveform can cover the full length of the coil and the internal measurement circuit can work normally; the simulation function is used to simulate the multi-class operation state of the motor coil, by inputting known parameters, the different operation states of the motor can be simulated, and the output data of the simulation model are used for evaluation and judgment.