Three-phase motor fault diagnosis system and method based on magnetic temperature vibration electric coupling

Abstract

This invention relates to the field of motor condition monitoring and fault diagnosis technology, and discloses a three-phase motor fault diagnosis system and method based on magneto-thermal-oscillatory-electric coupling. The system includes a signal acquisition component installed at the junction box of the three-phase motor and a diagnostic processing component connected to the signal acquisition component. This three-phase motor fault diagnosis system and method based on magneto-thermal-oscillatory-electric coupling synchronously acquires three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals at the junction box of the three-phase motor. Based on the fundamental frequency of the magnetic field, the real-time rotational frequency of the motor is determined. The vibration signal is then processed using this real-time rotational frequency, and simultaneously, magneto-electric coupling characteristics, magneto-oscillatory coupling characteristics, and thermal correlation coupling characteristics are generated by combining electrical parameter characteristics and temperature rise trend characteristics. This allows various detection signals to form mutually corroborating diagnostic evidence under the same operating conditions.

Description

Technical Field

[0001] This invention relates to the field of motor condition monitoring and fault diagnosis technology, and in particular to a three-phase motor fault diagnosis system and method based on magneto-thermal-oscillatory-electric coupling. Background Technology

[0002] Three-phase motors are commonly used power equipment in industrial production. During operation, abnormalities such as rotor bar breakage, winding turn-to-turn short circuits, air gap eccentricity, bearing damage, overload overheating, and poor contact may occur. Existing motor fault diagnosis technologies typically use vibration sensors to collect mechanical vibration signals, detect electrical operating status through current and voltage, or monitor motor heating through temperature sensors, and combine these with spectrum analysis, threshold judgment, or intelligent algorithms to identify the motor's condition.

[0003] However, the main problem with existing technologies is that most of the detection signals are only collected, analyzed independently, or simply fused, lacking a unified utilization of the inherent correlation between magnetic fields, temperature, vibration, and electrical parameters. This leads to the easy misjudgment of normal operating conditions as faults under complex conditions such as startup, frequency conversion, and load fluctuations, or to missed diagnoses due to weak early fault characteristics, thus affecting the accuracy and reliability of three-phase motor fault diagnosis. Summary of the Invention

[0004] The technical problem to be solved by this invention is that the existing technology suffers from the drawback of misdiagnosis caused by the fragmentation of multi-source signals. To address this, we propose a three-phase motor fault diagnosis system and method based on magneto-thermal-oscillatory-electric coupling.

[0005] To achieve the above objectives, this application adopts the following technical solution: a three-phase motor fault diagnosis system based on magnetic-temperature-vibration-electric coupling, including a signal acquisition component installed at the junction box of the three-phase motor and a diagnostic processing component connected to the signal acquisition component;

[0006] The signal acquisition component includes a system housing, a current transformer, a voltage sampling terminal, a temperature and vibration MEMS chip, and a magnetic field MEMS chip. The system housing is fixed to the bottom of the junction box. The current transformer is sleeved on the power supply line of the three-phase terminal block. The voltage sampling terminal is electrically connected to the three-phase terminal block. The temperature and vibration MEMS chip and the magnetic field MEMS chip are located on the side of the system housing close to the motor body, so as to acquire three-phase current signal, three-phase voltage signal, temperature signal, vibration signal, and spatial magnetic field signal at the same installation position.

[0007] The diagnostic processing component includes a microprocessor. The microprocessor performs time alignment on three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals within the same sampling time window. It extracts the fundamental frequency of the magnetic field from the spatial magnetic field signal and determines the real-time rotational frequency of the motor by combining it with the three-phase motor order. The microprocessor performs order processing on the vibration signal based on the real-time rotational frequency of the motor to obtain vibration order characteristics and extracts electrical parameter characteristics and temperature rise trend characteristics.

[0008] The microprocessor generates magnetoelectric coupling characteristics based on the fundamental frequency of the magnetic field and electrical parameter characteristics, generates magnetic vibration coupling characteristics based on the real-time rotation frequency and vibration order characteristics of the motor, generates thermal correlation coupling characteristics based on the temperature rise trend characteristics, electrical parameter characteristics, and vibration order characteristics, and identifies the operating condition and fault type of the three-phase motor based on the coupling characteristics.

[0009] Preferably, the microprocessor performs timestamp matching on the three-phase current signal, three-phase voltage signal, temperature signal, vibration signal and spatial magnetic field signal according to a preset sampling time window to generate a magnetic-temperature-vibration-electric synchronization data segment.

[0010] Preferably, the microprocessor filters and performs spectrum analysis on the spatial magnetic field signal, extracts the magnetic field fundamental frequency, and calculates the real-time motor rotation frequency based on the magnetic field fundamental frequency and the pre-recorded three-phase motor stage.

[0011] Preferably, the microprocessor calculates the ratio of the peak frequency in the vibration signal to the real-time rotational frequency of the motor to obtain vibration order characteristics. These vibration order characteristics are used to identify at least one of dynamic imbalance, misalignment, angular misalignment, and bearing failure.

[0012] Preferably, when the three-phase voltage signal is within the normal range and the three-phase current signal increases, the microprocessor determines whether the three-phase motor is in a starting state, an overload state, or a fault state based on the change in the fundamental frequency of the magnetic field, the change in the vibration order characteristics, and the temperature rise trend characteristics.

[0013] Preferably, the microprocessor generates a fault confidence score for different fault types, and the fault confidence score is obtained by weighting the magnetoelectric coupling characteristics, magnetic vibration coupling characteristics, and thermal correlation coupling characteristics.

[0014] Preferably, when a single signal does not reach the alarm threshold, the microprocessor performs trend accumulation of the fault confidence within multiple consecutive sampling time windows, and outputs an early fault warning signal when the corresponding fault confidence continues to rise.

[0015] The method for fault diagnosis of three-phase motors based on magneto-thermal-oscillatory coupling acquires the three-phase current signal, three-phase voltage signal, temperature signal, vibration signal and spatial magnetic field signal of the three-phase motor within the same sampling time window, and performs time alignment on each signal.

[0016] Extract the fundamental frequency of the magnetic field from the space magnetic field signal, and determine the real-time rotational frequency of the motor based on the fundamental frequency of the magnetic field and the number of stages of the three-phase motor;

[0017] The vibration signal is processed by ordering based on the real-time rotation frequency of the motor to obtain vibration order characteristics. Electrical parameter characteristics are extracted from the three-phase current signal and the three-phase voltage signal, and temperature rise trend characteristics are extracted from the temperature signal.

[0018] Magnetoelectric coupling characteristics are generated based on the fundamental frequency of the magnetic field and the characteristics of electrical parameters; magnetic vibration coupling characteristics are generated based on the real-time rotation frequency of the motor and the vibration order characteristics; and thermal correlation coupling characteristics are generated based on the temperature rise trend characteristics, electrical parameter characteristics, and vibration order characteristics.

[0019] Based on the characteristics of magnetoelectric coupling, magnetic vibration coupling, and thermal correlation coupling, the operating conditions and fault types of a three-phase motor are determined.

[0020] Preferably, when the three-phase voltage signal is within the normal range and the three-phase current signal increases, the three-phase motor is judged to be in a starting state, an overload state, or a fault state based on the changes in the fundamental frequency of the magnetic field, the changes in the vibration order characteristics, and the temperature rise trend characteristics.

[0021] Preferably, when a single signal does not reach the corresponding alarm threshold, the trend of magnetoelectric coupling characteristics, magnetic vibration coupling characteristics and thermal correlation coupling characteristics within multiple consecutive sampling time windows is judged, and an early fault warning signal is output when at least two types of coupling characteristics continue to increase.

[0022] The technical effects and advantages of this invention are as follows:

[0023] This invention synchronously acquires three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals at the junction box of a three-phase motor. Based on the fundamental frequency of the magnetic field, the real-time motor rotation frequency is determined. This real-time rotation frequency is then used to perform order-wise processing on the vibration signals. Simultaneously, combined with electrical parameter characteristics and temperature rise trend characteristics, magneto-electric coupling characteristics, magnetic-vibration coupling characteristics, and thermal correlation coupling characteristics are generated. This allows various detection signals to form mutually corroborating diagnostic evidence under the same operating conditions. Therefore, under complex operating conditions such as motor startup, variable frequency operation, and load fluctuations, the system can first identify the operating condition and then determine the fault type, reducing false alarms and missed alarms caused by single signal changes or simple signal superposition, and improving the accuracy, stability, and early warning capability of three-phase motor fault diagnosis. Attached Figure Description

[0024] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts:

[0025] Figure 1This is a system flowchart of the present invention;

[0026] Figure 2 This is a flowchart of the three-phase motor fault diagnosis method of the present invention;

[0027] Figure 3 This is a schematic diagram illustrating the cumulative trend of early fault warning in this invention. Detailed Implementation

[0028] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0029] Reference Figure 1-3 As shown, this invention provides a technical solution: a three-phase motor fault diagnosis system based on magneto-thermal-oscillatory-electric coupling. This system uses a signal acquisition component installed at the junction box of the three-phase motor to simultaneously acquire three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals. A diagnostic processing component then performs synchronous processing, feature extraction, coupling analysis, and fault diagnosis on these signals. Compared to diagnostic methods that rely solely on vibration, current, or temperature, this invention focuses more on the correspondence between different physical quantities, enabling the diagnostic results to reflect the actual operating state of the motor.

[0030] The system comprises a signal acquisition component, a diagnostic processing component, a power supply module, and a communication module. The signal acquisition component, installed at the junction box of the three-phase motor, includes a system housing, a current transformer, a voltage sampling terminal, a temperature and vibration MEMS chip, and a magnetic field MEMS chip. The system housing is fixed to the bottom of the junction box. The current transformer is mounted on the power supply line of the three-phase terminals. The voltage sampling terminal is electrically connected to the three-phase terminals. The temperature and vibration MEMS chips and the magnetic field MEMS chip are located on the side of the system housing closest to the motor body. This arrangement allows electrical parameters, temperature, vibration, and magnetic field signals to be acquired in a relatively concentrated location, reducing the complexity of on-site wiring and mitigating the time and space deviations caused by the dispersed installation of different sensors. The temperature and vibration MEMS chip is used to acquire the motor temperature signal and vibration signals, including at least radial and axial vibration components.

[0031] The diagnostic processing component includes a microprocessor. During system operation, the microprocessor receives three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals according to a preset sampling time window. It then aligns the signals at different sampling frequencies based on timestamps, forming a synchronous magnetic-temperature-vibration-electric data segment. Specifically, the three-phase current signals, three-phase voltage signals, vibration signals, and spatial magnetic field signals can be acquired using a higher sampling frequency, while the temperature signal can be acquired using a lower sampling frequency. After time alignment, each type of signal corresponds to the same segment of motor operation, preventing anomalies from different time points from being mixed together for subsequent diagnostic analysis.

[0032] After obtaining the magnetic-temperature-vibration-electric synchronization data segment, the system first processes the spatial magnetic field signal. The diagnostic processing component filters and performs spectral analysis on the spatial magnetic field signal to extract the fundamental frequency of the magnetic field, and calculates the real-time motor rotation frequency by combining it with the pre-recorded three-phase motor stage. This real-time motor rotation frequency serves as the reference parameter for vibration diagnosis, reflecting the current rotational state of the motor. After this processing, even if the motor is in startup, variable frequency operation, or load fluctuation state, the system can still obtain a rotation frequency reference adapted to the current operating conditions, avoiding the deviation caused by directly analyzing the vibration frequency based on a fixed speed.

[0033] Subsequently, the diagnostic processing component performs order-based processing on the vibration signal using the motor's real-time rotational frequency as a reference. Specifically, it performs spectral analysis on the vibration signal to obtain the peak vibration frequency, and calculates the ratio of the peak vibration frequency to the motor's real-time rotational frequency to obtain the vibration order characteristics. The vibration order characteristics include at least one of the following: radial first order, radial second order, axial first order, and bearing fault order. The radial first order can be used to determine dynamic balance abnormalities, the radial second order can be used to determine alignment abnormalities, the axial first order can be used to determine angular misalignment, and the bearing fault order can be used to determine bearing-related faults. Because the above vibration order characteristics are normalized according to the motor's real-time rotational frequency, they still maintain good comparability when the speed changes, and the judgment is closer to the motor's current operating state.

[0034] For electrical parameter signals, the diagnostic processing component extracts at least one of the following from the three-phase current signals and three-phase voltage signals: three-phase current imbalance, negative sequence component, current sideband component, and harmonic distortion characteristics. The three-phase current imbalance and negative sequence component reflect the asymmetrical state of the power supply or windings; the current sideband component reflects the spectral changes that may be caused by rotor bar breakage, end ring abnormalities, or air gap eccentricity; and the harmonic distortion characteristics reflect waveform changes caused by local short circuits in the windings, poor contact, or abnormal power supply. For temperature signals, the diagnostic processing component extracts at least one of the following: temperature rise slope, temperature rise duration, and steady-state temperature deviation, to determine whether the motor heating is continuous and whether this heating change corresponds to abnormal current or vibration.

[0035] Based on the above characteristics, the system further generates magnetoelectric coupling characteristics, magnetic vibration coupling characteristics, and thermal correlation coupling characteristics. Magnetoelectric coupling characteristics are generated based on the correspondence between the fundamental frequency of the magnetic field, the magnetic field spectrum distortion state, and electrical parameter characteristics. This is used to determine whether electromagnetic anomalies are accompanied by current imbalance, sideband enhancement, or harmonic distortion. Magnetic vibration coupling characteristics are generated based on the correspondence between the real-time motor rotation frequency and vibration order characteristics. This is used to determine whether the vibration peak value corresponds to the rotation frequency calculated from the magnetic field. Thermal correlation coupling characteristics are generated based on the changing relationship between temperature rise trend characteristics and electrical parameter characteristics and vibration order characteristics. This is used to determine whether the temperature rise mainly comes from overload heating, abnormal heat dissipation, or heat accumulation caused by mechanical friction or bearing abnormalities.

[0036] To reduce false alarms under complex operating conditions, the system first determines the operating status before outputting the fault type. When the three-phase voltage signal is within the normal range and the three-phase current signal increases, the diagnostic processing component combines changes in the magnetic field fundamental frequency, vibration order characteristics, and temperature rise trends to determine whether the three-phase motor is in a starting state, overload state, variable frequency operation state, or fault state. If the magnetic field fundamental frequency changes over time, the vibration order characteristics do not show a stable increase in fault order, and the temperature change conforms to the change pattern during starting or speed regulation, then it is determined to be in a starting state or variable frequency operation state, and no fault alarm is output temporarily. If the magnetic field fundamental frequency is relatively stable, the three-phase current continues to increase, the temperature rise slope increases, but the vibration order characteristics do not show a significant fault order, then it is determined to be in an overload state. If there is a continuous correlation between abnormal current, abnormal magnetic field, abnormal vibration order, and abnormal temperature rise, then it is determined to be in a fault state.

[0037] Under fault conditions, the diagnostic processing component generates fault confidence levels for different fault types and outputs the fault type and alarm level based on the fault confidence levels. The fault confidence level for rotor bar breakage is determined at least based on the current sideband component and magnetic field spectrum distortion state; the fault confidence level for winding inter-turn short circuit is determined at least based on the three-phase current imbalance, harmonic distortion characteristics, and temperature rise slope; the fault confidence level for air gap eccentricity is determined at least based on current spectrum anomalies, magnetic field spectrum anomalies, and vibration order changes; the fault confidence levels for dynamic balance anomalies, alignment anomalies, and angular misalignment are determined at least based on the corresponding radial first order, radial second order, and axial first order, respectively; the fault confidence level for bearing faults is determined at least based on the bearing fault order and temperature rise duration; and the fault confidence level for heat dissipation anomalies is determined at least based on the inconsistency between temperature rise trend characteristics and electrical parameter characteristics.

[0038] Furthermore, when a single signal has not yet reached the alarm threshold, the diagnostic processing component accumulates the fault confidence over multiple consecutive sampling time windows. If the fault confidence corresponding to the same fault type continues to increase, and at least two of the following—current harmonic distortion characteristics, magnetic field spectrum distortion state, vibration order amplitude, and temperature rise slope—show a unidirectional increase over multiple consecutive sampling time windows, an early fault warning signal is output. This processing method can detect risk changes in advance when fault characteristics are not yet obvious, and is suitable for the early identification of gradual faults such as those in rotors, windings, and bearings.

[0039] This invention also provides a method for diagnosing three-phase motor faults using the above system, comprising: acquiring three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals at a junction box; performing time alignment on the above signals to form a magnetic-temperature-vibration-electric synchronous data segment; extracting the fundamental frequency of the magnetic field from the spatial magnetic field signal, and determining the real-time rotational frequency of the motor based on the fundamental frequency of the magnetic field and the number of stages of the three-phase motor; performing order-based processing on the vibration signal based on the real-time rotational frequency of the motor to obtain vibration order characteristics; extracting electrical parameter characteristics and temperature rise trend characteristics; generating magnetoelectric coupling characteristics, magnetic vibration coupling characteristics, and thermal correlation coupling characteristics based on the fundamental frequency of the magnetic field, the real-time rotational frequency of the motor, the vibration order characteristics, the electrical parameter characteristics, and the temperature rise trend characteristics; determining the operating condition based on the above coupling characteristics, and outputting the fault type, alarm level, or early fault warning signal under the constraints of the operating condition.

[0040] Example 1:

[0041] This embodiment is applied to a three-phase motor used to drive a circulating water pump in an industrial pump room. During system installation, the signal acquisition component is fixed to the bottom of the junction box of the three-phase motor, the current transformer is sleeved on the power supply line of the three-phase terminals, the voltage sampling terminal is electrically connected to the three-phase terminals, and the temperature vibration MEMS chip and magnetic field MEMS chip are located on the side of the system casing near the motor body. After installation, the number of poles, rated voltage, rated current, bearing fault characteristic parameters, and corresponding judgment thresholds of the three-phase motor are entered into the diagnostic processing component.

[0042] During system operation, the diagnostic processing component synchronously acquires three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals according to a preset sampling time window. It then aligns these signals based on timestamps to form a synchronous magnetic-temperature-vibration-electric data segment. For current, voltage, vibration, and magnetic field signals with rapidly changing waveforms, a higher sampling frequency is used; for temperature signals, a lower sampling frequency is used. This processing ensures that each diagnostic corresponds to the same segment of motor operation, preventing the mixing and judgment of data from different time periods.

[0043] During the stable operation phase of the motor, the three-phase voltage is within the normal range, the three-phase current does not exceed the rated range, and the temperature is within the normal operating range. The diagnostic processing component filters and performs spectrum analysis on the spatial magnetic field signal to extract the fundamental frequency of the magnetic field, and determines the real-time rotational frequency of the motor based on the fundamental frequency of the magnetic field and the number of three-phase motor stages. Subsequently, the diagnostic processing component performs spectrum analysis on the vibration signal, calculates the ratio of the vibration peak frequency to the real-time rotational frequency of the motor, and obtains the vibration order characteristics.

[0044] During operation, the system detected a gradual increase in the vibration order amplitude corresponding to the bearing fault characteristic parameters, but the three-phase current imbalance, current harmonic distortion, and three-phase voltage did not reach the electrical abnormality alarm conditions. Simultaneously, the fundamental frequency of the spatial magnetic field remained relatively stable, indicating that the motor speed did not fluctuate significantly, while the temperature signal showed a gradual upward trend with the increase in vibration order amplitude. Therefore, the system determined that the abnormality was mainly manifested as increased mechanical-side vibration, accompanied by a certain amount of heat accumulation.

[0045] The diagnostic processing component further generates magnetic vibration coupling features and thermal correlation coupling features. Magnetic vibration coupling features are used to characterize the correspondence between vibration order characteristics and the real-time motor speed obtained by inverse calculation of the magnetic field; thermal correlation coupling features are used to characterize the relationship between the temperature rise trend and the increase in vibration abnormality. Since the electrical parameter characteristics do not show obvious abnormalities, the system reduces the diagnostic weight of electrical faults and increases the confidence level of bearing fault correlation.

[0046] When the bearing fault-related confidence level continuously increases within multiple consecutive sampling time windows, and the vibration order amplitude and temperature rise trend increase in the same direction, the system outputs an early bearing fault warning signal. At this time, since the three-phase current and three-phase voltage do not show overload or power supply abnormality characteristics, and the fundamental frequency of the spatial magnetic field does not show significant changes during startup or frequency conversion, the system does not classify this state as a startup state, overload state, or frequency conversion operation state, but rather identifies it as an early bearing damage risk.

[0047] During another operating period, the circulating water pump load increased. The system detected that the three-phase voltage remained within the normal range, the three-phase current continued to rise, and the temperature rise trend intensified. However, the fundamental frequency of the spatial magnetic field remained relatively stable, and the vibration order characteristics did not show a stable increase in fault order. Based on the combination of increased current, enhanced temperature rise, stable magnetic field fundamental frequency, and no obvious fault characteristics in the vibration, the diagnostic processing component judged the current operating condition as an overload state and output an overload alarm, but did not output a bearing fault alarm.

[0048] During the motor startup phase, the system detects a short-term increase in the three-phase current, a gradual change in the fundamental frequency of the spatial magnetic field from low to high, and a corresponding increase in the motor's real-time rotational frequency. The peak vibration frequency also shifts synchronously with the real-time rotational frequency, while the temperature exhibits only normal changes during startup. Based on this, the diagnostic processing component determines that the motor is in a startup state and suppresses fault alarms. This process avoids misinterpreting the startup inrush current as an overload or electrical fault.

[0049] As can be seen from this embodiment, in practical applications, the real-time motor speed is first determined by the spatial magnetic field signal. Then, the vibration signal is processed by order reduction based on the real-time motor speed, and the operating condition constraints are judged by combining electrical parameter characteristics and temperature rise trend characteristics. For early bearing failures, the system can output early warnings based on the continuous correlation between vibration order characteristics and temperature rise trends when the current and voltage are not yet obviously abnormal. For overload and startup processes, the system can distinguish based on the differences between the magnetic field fundamental frequency, vibration order, and temperature rise trend, thereby improving the diagnostic accuracy under complex operating conditions. The content regarding junction box installation, magnetic field frequency calculation of speed, and magnetic vibration coupling alarm in the original disclosure can serve as the basis for this embodiment.

[0050] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A three-phase motor fault diagnosis system based on magneto-thermal-oscillatory-electric coupling, characterized in that, This includes a signal acquisition component installed at the junction box of a three-phase motor and a diagnostic processing component connected to the signal acquisition component; The signal acquisition component includes a system housing, a current transformer, a voltage sampling terminal, a temperature and vibration MEMS chip, and a magnetic field MEMS chip. The system housing is fixed to the bottom of the junction box. The current transformer is sleeved on the power supply line of the three-phase terminal block. The voltage sampling terminal is electrically connected to the three-phase terminal block. The temperature and vibration MEMS chip and the magnetic field MEMS chip are located on the side of the system housing close to the motor body, so as to acquire three-phase current signal, three-phase voltage signal, temperature signal, vibration signal, and spatial magnetic field signal at the same installation position. The diagnostic processing component includes a microprocessor. The microprocessor performs time alignment on three-phase current signals, three-phase voltage signals, temperature signals, vibration signals, and spatial magnetic field signals within the same sampling time window. It extracts the fundamental frequency of the magnetic field from the spatial magnetic field signal and determines the real-time rotational frequency of the motor by combining it with the three-phase motor order. The microprocessor performs order processing on the vibration signal based on the real-time rotational frequency of the motor to obtain vibration order characteristics and extracts electrical parameter characteristics and temperature rise trend characteristics. The microprocessor generates magnetoelectric coupling characteristics based on the fundamental frequency of the magnetic field and electrical parameter characteristics, generates magnetic vibration coupling characteristics based on the real-time rotation frequency and vibration order characteristics of the motor, generates thermal correlation coupling characteristics based on the temperature rise trend characteristics, electrical parameter characteristics, and vibration order characteristics, and identifies the operating condition and fault type of the three-phase motor based on the coupling characteristics.

2. The three-phase motor fault diagnosis system based on magneto-thermal-oscillatory coupling according to claim 1, characterized in that: The microprocessor timestamps the three-phase current signal, three-phase voltage signal, temperature signal, vibration signal, and spatial magnetic field signal according to a preset sampling time window, and generates a magnetic-temperature-vibration-electric synchronization data segment.

3. The three-phase motor fault diagnosis system based on magneto-thermal-electric coupling according to claim 1, characterized in that: The microprocessor filters and performs spectrum analysis on the spatial magnetic field signal, extracts the fundamental frequency of the magnetic field, and calculates the real-time rotational frequency of the motor based on the fundamental frequency of the magnetic field and the pre-recorded three-phase motor stage.

4. The three-phase motor fault diagnosis system based on magneto-thermal-electric coupling according to claim 1, characterized in that: The microprocessor calculates the ratio of the peak frequency in the vibration signal to the real-time rotation frequency of the motor to obtain the vibration order characteristics. The vibration order characteristics are used to identify at least one of dynamic balance abnormality, alignment abnormality, angle misalignment, and bearing failure.

5. The three-phase motor fault diagnosis system based on magneto-thermal-electric coupling according to claim 1, characterized in that: When the three-phase voltage signal is within the normal range and the three-phase current signal increases, the microprocessor determines whether the three-phase motor is in a starting state, an overload state, or a fault state based on the changes in the magnetic field fundamental frequency, the changes in vibration order characteristics, and the temperature rise trend characteristics.

6. The three-phase motor fault diagnosis system based on magneto-thermal-electric coupling according to claim 5, characterized in that: The microprocessor generates fault confidence scores for different fault types, and the fault confidence scores are obtained by weighting the magnetoelectric coupling characteristics, magnetic vibration coupling characteristics, and thermal correlation coupling characteristics.

7. The three-phase motor fault diagnosis system based on magneto-thermal-electric coupling according to claim 6, characterized in that: When a single signal does not reach the alarm threshold, the microprocessor performs trend accumulation of the fault confidence within multiple consecutive sampling time windows, and outputs an early fault warning signal when the corresponding fault confidence continues to rise.

8. A method for diagnosing three-phase motor faults using the three-phase motor fault diagnosis system based on magneto-thermal-oscillatory coupling as described in any one of claims 1 to 7, characterized in that: The three-phase current signal, three-phase voltage signal, temperature signal, vibration signal and spatial magnetic field signal of the three-phase motor are acquired within the same sampling time window, and the time of each signal is aligned. Extract the fundamental frequency of the magnetic field from the space magnetic field signal, and determine the real-time rotational frequency of the motor based on the fundamental frequency of the magnetic field and the number of stages of the three-phase motor; The vibration signal is processed by ordering based on the real-time rotation frequency of the motor to obtain vibration order characteristics. Electrical parameter characteristics are extracted from the three-phase current signal and the three-phase voltage signal, and temperature rise trend characteristics are extracted from the temperature signal. Magnetoelectric coupling characteristics are generated based on the fundamental frequency of the magnetic field and the characteristics of electrical parameters; magnetic vibration coupling characteristics are generated based on the real-time rotation frequency of the motor and the vibration order characteristics; and thermal correlation coupling characteristics are generated based on the temperature rise trend characteristics, electrical parameter characteristics, and vibration order characteristics. Based on the characteristics of magnetoelectric coupling, magnetic vibration coupling, and thermal correlation coupling, the operating conditions and fault types of a three-phase motor are determined.

9. The three-phase motor fault diagnosis method based on magneto-thermal-electric coupling according to claim 8, characterized in that: When the three-phase voltage signal is within the normal range and the three-phase current signal increases, the three-phase motor is judged to be in the starting state, overload state or fault state based on the changes in the fundamental frequency of the magnetic field, the changes in the vibration order characteristics and the temperature rise trend characteristics.

10. The three-phase motor fault diagnosis method based on magneto-thermal-oscillatory coupling according to claim 8, characterized in that: When a single signal does not reach the corresponding alarm threshold, the trend of magnetoelectric coupling characteristics, magnetic vibration coupling characteristics and thermal correlation coupling characteristics within multiple consecutive sampling time windows is judged, and an early fault warning signal is output when at least two types of coupling characteristics continue to increase.

Images