A method and system for motor fault diagnosis based on the linkage of acoustic signature features and frequency conversion parameters

Abstract

This invention discloses a method and system for motor fault diagnosis based on the linkage of acoustic signature features and frequency converter parameters, belonging to the field of mechanical vibration and acoustic measurement technology. It includes: acquiring the operating frequency range of the motor to divide it into characteristic frequency bands; collecting historical acoustic signature signals of the motor operation and extracting features to generate a dynamic baseline threshold library; selecting the dynamic baseline threshold corresponding to the characteristic frequency band for acoustic signature feature comparison; outputting a preliminary fault diagnosis result; controlling the frequency converter to perform micro-amplitude frequency sweep within a preset frequency amplitude range and collecting the swept acoustic signature signal; determining whether the fault persists; if so, decoupling the energy of the swept acoustic signature signal, calculating the true fault intensity, and generating a frequency-fault intensity mapping map and fault classification results. This invention, by establishing a dynamic baseline threshold library, performing micro-amplitude frequency sweep to verify fault persistence, decoupling frequency converter electromagnetic interference, and finally generating classification results, can achieve high-precision, interference-resistant quantitative diagnosis of motor faults.

Description

Technical Field

[0001] This invention relates to the fields of mechanical vibration and acoustic measurement technology, and in particular to a method and system for diagnosing motor faults based on the linkage of acoustic signature characteristics and frequency conversion parameters. Background Technology

[0002] In the field of mechanical vibration measurement technology, motor fault diagnosis is often achieved by analyzing acoustic signature signals during operation. These signals contain sound wave information generated by the vibration of the motor's mechanical structure, and their frequency and time domain characteristics can reflect fault states such as bearing damage and rotor imbalance. Especially for variable frequency motors, the dynamic changes in operating frequency result in a strong correlation between acoustic signature characteristics and frequency conversion parameters.

[0003] Existing technologies typically employ a fixed threshold method for fault diagnosis. This method involves acquiring acoustic signature signals in a specific frequency band using a single sensor, extracting the energy amplitude or harmonic components as characteristic quantities, and comparing them with a preset fixed threshold. It relies on vibration sensor data to build a static model, ignoring the interference of electromagnetic noise during frequency conversion on the acoustic signature signal. While some solutions introduce frequency segmentation, they fail to dynamically divide the frequency bands in conjunction with the motor's resonance characteristics.

[0004] However, fixed thresholds cannot adapt to fluctuations in motor load and changes in operating frequency, leading to an increased false alarm rate; the pollution of acoustic signature features by electromagnetic harmonics of the frequency converter is not considered, making it easy to misjudge electromagnetic interference as mechanical faults; the lack of a verification mechanism for the persistence of faults makes it difficult to distinguish between transient anomalies and real damage; in addition, existing methods cannot quantify fault intensity and accurately locate characteristic frequencies, which limits the accuracy of fault classification. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a motor fault diagnosis method and system based on the linkage of acoustic signature features and frequency conversion parameters. By establishing a dynamic baseline threshold library, performing micro-amplitude frequency sweep to verify fault persistence, and decoupling frequency conversion electromagnetic interference, a frequency-fault intensity mapping map and classification results are finally generated, enabling high-precision and interference-resistant quantitative diagnosis of motor faults.

[0006] The above objectives can be achieved through the following approach:

[0007] A motor fault diagnosis method based on the linkage of acoustic signature features and frequency converter parameters includes: acquiring the operating frequency range of the motor; dividing the operating frequency range into multiple characteristic frequency bands according to the motor's resonance point; collecting historical acoustic signature signals of the motor operation within each characteristic frequency band and extracting features to generate a dynamic baseline threshold library; monitoring the actual operating frequency and actual acoustic signature signals of the motor in real time; selecting the corresponding dynamic baseline threshold in the dynamic baseline threshold library for acoustic signature feature comparison based on the characteristic frequency band where the actual operating frequency is located, and outputting a primary fault diagnosis result; based on the primary fault diagnosis result, controlling the frequency converter to perform micro-amplitude frequency sweep within a preset frequency amplitude range, collecting the frequency sweep acoustic signature signals during the frequency sweep process, and determining whether the fault persists based on the frequency sweep acoustic signature signals; if the fault persists, decoupling the energy of the frequency sweep acoustic signature signals and calculating the true fault intensity; and generating a frequency-fault intensity mapping map and fault classification results based on the true fault intensity.

[0008] Optionally, obtaining the operating frequency range of the motor and dividing the operating frequency range into multiple characteristic frequency bands based on the motor's resonance point includes: obtaining historical vibration spectrum data of the motor and identifying the point of sudden increase in vibration energy as the resonance point; dividing the operating frequency range into high-frequency band, mid-frequency band, and low-frequency band using adjacent resonance points as boundaries.

[0009] Optionally, the step of collecting historical acoustic signature signals of motor operation within each characteristic frequency band and extracting features to generate a dynamic baseline threshold library includes: collecting historical acoustic signature signals of motor operation within each characteristic frequency band and extracting the preset fault band energy, full-band energy, acoustic signal amplitude, fundamental acoustic energy, and electromagnetic harmonic acoustic energy of the corresponding characteristic frequency band; calculating the ratio of the preset fault band energy to the full-band energy as the frequency domain energy focusing value, and calculating the reference energy focusing value based on the frequency domain energy focusing value; counting the number of times the acoustic signal amplitude exceeds a preset impact threshold per unit time as the time domain pulse density, and calculating the reference pulse density based on the time domain pulse density; measuring the ratio of the fundamental acoustic energy to the electromagnetic harmonic acoustic energy as the harmonic distortion rate, and calculating the reference harmonic distortion rate based on the harmonic distortion rate; and integrating the reference energy focusing value, reference pulse density, and reference harmonic distortion rate of each characteristic frequency band to obtain the dynamic baseline threshold library.

[0010] Optionally, the output of the primary fault diagnosis result includes: real-time monitoring of the actual operating frequency and actual acoustic signature signal of the motor; based on the actual operating frequency, selecting the reference energy focusing value, reference pulse density, and reference harmonic distortion rate of the corresponding characteristic frequency band from the dynamic baseline threshold library to obtain a reference threshold set; performing feature extraction on the actual acoustic signature signal to obtain the actual energy focusing value, actual pulse density, and actual harmonic distortion rate, and an actual feature set; calculating the deviation between the actual feature set and the reference threshold set, and generating the primary fault diagnosis result based on the deviation.

[0011] Optionally, calculating the deviation between the actual feature set and the reference threshold set, and generating a preliminary fault diagnosis result based on the deviation, includes: calculating the actual pulse density deviation based on the actual pulse density and the corresponding reference pulse density; calculating the actual energy focusing deviation based on the actual energy focusing value and the corresponding reference energy focusing value; calculating the actual distortion deviation based on the actual harmonic distortion rate and the corresponding reference harmonic distortion rate; determining whether the actual pulse density deviation is greater than a preset pulse density deviation threshold, or the actual energy focusing deviation is greater than a preset energy focusing deviation threshold, or the actual distortion deviation is greater than a preset distortion deviation threshold; if yes, then a fault is determined to exist; if no, then a fault is determined to not exist.

[0012] Optionally, determining whether a fault persists based on the frequency sweep acoustic signal includes: extracting features from the frequency sweep acoustic signal to determine the fault feature intensity at the start of the frequency sweep; obtaining the fault feature intensity at the end of the frequency sweep based on the fault feature intensity at the start of the frequency sweep, and calculating the attenuation rate of the fault feature intensity; determining that the fault persists when the attenuation rate of the fault feature intensity is less than a preset attenuation threshold.

[0013] Optionally, the step of extracting features from the swept acoustic signal to determine the fault feature intensity at the start of the swept frequency includes: extracting features from the swept acoustic signal at the start of the swept frequency to obtain the swept frequency energy focusing value, swept frequency pulse density, and swept frequency harmonic distortion rate, and a swept frequency feature set; calculating the swept frequency pulse density deviation, swept frequency energy focusing deviation, and swept frequency distortion deviation based on the swept frequency feature set and the reference threshold set; sorting the swept frequency pulse density deviation, swept frequency energy focusing deviation, and swept frequency distortion deviation from largest to smallest, and selecting the value ranked first as the fault feature intensity at the start of the swept frequency.

[0014] Optionally, if the fault persists, the energy decoupling of the swept frequency acoustic signal to calculate the true fault intensity includes: if the fault persists, collecting the percentage value of the motor's real-time output power to the rated power during the swept frequency process to obtain the motor load rate; extracting the swept frequency harmonic distortion rate during the swept frequency process based on the swept frequency acoustic signal; calculating the frequency converter interference factor based on the motor load rate and the swept frequency harmonic distortion rate; and calculating the true fault intensity using the frequency converter interference factor and the fault characteristic intensity during the swept frequency process.

[0015] Optionally, generating the frequency-fault intensity mapping map and fault classification results based on the actual fault intensity includes: generating a frequency-fault intensity mapping map using sampling frequency points within a frequency sweep range and the actual fault intensity corresponding to the sampling frequency points; obtaining the actual fault intensity peak value based on the frequency-fault intensity mapping map; outputting a normal classification when the actual fault intensity peak value is less than a preset first threshold; outputting a minor fault classification when the actual fault intensity peak value is greater than or equal to the first threshold and less than a preset second threshold; outputting a severe fault classification when the actual fault intensity peak value is greater than or equal to the second threshold; when outputting either the minor fault classification or the severe fault classification, obtaining the fault feature frequency according to the actual fault intensity peak value and the frequency-fault intensity mapping map; and matching the fault feature frequency with a preset fault feature frequency library to obtain the fault classification result.

[0016] Based on the same inventive concept, this invention also provides a motor fault diagnosis system based on the linkage of acoustic signature features and frequency conversion parameters. The system includes: a frequency band division module for acquiring the operating frequency range of the motor and dividing the operating frequency range into multiple characteristic frequency bands according to the motor's resonance point; a dynamic baseline generation module for collecting historical acoustic signature signals of the motor within each characteristic frequency band and extracting features to generate a dynamic baseline threshold library; a forward diagnosis module for real-time monitoring of the motor's actual operating frequency and actual acoustic signature signals, selecting the corresponding dynamic baseline threshold from the dynamic baseline threshold library based on the characteristic frequency band where the actual operating frequency is located, comparing acoustic signature features, and outputting a primary fault diagnosis result; a reverse diagnosis module for controlling the frequency converter to perform micro-amplitude frequency sweep within a preset frequency amplitude range based on the primary fault diagnosis result, collecting the frequency sweep acoustic signature signal during the frequency sweep process, and determining whether the fault persists based on the frequency sweep acoustic signature signal; a fault decoupling module for performing energy decoupling on the frequency sweep acoustic signature signal if the fault persists, and calculating the true fault intensity; and a fault determination module for generating a frequency-fault intensity mapping diagram and fault classification results based on the true fault intensity.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] 1. This invention significantly improves the robustness and early warning capability of fault diagnosis through the adaptive matching mechanism of dynamic baseline threshold library; based on the motor resonance characteristics, it divides the characteristic frequency bands and establishes a multi-dimensional acoustic signature benchmark library, which can adaptively match the normal operating state characteristics under different working conditions, effectively reducing the risk of misjudgment caused by environmental noise and load fluctuations.

[0019] 2. This invention utilizes micro-amplitude frequency sweep to excite fault response characteristics and combines them with attenuation rate analysis to achieve accurate verification of fault persistence. This method actively controls the frequency of the inverter to change slightly, captures the dynamic attenuation characteristics of the fault characteristic intensity, and effectively distinguishes between real mechanical faults and instantaneous electromagnetic interference or random noise.

[0020] 3. This invention eliminates the influence of inverter electromagnetic interference on acoustic signature characteristics based on the energy decoupling mechanism of motor load rate and harmonic distortion rate; by calculating the inverter interference factor and decoupling the electromagnetic noise component in the swept frequency acoustic signature signal, the true fault intensity that only reflects mechanical damage is extracted, thereby improving the accuracy of fault quantitative assessment.

[0021] 4. This invention achieves collaborative determination of fault type and severity through a frequency-fault intensity mapping graph; by combining the hierarchical comparison of the actual fault intensity peak value and the preset threshold, as well as the matching of fault characteristic frequency with the typical fault database, the fault classification results and severity level are output simultaneously, providing multi-dimensional and accurate basis for maintenance decisions.

[0022] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating the motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of frequency band division according to an embodiment of the present invention.

[0026] Figure 3This is a schematic diagram of the motor fault diagnosis system based on the linkage of acoustic features and frequency conversion parameters according to an embodiment of the present invention.

[0027] Figure 4 This is a frequency-fault intensity mapping diagram according to an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Reference Figure 1 One embodiment of the present invention proposes a motor fault diagnosis method based on the linkage of acoustic features and frequency conversion parameters. By establishing a dynamic baseline threshold library, performing micro-amplitude frequency sweep to verify fault persistence, and decoupling frequency conversion electromagnetic interference, a frequency-fault intensity mapping map and classification results are finally generated, which can achieve high-precision and interference-resistant quantitative diagnosis of motor faults.

[0030] The method described in this embodiment specifically includes:

[0031] Obtain the operating frequency range of the motor, and divide the operating frequency range into multiple characteristic frequency bands based on the motor's resonance point;

[0032] Collect historical acoustic signature signals of motor operation in each characteristic frequency band and extract features to generate a dynamic baseline threshold library;

[0033] The actual operating frequency and actual acoustic signature of the motor are monitored in real time. Based on the characteristic frequency band of the actual operating frequency, the corresponding dynamic baseline threshold in the dynamic baseline threshold library is selected for acoustic signature feature comparison, and the primary fault diagnosis result is output.

[0034] Based on the initial fault diagnosis results, the frequency converter is controlled to perform micro-sweep within a preset frequency amplitude range, and the sweep acoustic signal during the sweep process is collected. Based on the sweep acoustic signal, it is determined whether the fault persists.

[0035] If the fault persists, the swept frequency acoustic signal is decoupled by energy to calculate the true fault intensity.

[0036] Based on the actual fault intensity, a frequency-fault intensity mapping diagram and fault classification results are generated.

[0037] Specifically, a dynamic collaborative mechanism between acoustic signature features and frequency converter parameters is established. First, the operating frequency range is divided based on the motor's resonance characteristics, generating characteristic frequency bands with clear physical meaning. Multi-dimensional features such as frequency domain energy focusing value, time domain pulse density, and harmonic distortion rate are extracted from historical acoustic signature signals to construct an adaptive dynamic baseline threshold library. During real-time monitoring, the acoustic signature feature deviation is calculated by matching the baseline threshold of the corresponding frequency band to the actual operating frequency, achieving primary fault diagnosis. If an anomaly is detected, the frequency converter is triggered to perform a micro-amplitude frequency sweep within a preset range. The persistence of the fault is confirmed by analyzing the attenuation rate of the fault feature intensity of the swept acoustic signature signal. For persistent faults, the frequency converter interference factor is calculated by combining the motor load rate and the swept harmonic distortion rate to decouple the influence of electromagnetic noise on the acoustic signature features. Finally, a frequency-fault intensity mapping map and classification results are generated based on the actual fault intensity. This method significantly improves the reliability and accuracy of fault diagnosis. Adaptive matching using a dynamic baseline threshold library enhances the sensitivity to detect early, subtle faults. Micro-amplitude frequency sweeping excites fault response characteristics and combines this with an energy decoupling mechanism to effectively distinguish between real mechanical faults and transient electromagnetic interference. A quantitative assessment of fault type and severity is achieved based on frequency-fault intensity mapping, providing a clear basis for maintenance decisions. Simultaneously, it reduces the misjudgment rate caused by environmental noise and operating condition fluctuations, optimizing the robustness of the diagnostic process.

[0038] Optionally, obtaining the operating frequency range of the motor, and dividing the operating frequency range into multiple characteristic frequency bands based on the motor's resonance point, includes:

[0039] Acquire historical vibration spectrum data of the motor and identify points of sudden increase in vibration energy as resonance points;

[0040] Using adjacent resonance points as boundaries, the operating frequency range is divided into high-frequency, mid-frequency, and low-frequency bands.

[0041] Specifically, historical vibration spectrum data of the motor is acquired through vibration sensors installed on the motor housing, with a sampling frequency no less than twice the motor's highest operating frequency. The method for identifying vibration energy spikes is as follows: peak detection is performed on the vibration spectrum data. When the difference between the vibration energy amplitude at a certain frequency point and the energy amplitudes at its left and right adjacent frequency points both exceed a preset energy spike threshold, that point is determined to be a vibration energy spike point, i.e., a resonance point. Using the frequency range between all adjacent resonance points as boundaries, the motor's operating frequency range is divided into continuous characteristic frequency bands, specifically: a high-frequency band (from the highest resonance point to the motor's maximum allowable frequency), a mid-frequency band (the main operating frequency band between two adjacent resonance points), and a low-frequency band (from the motor's lowest allowable frequency to the lowest resonance point). For example, as... Figure 2As shown, when resonance points f1=25Hz, f2=50Hz, and f3=75Hz are identified, the low-frequency band is divided into 0-25Hz, the mid-frequency band into 25-50Hz and 50-75Hz, and the high-frequency band into 75-100Hz. Resonance point identification accurately segments regions with varying frequency response characteristics of the motor, improving the clarity of the physical meaning of the characteristic frequency band division. This makes subsequent acoustic signature comparisons more closely aligned with the actual dynamic behavior of the motor, avoiding misjudgments caused by unreasonable frequency band division.

[0042] Optionally, the step of collecting historical acoustic signature signals of motor operation within each characteristic frequency band and extracting features to generate a dynamic baseline threshold library includes:

[0043] Collect historical acoustic signature signals of motor operation within each characteristic frequency band, and extract the preset fault band energy, full band energy, acoustic signal amplitude, fundamental acoustic energy, and electromagnetic harmonic acoustic energy of the corresponding characteristic frequency band.

[0044] The ratio of the preset fault frequency band energy to the full frequency band energy is calculated as the frequency domain energy focusing value, and a reference energy focusing value is calculated based on the frequency domain energy focusing value.

[0045] The number of times the amplitude of the acoustic signal exceeds a preset impact threshold per unit time is counted as the time-domain pulse density, and a reference pulse density is calculated based on the time-domain pulse density.

[0046] The ratio of fundamental acoustic energy to electromagnetic harmonic acoustic energy is measured as the harmonic distortion rate, and a reference harmonic distortion rate is calculated based on the harmonic distortion rate.

[0047] By integrating the reference energy focusing value, reference pulse density, and reference harmonic distortion rate of each characteristic frequency band, a dynamic baseline threshold library is obtained.

[0048] Specifically, for each characteristic frequency band, multiple historical operating cycle acoustic signature signal samples are collected, with each sample being time-series data. First, a Fast Fourier Transform (FFT) is performed on the historical acoustic signature signals to obtain a frequency domain representation. Energy in a preset fault frequency band is extracted, obtained by integrating the sum of squares of signal energy amplitudes within that band (e.g., the common frequency band for motor bearing faults, 2000-5000Hz). Energy across the entire frequency band is extracted by integrating the sum of squares of signal energy amplitudes across the entire frequency range (e.g., 0-10000Hz). The amplitude of the acoustic wave signal is extracted by directly reading the peak or root-mean-square value from the time domain signal. Fundamental acoustic energy is extracted by using FFT to extract the energy amplitude at the motor's current operating fundamental frequency. Electromagnetic harmonic acoustic energy is extracted by using FFT to extract the sum of harmonic component energy amplitudes at integer multiples of the fundamental frequency. The frequency domain energy focusing value is then calculated. :

[0049] ,

[0050] in To preset the fault frequency band energy, Both the full-band energy and the energy value are obtained through FFT energy integration, with consistent dimensions in energy units, and the ratio is dimensionless. A baseline energy focusing value is calculated based on the frequency domain energy focusing value, obtained by statistically averaging the frequency domain energy focusing values ​​of multiple historical samples. The time-domain pulse density is statistically analyzed by setting a unit time (e.g., 1 second) and a preset impact threshold (based on historical signal amplitude statistics), counting the number of times the acoustic signal amplitude exceeds the preset impact threshold within a unit time. :

[0051] ,

[0052] in For the number of times exceeding the threshold, The unit of time is measured in terms of cycles / time. A baseline pulse density is calculated based on the time-domain pulse density, obtained by statistically averaging the time-domain pulse densities of multiple historical samples. Harmonic distortion rate is then measured. :

[0053] ,

[0054] in Electromagnetic harmonic acoustic energy, Both fundamental acoustic energy and harmonic distortion (HID) are obtained through FFT energy extraction, with consistent dimensions in energy units and dimensionless ratios. A baseline HID is calculated based on the harmonic distortion rate, obtained by statistically averaging the HIDs of multiple historical samples. A dynamic baseline threshold library is formed by integrating the baseline energy focus value, baseline pulse density, and baseline HID of each characteristic frequency band. This library stores three baseline value datasets corresponding to each characteristic frequency band. Baseline thresholds are established through feature statistical analysis of historical acoustic signals to capture the frequency domain energy distribution, time domain pulse characteristics, and harmonic component ratios under normal motor operation, thus providing a reliable reference for real-time fault diagnosis. The robustness and adaptability of fault feature extraction are enhanced, reducing the risk of misjudgment due to environmental noise interference, improving sensitivity to early, subtle faults, and optimizing diagnostic accuracy through multi-dimensional feature integration.

[0055] Optionally, the output of the primary fault diagnosis result includes:

[0056] Real-time monitoring of the motor's actual operating frequency and actual acoustic signature signal;

[0057] Based on the actual operating frequency, the reference energy focusing value, reference pulse density and reference harmonic distortion rate of the corresponding characteristic frequency band are selected from the dynamic baseline threshold library to obtain the reference threshold set;

[0058] Feature extraction is performed on the actual voiceprint signal to obtain the actual energy focusing value, actual pulse density, actual harmonic distortion rate, and actual feature set;

[0059] Calculate the deviation between the actual feature set and the benchmark threshold set, and generate a preliminary fault diagnosis result based on the deviation.

[0060] Optionally, calculating the deviation between the actual feature set and the benchmark threshold set, and generating a preliminary fault diagnosis result based on the deviation, includes:

[0061] The deviation of the actual pulse density is calculated based on the actual pulse density and the corresponding reference pulse density.

[0062] The actual energy focusing deviation is calculated based on the actual energy focusing value and the corresponding reference energy focusing value.

[0063] The actual distortion deviation is calculated based on the actual harmonic distortion rate and the corresponding reference harmonic distortion rate.

[0064] Determine whether the actual pulse density deviation is greater than a preset pulse density deviation threshold, or the actual energy focusing deviation is greater than a preset energy focusing deviation threshold, or the actual distortion deviation is greater than a preset distortion deviation threshold.

[0065] If so, then a fault is determined to exist;

[0066] If not, then it is determined that there is no fault.

[0067] Specifically, based on the aforementioned method, the deviation of the actual pulse density is calculated by comparing the actual pulse density obtained from the feature extraction of the actual voiceprint signal with the reference pulse density of the corresponding feature frequency band in the dynamic baseline threshold library. :

[0068] ,

[0069] in This refers to the number of times the amplitude of the acoustic signal exceeds the preset impact threshold per unit time, measured in real time. The unit is the number of times per time. The baseline pulse density is defined as historical statistics, with the unit being pulses per time. The actual energy focus deviation is calculated based on the actual energy focus value extracted from the actual voiceprint signal features and the baseline energy focus value in the baseline threshold set. :

[0070] ,

[0071] in The ratio of the preset fault band energy to the full band energy, calculated in real time, is dimensionless. The historical baseline value is dimensionless. The actual distortion deviation is calculated based on the actual harmonic distortion rate extracted from the actual voiceprint signal features and the baseline harmonic distortion rate in the baseline threshold set. :

[0072] ,

[0073] in The ratio of fundamental acoustic energy to electromagnetic harmonic acoustic energy is measured in real time and is dimensionless. The values ​​are historical baselines and are dimensionless. The system checks whether the actual pulse density deviation, actual energy focusing deviation, and actual distortion deviation exceed preset thresholds. These thresholds are determined based on historical operating data of the motor model. If any deviation exceeds its corresponding threshold, a fault is identified; if none exceed their respective thresholds, no fault is identified. The initial fault diagnosis result is output as a Boolean value. Through multi-dimensional comparison of real-time acoustic signature features with historical baseline thresholds, feature deviations are detected to identify anomalies. Complementary frequency and time domain features enhance diagnostic reliability. This significantly improves the accuracy and robustness of fault diagnosis, reduces false alarms caused by environmental noise interference, and increases the sensitivity to detect subtle early-stage faults.

[0074] Optionally, determining whether the fault persists based on the swept frequency acoustic signal includes:

[0075] Feature extraction is performed on the frequency sweep acoustic signal to determine the fault feature intensity at the start of the frequency sweep;

[0076] Based on the fault characteristic intensity at the start of the frequency sweep, the fault characteristic intensity at the end of the frequency sweep is obtained, and the attenuation rate of the fault characteristic intensity is calculated.

[0077] When the attenuation rate of the fault characteristic intensity is less than the preset attenuation threshold, the fault is determined to persist.

[0078] Optionally, the step of extracting features from the frequency sweep acoustic signal to determine the fault feature intensity at the start of the frequency sweep includes:

[0079] Feature extraction is performed on the frequency sweep acoustic signal at the start of the frequency sweep to obtain the frequency sweep energy focusing value, frequency sweep pulse density, frequency sweep harmonic distortion rate, and frequency sweep feature set;

[0080] Based on the sweep frequency feature set and the benchmark threshold set, the sweep frequency pulse density deviation, sweep frequency energy focusing deviation, and sweep frequency distortion deviation are calculated.

[0081] The frequency sweep pulse density deviation, frequency sweep energy focusing deviation, and frequency sweep distortion deviation are sorted from largest to smallest, and the value with the highest ranking is selected as the fault characteristic intensity at the start of the frequency sweep.

[0082] Specifically, based on the benchmark threshold set used in the initial fault diagnosis result judgment, the sweep pulse density deviation, sweep energy focusing deviation, and sweep distortion deviation are calculated according to the aforementioned method. These deviations are then sorted from largest to smallest, and the value ranked first is selected as the fault characteristic intensity at the start of the sweep. The attenuation rate of this fault characteristic intensity is then calculated. :

[0083] ,

[0084] in The fault characteristic intensity at the start of the frequency sweep is obtained by extracting signal features and calculating deviation at the start of the frequency sweep. The fault characteristic intensity at the end of the frequency sweep is obtained by extracting the signal features at the end of the frequency sweep and calculating the deviation. If the fault characteristic intensity at the beginning of the frequency sweep is the frequency sweep energy focusing deviation, then the fault characteristic intensity at the end of the frequency sweep is also the frequency sweep energy focusing deviation. The frequency sweep time interval, obtained from the inverter control parameters, is measured in seconds; the attenuation rate is measured in seconds. It determines whether the attenuation rate is less than a preset attenuation threshold, which is set based on statistical data from the motor's historical normal operation, also measured in seconds. If the attenuation rate is less than the preset threshold, the fault is considered persistent; otherwise, the fault is considered non-persistent. By exciting the motor's dynamic response through micro-amplitude frequency sweeps and analyzing the rate of change of fault characteristic intensity, slow attenuation indicates a stable and persistent fault characteristic rather than a transient disturbance, thus enhancing diagnostic reliability. It effectively distinguishes between transient anomalies such as electromagnetic noise and actual mechanical faults, reducing the false positive rate and improving the accuracy of confirming persistent faults.

[0085] Optionally, if the fault persists, the energy is decoupled from the swept frequency acoustic signal to calculate the true fault intensity, including:

[0086] If the fault persists, the percentage of the motor's real-time output power to its rated power during the frequency sweep process is collected to obtain the motor load rate.

[0087] Based on the frequency sweep acoustic signal, the frequency sweep harmonic distortion rate during the frequency sweep process is extracted;

[0088] The frequency conversion interference factor is calculated based on the motor load rate and the swept frequency harmonic distortion rate.

[0089] The true fault intensity is calculated using the frequency conversion interference factor and the fault characteristic intensity during the frequency sweep process.

[0090] Specifically, when a fault is determined to be persistent, the real-time output power of the motor during the frequency sweep process is first collected using a power sensor, and the rated power value of the motor is obtained. The motor load rate is then calculated as the real-time output power divided by the rated power. The real-time output power is directly measured by the power sensor, and the rated power is the parameter on the motor nameplate or a preset value. Next, based on the frequency sweep acoustic signal collected during the frequency sweep process, a Fast Fourier Transform analysis is performed. The acoustic energy at the motor's current operating fundamental frequency is extracted as the fundamental acoustic energy, and the sum of the harmonic components at integer multiples of the fundamental frequency is extracted as the electromagnetic harmonic acoustic energy. The frequency sweep harmonic distortion rate is calculated using the aforementioned method. Finally, based on the motor load rate and the frequency sweep harmonic distortion rate, the frequency converter interference factor is calculated. :

[0091] ,

[0092] in Motor load rate, The frequency sweep harmonic distortion rate is used. Next, the fault characteristic intensity is obtained, which is the maximum deviation value in the frequency sweep characteristic set. This value is determined by ranking the maximum values ​​from the frequency sweep pulse density deviation, frequency sweep energy focusing deviation, and frequency sweep distortion deviation using the aforementioned method. Based on the determined deviation, the values ​​during the frequency sweep process are obtained to obtain the fault characteristic intensity during the frequency sweep process. That is, if the fault characteristic intensity at the start of the frequency sweep is the frequency sweep distortion deviation, then the fault characteristic intensity during the frequency sweep process is the corresponding frequency sweep distortion deviation during the frequency sweep process. Finally, the actual fault intensity is calculated using the frequency converter interference factor and the fault characteristic intensity. :

[0093] ,

[0094] in, To determine the fault characteristic intensity during the frequency sweep process, the actual fault intensity changes during the sweep are obtained. Through the linkage analysis of acoustic signature features and frequency converter parameters, characteristic frequency bands are first divided based on the motor resonance point, and a dynamic baseline threshold library is established to achieve primary fault diagnosis. Subsequently, acoustic signature signals are collected during the micro-amplitude frequency sweep process to determine fault persistence. Finally, energy decoupling is performed by combining motor load rate and frequency sweep harmonic distortion rate to eliminate frequency converter interference, calculate the actual fault intensity, and generate fault classifications. This effectively distinguishes between real mechanical faults and electromagnetic noise interference, improves the reliability and accuracy of fault diagnosis, enhances the detection capability of weak early faults, reduces the false positive rate, and optimizes the quantitative assessment of fault severity, thereby providing more accurate support for motor maintenance decisions.

[0095] Optionally, generating the frequency-fault intensity mapping map and fault classification results based on the actual fault intensity includes:

[0096] A frequency-fault intensity mapping map is generated by using the sampling frequency points within the frequency sweep range and the actual fault intensity corresponding to the sampling frequency points.

[0097] Based on the frequency-fault intensity mapping, the true fault intensity peak value is obtained;

[0098] When the peak value of the actual fault intensity is less than a preset first threshold, a normal classification is output.

[0099] When the peak value of the actual fault intensity is greater than or equal to the first threshold and less than the preset second threshold, a minor fault classification is output.

[0100] When the peak value of the actual fault intensity is greater than or equal to the second threshold, a severe fault classification is output;

[0101] When the minor fault classification or the severe fault classification is output, the fault characteristic frequency is obtained based on the actual fault intensity peak value and the frequency-fault intensity mapping diagram.

[0102] The fault characteristic frequencies are matched with a preset fault characteristic frequency database to obtain the fault classification results.

[0103] Specifically, firstly, the sequence of sampling frequency points recorded during the frequency sweep process and the actual fault intensity value corresponding to each sampling frequency point are obtained; the sampling frequency point sequence is extracted from the inverter control log, and the unit is Hertz; the actual fault intensity value is calculated by the aforementioned method. A scatter plot is drawn with frequency as the abscissa and actual fault intensity as the ordinate, and a continuous curve is generated by cubic spline interpolation to form a frequency-fault intensity mapping map. The peak value of the actual fault intensity is obtained from the frequency-fault intensity mapping map. Then, the peak value of the actual fault intensity is read and compared with a preset first threshold and a second threshold; the first and second thresholds are obtained statistically from the historical fault database based on the motor model, and the second threshold is greater than the first threshold. When the peak value of the actual fault intensity is less than the first threshold, a normal classification result is output. When the peak value of the actual fault intensity is greater than or equal to the first threshold and less than the second threshold, a minor fault classification result is output. When the peak value of the actual fault intensity is greater than or equal to the second threshold, a severe fault classification result is output. If a minor or severe fault classification result is output, the frequency point corresponding to the peak value of the actual fault intensity value is located in the frequency-fault intensity mapping map as the fault characteristic frequency. Finally, the fault characteristic frequencies are matched with a pre-set fault characteristic frequency library. This library stores the characteristic frequency ranges corresponding to typical fault types (such as the bearing outer ring fault characteristic frequency range, rotor bar breakage characteristic frequency range, etc.), and is pre-calculated using motor model parameters and physical fault models. A nearest neighbor matching algorithm is used to calculate the absolute difference between the fault characteristic frequency and the median value of each fault characteristic frequency range in the library, selecting the fault type with the smallest difference as the fault classification result. By obtaining the mapping relationship between frequency and actual fault intensity through a frequency sweep process, combined with multi-level threshold judgment and a fault characteristic frequency matching mechanism, the quantitative classification of motor fault severity and accurate identification of fault types are achieved. This effectively improves the interpretability of fault diagnosis results and the value of maintenance guidance, while avoiding misclassification problems caused by frequency converter interference.

[0104] Based on the same inventive concept, such as Figure 3 As shown, the present invention also provides a motor fault diagnosis system based on the linkage of acoustic signature features and frequency conversion parameters, the system comprising:

[0105] The frequency band division module is used to obtain the operating frequency range of the motor and divide the operating frequency range into multiple characteristic frequency bands according to the resonance point of the motor.

[0106] The dynamic baseline generation module is used to collect historical acoustic signature signals of motor operation in various characteristic frequency bands and extract features to generate a dynamic baseline threshold library.

[0107] The forward diagnostic module is used to monitor the actual operating frequency and actual acoustic signature signal of the motor in real time. Based on the characteristic frequency band of the actual operating frequency, it selects the corresponding dynamic baseline threshold in the dynamic baseline threshold library to perform acoustic signature feature comparison and outputs the primary fault diagnosis result.

[0108] The reverse diagnostic module is used to control the frequency converter to perform micro-sweep within a preset frequency amplitude range based on the primary fault diagnosis results, collect the sweep acoustic signal during the sweep process, and determine whether the fault persists based on the sweep acoustic signal.

[0109] The fault decoupling module is used to decouple the swept frequency acoustic signal by energy if the fault persists, and calculate the true fault intensity.

[0110] The fault determination module is used to generate a frequency-fault intensity mapping diagram and fault classification results based on the actual fault intensity.

[0111] Example 1

[0112] To verify the feasibility and advancement of this invention in practice, it was applied to the logistics sorting system of a large-scale automated warehousing center. This system has hundreds of motors controlled by frequency converters to drive high-speed conveyor belts. The stable operation of these motors is crucial to ensuring the efficiency of the entire warehousing center. Traditional periodic maintenance methods are costly and struggle to detect early-stage faults. Sudden motor failures (such as bearing wear or rotor imbalance) often cause sorting lines to shut down, resulting in significant economic losses.

[0113] In this embodiment, the automated warehousing center deployed the fault diagnosis system described in this invention to continuously monitor and analyze the data of 50 "SEW-Eurodrive DFV100M4" variable frequency motors on one of the key sorting lines for 6 months.

[0114] The system first acquired historical vibration spectrum data of the motor model under normal operating conditions. Using a peak detection algorithm, it identified significant structural resonance points around 25Hz and 60Hz. Therefore, the system divided the motor's rated operating frequency range (0-100Hz) into three characteristic frequency bands: low frequency (0-25Hz), mid-frequency (25-60Hz), and high-frequency (60-100Hz). Subsequently, the system collected historical acoustic signature signals from healthy motors operating for over 1000 hours within each characteristic frequency band. By extracting features from these signals, a dynamic baseline threshold was established for each characteristic frequency band. Taking the mid-frequency band (25-60Hz) as an example, its baseline threshold was determined as follows: baseline energy focus value 0.12, baseline pulse density 8 Hz, and baseline harmonic distortion rate 0.05. These data constitute a dynamic baseline threshold library, providing an adaptive reference standard for subsequent real-time diagnosis.

[0115] During a six-month monitoring period, the system continuously analyzed the real-time operating frequency and acoustic signature of the motors. At 10:30 AM on November 15, 2024, the system detected an anomaly in the real-time acoustic signature of motor M-07 on conveyor belt 7, operating at 48Hz (a mid-frequency band). The system extracted its features and compared them with the mid-frequency baseline in the dynamic baseline library, finding an actual pulse density as high as 32 pulses / second. The calculated deviation was (32-8) / 8 = 3.0, or 300%, far exceeding the preset 100% deviation threshold. The system therefore issued a "primary fault" alarm. To verify the persistence of the fault, the system immediately controlled the motor's inverter, performing a ±2Hz micro-sweep (from 46Hz to 50Hz) around the current operating frequency of 48Hz for 4 seconds. The system collected the acoustic signature during the sweep and analyzed the pulse density, the primary deviation indicator. At the start of the frequency sweep (46Hz), the pulse density deviation was 2.9; at the end of the sweep (50Hz), the deviation was 2.85. The calculated attenuation rate of the fault characteristic intensity was only (2.9-2.85) / 4=0.0125 / s. This value is much smaller than the attenuation threshold of 0.5 / s used to distinguish transient interference, therefore the system determines that the fault is a persistent real mechanical fault.

[0116] After confirming the persistent existence of the fault, the system entered the fault decoupling and quantification phase. The system acquired a real-time load rate of 0.75 for motor M-07 during the frequency sweep process, and simultaneously measured its frequency sweep harmonic distortion rate to be 0.20. Based on this, the system calculated the frequency converter interference factor to be 0.75 × 0.20 = 0.15. Using this interference factor, the system decoupled the fault characteristic intensity (i.e., pulse density deviation) measured during the frequency sweep process to eliminate the influence of electromagnetic noise. The calculated true fault intensity was 2.9 × (1 - 0.15) ≈ 2.47. This value more accurately reflects the degree of damage to the mechanical structure itself.

[0117] The system generates a "frequency-fault intensity mapping map" by utilizing the actual fault intensity corresponding to each frequency point during the frequency sweep process. For example... Figure 4 As shown, the actual fault intensity reaches a peak of 2.55 at 47.5 Hz. This peak value is between the preset "minor fault" threshold (1.5) and "serious fault" threshold (4.0), so the system classifies the severity of the fault as "minor fault".

[0118] Simultaneously, the system matched the fault characteristic frequency of 47.5Hz corresponding to the peak value with a preset fault characteristic frequency library. This frequency highly matched the characteristic frequency range of "early wear of bearing outer ring" (45Hz-50Hz, coupled with rotational frequency and harmonics) in the library. Ultimately, the system output the diagnostic result: "Motor M-07 has a minor bearing outer ring wear fault; repair is recommended during the next scheduled maintenance window." Based on this accurate diagnosis, the maintenance team prepared spare parts in advance and replaced the bearing of the motor over the weekend, avoiding an unplanned downtime and saving approximately 8 hours of production delay.

[0119] Table 1 Comparison of Accuracy and False Alarm Rate in Motor Fault Diagnosis

[0120]

[0121] Table 2 Typical Fault Diagnosis Event Record Table

[0122]

[0123] As can be seen from the above data and tables, the method of this invention exhibits superior performance in practical applications. Table 1 clearly demonstrates the significant advantages of this invention in terms of early warning accuracy and false alarm rate compared to traditional methods, proving the effectiveness of the dynamic baseline and micro-amplitude frequency sweep verification mechanism. Table 2 records several typical diagnostic cases, which not only successfully identified real mechanical faults but also accurately distinguished instantaneous electromagnetic interference, avoiding unnecessary maintenance and fully demonstrating the system's anti-interference capability and high reliability. By quantifying fault intensity and accurately classifying faults, the system provides the maintenance team with clear and actionable decision-making basis, ultimately realizing the transformation from "passive maintenance" to "predictive maintenance," significantly improving the operational stability and economic benefits of the entire warehousing system.

[0124] It should be noted that the formulas described above, through the principle of dimensional consistency and mathematical standardization methods (such as normalization, dimensionless parameter conversion, or unit system unification), can translate physical quantities with different properties into unitless standard values ​​or parameters that can be superimposed in the same dimension. This eliminates the interference of different dimensions on the computational logic, allowing the formulas to retain the original data distribution characteristics while possessing mathematical rationality and adaptability to objective laws. These are conventional technical methods and will not be elaborated further. The electrical connections between the various units described above do not necessarily represent direct or indirect connections; any indirect connection method is applicable to the embodiments of this invention as long as it achieves the purpose of this invention. The above descriptions are merely exemplary embodiments of this invention and should not be construed as limiting the scope of this invention.

[0125] All equivalent changes and modifications made in accordance with the teachings of this invention are still within the scope of this invention. Those skilled in the art will readily conceive of other embodiments of this invention upon considering the specification and the disclosure of practical truth. This application is intended to cover any variations, uses, or adaptations of this invention that follow the general principles of this invention and include common knowledge or conventional techniques in the art not described herein.

Claims

1. A motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters, characterized in that, The method includes: Obtain the operating frequency range of the motor, and divide the operating frequency range into multiple characteristic frequency bands based on the motor's resonance point; Historical acoustic signature signals of motor operation within each characteristic frequency band are collected and feature extracted to generate a dynamic baseline threshold library. This includes collecting historical acoustic signature signals of motor operation within each characteristic frequency band and extracting the preset fault band energy, full-band energy, acoustic signal amplitude, fundamental acoustic energy, and electromagnetic harmonic acoustic energy for the corresponding characteristic frequency band; calculating the ratio of the preset fault band energy to the full-band energy as the frequency domain energy focusing value, and calculating a reference energy focusing value based on the frequency domain energy focusing value; counting the number of times the acoustic signal amplitude exceeds a preset impact threshold per unit time as the time domain pulse density, and calculating a reference pulse density based on the time domain pulse density; measuring the ratio of the fundamental acoustic energy to the electromagnetic harmonic acoustic energy as the harmonic distortion rate, and calculating a reference harmonic distortion rate based on the harmonic distortion rate; and integrating the reference energy focusing value, reference pulse density, and reference harmonic distortion rate of each characteristic frequency band to obtain the dynamic baseline threshold library. The actual operating frequency and actual acoustic signature of the motor are monitored in real time. Based on the characteristic frequency band of the actual operating frequency, the corresponding dynamic baseline threshold in the dynamic baseline threshold library is selected for acoustic signature feature comparison, and the primary fault diagnosis result is output. Based on the initial fault diagnosis results, the frequency converter is controlled to perform micro-sweep within a preset frequency amplitude range, and the sweep acoustic signal during the sweep process is collected. Based on the sweep acoustic signal, it is determined whether the fault persists. If the fault persists, the energy of the swept frequency acoustic signal is decoupled to calculate the true fault intensity. Based on the actual fault intensity, a frequency-fault intensity mapping diagram and fault classification results are generated.

2. The motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to claim 1, characterized in that, The step of obtaining the operating frequency range of the motor, and dividing the operating frequency range into multiple characteristic frequency bands based on the motor's resonance point, includes: Acquire historical vibration spectrum data of the motor and identify points of sudden increase in vibration energy as resonance points; Using adjacent resonance points as boundaries, the operating frequency range is divided into high-frequency, mid-frequency, and low-frequency bands.

3. The motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to claim 1, characterized in that, The output of the primary fault diagnosis results includes: Real-time monitoring of the motor's actual operating frequency and actual acoustic signature signal; Based on the actual operating frequency, the reference energy focusing value, reference pulse density and reference harmonic distortion rate of the corresponding characteristic frequency band are selected from the dynamic baseline threshold library to obtain the reference threshold set; Feature extraction is performed on the actual voiceprint signal to obtain the actual energy focusing value, actual pulse density, actual harmonic distortion rate, and actual feature set; Calculate the deviation between the actual feature set and the benchmark threshold set, and generate a preliminary fault diagnosis result based on the deviation.

4. The motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to claim 3, characterized in that, The step of calculating the deviation between the actual feature set and the benchmark threshold set, and generating a preliminary fault diagnosis result based on the deviation, includes: The deviation of the actual pulse density is calculated based on the actual pulse density and the corresponding reference pulse density. The actual energy focusing deviation is calculated based on the actual energy focusing value and the corresponding reference energy focusing value. The actual distortion deviation is calculated based on the actual harmonic distortion rate and the corresponding reference harmonic distortion rate. Determine whether the actual pulse density deviation is greater than a preset pulse density deviation threshold, or the actual energy focusing deviation is greater than a preset energy focusing deviation threshold, or the actual distortion deviation is greater than a preset distortion deviation threshold. If so, then a fault is determined to exist; If not, then it is determined that there is no fault.

5. The motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to claim 3, characterized in that, The step of determining whether the fault persists based on the swept frequency acoustic signal includes: Feature extraction is performed on the frequency sweep acoustic signal to determine the fault feature intensity at the start of the frequency sweep; Based on the fault characteristic intensity at the start of the frequency sweep, the fault characteristic intensity at the end of the frequency sweep is obtained, and the attenuation rate of the fault characteristic intensity is calculated. When the attenuation rate of the fault characteristic intensity is less than the preset attenuation threshold, the fault is determined to persist.

6. The motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to claim 5, characterized in that, The step of extracting features from the frequency sweep acoustic signal to determine the fault feature intensity at the start of the frequency sweep includes: Feature extraction is performed on the frequency sweep acoustic signal at the start of the frequency sweep to obtain the frequency sweep energy focusing value, frequency sweep pulse density, frequency sweep harmonic distortion rate, and frequency sweep feature set; Based on the sweep frequency feature set and the benchmark threshold set, the sweep frequency pulse density deviation, sweep frequency energy focusing deviation, and sweep frequency distortion deviation are calculated. The frequency sweep pulse density deviation, frequency sweep energy focusing deviation, and frequency sweep distortion deviation are sorted from largest to smallest, and the value with the highest ranking is selected as the fault characteristic intensity at the start of the frequency sweep.

7. The motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to claim 5, characterized in that, If the fault persists, the swept frequency acoustic signature signal is decoupled by energy, and the actual fault intensity is calculated, including: If the fault persists, the percentage of the motor's real-time output power to its rated power during the frequency sweep process is collected to obtain the motor load rate. Based on the frequency sweep acoustic signal, the frequency sweep harmonic distortion rate during the frequency sweep process is extracted; The frequency conversion interference factor is calculated based on the motor load rate and the swept frequency harmonic distortion rate. The true fault intensity is calculated using the frequency conversion interference factor and the fault characteristic intensity during the frequency sweep process.

8. The motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters according to claim 1, characterized in that, The generation of the frequency-fault intensity mapping map and fault classification results based on the actual fault intensity includes: A frequency-fault intensity mapping map is generated by using the sampling frequency points within the frequency sweep range and the actual fault intensity corresponding to the sampling frequency points. Based on the frequency-fault intensity mapping, the true fault intensity peak value is obtained; When the peak value of the actual fault intensity is less than a preset first threshold, a normal classification is output. When the peak value of the actual fault intensity is greater than or equal to the first threshold and less than the preset second threshold, a minor fault classification is output. When the peak value of the actual fault intensity is greater than or equal to the second threshold, a severe fault classification is output; When the minor fault classification or the severe fault classification is output, the fault characteristic frequency is obtained based on the actual fault intensity peak value and the frequency-fault intensity mapping diagram. The fault characteristic frequencies are matched with a preset fault characteristic frequency database to obtain the fault classification results.

9. A motor fault diagnosis system based on the linkage of acoustic signature features and frequency conversion parameters, applied to the motor fault diagnosis method based on the linkage of acoustic signature features and frequency conversion parameters as described in any one of claims 1-8, characterized in that, The system includes: The frequency band division module is used to obtain the operating frequency range of the motor and divide the operating frequency range into multiple characteristic frequency bands according to the resonance point of the motor. A dynamic baseline generation module is used to collect historical acoustic signature signals of motor operation within each characteristic frequency band and extract features to generate a dynamic baseline threshold library. This includes collecting historical acoustic signature signals of motor operation within each characteristic frequency band and extracting the preset fault band energy, full-band energy, acoustic signal amplitude, fundamental acoustic energy, and electromagnetic harmonic acoustic energy for the corresponding characteristic frequency band; calculating the ratio of the preset fault band energy to the full-band energy as the frequency domain energy focusing value, and calculating a reference energy focusing value based on the frequency domain energy focusing value; counting the number of times the acoustic signal amplitude exceeds a preset impact threshold per unit time as the time domain pulse density, and calculating a reference pulse density based on the time domain pulse density; measuring the ratio of the fundamental acoustic energy to the electromagnetic harmonic acoustic energy as the harmonic distortion rate, and calculating a reference harmonic distortion rate based on the harmonic distortion rate; and integrating the reference energy focusing value, reference pulse density, and reference harmonic distortion rate of each characteristic frequency band to obtain the dynamic baseline threshold library. The forward diagnostic module is used to monitor the actual operating frequency and actual acoustic signature signal of the motor in real time. Based on the characteristic frequency band of the actual operating frequency, it selects the corresponding dynamic baseline threshold in the dynamic baseline threshold library to perform acoustic signature feature comparison and outputs the primary fault diagnosis result. The reverse diagnostic module is used to control the frequency converter to perform micro-sweep within a preset frequency amplitude range based on the primary fault diagnosis results, collect the sweep acoustic signal during the sweep process, and determine whether the fault persists based on the sweep acoustic signal. The fault decoupling module is used to decouple the energy of the swept frequency acoustic signal and calculate the true fault intensity if the fault persists. The fault determination module is used to generate a frequency-fault intensity mapping diagram and fault classification results based on the actual fault intensity.

Images