Brushless direct current motor health management method and system

Abstract

The invention provides a brushless direct current motor health management method and system. The method comprises the following steps: S1, collecting a vibration signal and a temperature signal of a brushless direct current motor; s2, performing denoising and filtering preprocessing on the vibration signal and the temperature signal to obtain a preprocessed vibration signal and a preprocessed temperature signal; s3, extracting vibration characteristic parameters of the brushless direct current motor; acquiring a temperature characteristic parameter reflecting the state of the heating component of the motor; s4, fusing the temperature characteristic parameters and the vibration characteristic parameters in a characteristic layer, matching with a pre-established fault characteristic database, and determining a fault type and a fault position of the brushless direct current motor based on a matching result; and S5, learning the fault type and the fault position by using an artificial intelligence algorithm, and updating the fault feature database based on a learning result. The problems that an existing brushless direct current motor fault diagnosis method is low in diagnosis accuracy, low in efficiency and incapable of early warning in advance are solved.

Description

Technical Field

[0001] This invention relates to the field of motor fault diagnosis technology, and more specifically, to a method and system for health management of brushless DC motors. Background Technology

[0002] With the rapid development of industrial automation, smart home appliances, and new energy vehicles, the stable operation of drive equipment has become a core element in ensuring production efficiency and system reliability. Brushless DC motors, with their significant advantages such as high efficiency, long lifespan, and low maintenance costs, are gradually replacing traditional motors and becoming the mainstream drive device in these fields, finding widespread application in intelligent manufacturing production lines, variable frequency home appliances, and electric vehicle power systems.

[0003] However, brushless DC motors are prone to various potential faults during long-term high-load operation due to factors such as mechanical wear, environmental interference, and electromagnetic stress. Among these, mechanical faults such as abnormal bearing wear, shaft eccentricity, and rotor dynamic imbalance are particularly common. If these faults are not detected in time, they can not only lead to a sharp drop in motor operating efficiency, but in severe cases, they can also cause sudden motor shutdowns, resulting in production line stoppages and cascading damage to equipment, causing huge economic losses to enterprises.

[0004] Traditional methods for diagnosing brushless DC motor faults heavily rely on manual experience and simple testing techniques: technicians sense vibration intensity by touching the motor casing, judge whether operating noise is abnormal by hearing, or use tools such as multimeters to measure basic parameters such as current and voltage. This approach has significant limitations: firstly, the diagnostic results are greatly influenced by the operator's experience level, are highly subjective, and are prone to missed or misdiagnosed cases; secondly, it can only make judgments after a fault has appeared, failing to provide early warnings and thus failing to meet the needs of modern industry for predictive maintenance of equipment.

[0005] With the rapid development of sensor technology, signal processing algorithms, and artificial intelligence, fault diagnosis methods based on vibration signal analysis have gradually become a research focus. This method involves installing vibration sensors at key parts of the motor to collect vibration signals in real time during operation. The vibration signals are then subjected to spectral decomposition and noise suppression to extract fault features, thereby achieving motor fault diagnosis. Although this represents a significant improvement over traditional methods, existing technologies still have shortcomings: the diagnostic accuracy is low for complex scenarios such as multi-fault coupling and early, weak faults; vibration signals often contain inherent equipment noise and environmental interference, leading to insufficient stability and reliability of fault feature extraction, making it difficult to meet the requirements of industrial sites.

[0006] A patent search revealed invention patent CN116859240A, which discloses a method for fault identification of brushless DC motors based on multi-sensor and edge computing. The method includes demodulating the real-time vibration signal of the brushless DC motor and acquiring its three-phase current signal; extracting features from the real-time vibration signal and three-phase current signal to obtain motor fault identification features; using a backpropagation neural network to fuse the motor fault identification features and generate a feature fusion result; and using a neural network to infer the feature fusion result to identify the motor fault state. This patent does not clearly define fault location and relies solely on vibration and current signals, making it difficult to comprehensively reflect the motor's condition.

[0007] In summary, given the problems of the existing technologies, researching a health management method and system for brushless DC motors has become a critical task that urgently needs to be addressed. Summary of the Invention

[0008] To address the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for health management of brushless DC motors.

[0009] A brushless DC motor health management method according to the present invention includes the following steps: Step S1: Collect vibration and temperature signals of the brushless DC motor using various sensors installed on the brushless DC motor. Step S2: Denoise and filter the vibration signal and temperature signal to obtain the preprocessed vibration signal and preprocessed temperature signal. Step S3: Perform spectrum analysis and noise analysis on the preprocessed vibration signal to extract the vibration characteristic parameters of the brushless DC motor; perform feature extraction on the preprocessed temperature signal to obtain the temperature characteristic parameters reflecting the state of the motor's heating components. Step S4: The temperature characteristic parameters and vibration characteristic parameters are fused in the feature layer, and the fused comprehensive characteristic parameters are matched with the pre-established fault characteristic database. Based on the matching results, the fault type and fault location of the brushless DC motor are determined. Step S5: Use artificial intelligence algorithms to learn the fault type and fault location, and update the fault feature database based on the learning results.

[0010] Preferably, in step S1, the sensor includes a vibration sensor and a temperature sensor. The vibration sensor converts the collected vibration analog signal into a vibration electrical signal, and the temperature sensor converts the collected temperature analog signal into a temperature electrical signal.

[0011] Preferably, in step S1, the vibration sensor is disposed on the bearing housing, and the temperature sensor is installed in the corresponding winding area of ​​the housing.

[0012] Preferably, in step S2, bandpass filtering and lowpass filtering are used to filter the frequency band of the vibration electrical signal, and then wavelet transform and mean filtering are used for noise reduction to obtain the pre-processed vibration signal; the temperature electrical signal is filtered to obtain the pre-processed temperature signal.

[0013] Preferably, in step S3, the preprocessed vibration signal is converted from the time domain to the frequency domain using a fast Fourier transform to extract the frequency characteristics of the preprocessed vibration signal; a combination of power spectrum estimation and cepstral analysis is used to perform noise analysis on the preprocessed vibration signal to extract the noise characteristics of the preprocessed vibration signal; the frequency characteristics and noise characteristics constitute the vibration characteristic parameters of the brushless DC motor. Feature extraction is performed on the preprocessed temperature signal to calculate its time-domain and frequency-domain features. The time-domain features include mean, variance, temperature fluctuation amplitude, and temperature change rate, while the frequency-domain features include the temperature spectrum extracted through Fourier transform.

[0014] Preferably, in step S4, the fault feature database includes characteristic parameters of the brushless DC motor under normal operating conditions and characteristic parameters under various typical fault conditions, and is established based on statistical analysis of experimental data and actual operating data.

[0015] Preferably, in step S4, the fused integrated feature parameters are matched and compared with the fault feature database. Based on the similarity and difference between the integrated feature parameters and the feature parameters in the fault feature database, the fault type and fault location of the brushless DC motor are determined.

[0016] Preferably, in step S5, an artificial intelligence algorithm combining support vector machine and deep learning is used to learn the fault type and fault location, and the fault feature database is updated based on the learning results.

[0017] The present invention also provides a brushless DC motor health management system, comprising: The sensor module collects vibration and temperature signals from the brushless DC motor through various sensors installed on the brushless DC motor. The signal preprocessing module includes hardware filtering circuits and digital filtering algorithms to perform noise reduction and filtering preprocessing on vibration and temperature signals, resulting in preprocessed vibration and temperature signals. The feature extraction module uses a digital signal processor or field-programmable gate array to perform spectrum analysis and noise analysis on the preprocessed vibration signal to extract the vibration characteristic parameters of the brushless DC motor; it also performs feature extraction on the preprocessed temperature signal to obtain temperature characteristic parameters reflecting the state of the motor's heating components. The fault location and database module is deployed on the server platform. It fuses temperature characteristic parameters and vibration characteristic parameters at the feature layer, and matches the fused comprehensive characteristic parameters with the pre-established fault characteristic database. Based on the matching results, it determines the fault type and fault location of the brushless DC motor. The AI ​​learning module uses artificial intelligence algorithms to learn fault types and fault locations, and updates the fault feature database based on the learning results.

[0018] Preferably, the brushless DC motor health management system further includes: The display module is used to display the operating status, fault type and fault location of the brushless DC motor. The display module is implemented using an LCD or touch screen. The alarm module is used to issue an alarm signal when a fault in the brushless DC motor is detected. The alarm module supports audible and visual alarms as well as remote notification.

[0019] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention extracts vibration characteristic parameters of multiple types of faults by fusing the frequency and noise characteristics of vibration signals; and combines artificial intelligence algorithms to diagnose faults based on the vibration characteristic parameters, thereby achieving accurate identification of fault type and fault location.

[0020] 2. This invention optimizes the installation position and fixing method of the sensor based on the structure and fault characteristics of the motor, ensuring the quality of the original vibration signal acquisition, reducing the impact of interference on the diagnostic results from the source, and improving the accuracy and stability of fault judgment.

[0021] 3. This invention, through real-time monitoring of the motor's operating status and analysis of vibration signals, can identify potential hazards in the early stages of a fault and automatically trigger an early warning mechanism. To ensure the accuracy of the early warning, a fault feature database is constructed based on massive experimental data and field operation data. Through rigorous data cleaning and feature screening, the representativeness of fault features under different operating conditions and environments is ensured, thereby providing reliable data support for early warning and avoiding downtime losses caused by the escalation of faults.

[0022] 4. This invention minimizes the number of sudden shutdowns by promptly detecting and addressing motor faults, ensuring the continuous and stable operation of the motor and related systems. Simultaneously, it periodically calibrates sensors to ensure signal acquisition accuracy. Utilizing the self-learning mechanism of artificial intelligence algorithms, it continuously iterates and optimizes the fault feature model based on historical diagnostic results, achieving dynamic updates to the fault feature database. Furthermore, by adaptively adjusting algorithm parameters to address the varying operating conditions of different application scenarios, it flexibly adapts to diverse fault diagnosis needs, enhancing its adaptability and generalization capabilities in complex environments. Attached Figure Description

[0023] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart of a brushless DC motor health management method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a brushless DC motor health management system according to an embodiment of the present invention. Detailed Implementation

[0024] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0025] This invention provides a health management method and system for brushless DC motors. The method includes the following steps: S1, collecting vibration and temperature signals of the brushless DC motor using various sensors installed on the motor; S2, performing noise reduction and filtering preprocessing on the vibration and temperature signals to obtain preprocessed vibration and temperature signals; S3, performing spectrum and noise analysis on the preprocessed vibration signals to extract vibration characteristic parameters of the brushless DC motor; and extracting features from the preprocessed temperature signals to obtain temperature characteristic parameters reflecting the state of the motor's heating components; S4, fusing the temperature and vibration characteristic parameters at a feature layer, and matching the fused comprehensive characteristic parameters with a pre-established fault characteristic database to determine the fault type and location of the brushless DC motor based on the matching results; and S5, using artificial intelligence algorithms to learn the fault type and location, and updating the fault characteristic database based on the learning results. This invention solves the problems of low diagnostic accuracy, low efficiency, and inability to provide early warnings in existing brushless DC motor fault diagnosis methods.

[0026] Example: Figure 1 This is a flowchart of a brushless DC motor health management method according to an embodiment of the present invention.

[0027] like Figure 1 As shown, this embodiment provides a health management method for a brushless DC motor, including the following steps: Step S1: Vibration and temperature signals of the brushless DC motor are collected using various sensors installed on the brushless DC motor.

[0028] Specifically, the sensors include vibration sensors and temperature sensors. The vibration sensor converts the collected vibration analog signals into vibration electrical signals, and the temperature sensor converts the collected temperature analog signals into temperature electrical signals.

[0029] Furthermore, vibration and temperature sensors are installed in key locations based on the structure and fault characteristics of the brushless DC motor to obtain signals that best reflect its operating status.

[0030] Specifically, based on the occurrence mechanism of common fault types in brushless DC motors, the characteristic regions with the most significant vibration and temperature signal responses are identified, and sensors are preferentially placed in these sensitive areas. Simultaneously, in conjunction with the motor's structural design, locations that can directly or indirectly characterize the operating status of key components are selected for installation. Furthermore, vibration sensors are placed in the bearing housing to directly capture bearing vibration, and temperature sensors are installed in the corresponding winding area of ​​the housing to monitor winding temperature changes.

[0031] In this embodiment, vibration sensors are arranged on the bearing housing end face, near the rotor ends of the outer casing, and at the connection between the stator core and the outer casing of the brushless DC motor, either by bolt fixing or magnetic adsorption. Bolting is suitable for long-term stable monitoring, ensuring a tight fit between the sensor and the motor surface, reducing signal attenuation during vibration transmission, and is particularly suitable for critical vibration sources such as bearing housings. Magnetic adsorption allows for flexible position adjustment and is suitable for temporary testing or scenarios requiring periodic replacement of monitoring points. When used near the rotor ends of the outer casing, it acquires vibration signals generated by rotor rotation. These positions can directly capture vibration signals generated by mechanical faults such as abnormal bearing wear and rotor dynamic imbalance, and the signals contain richer fault characteristic information.

[0032] Temperature sensors are mounted on the winding ends, bearing outer rings, and the base of corresponding heat sink fins on the brushless DC motor, either by adhesive bonding or embedded installation. Adhesive bonding is suitable for winding ends and the surface of the housing, enabling real-time monitoring of winding heat generation and housing heat dissipation; it is convenient to install and does not damage the motor structure. Embedded installation is used for areas requiring deeper monitoring, such as the bearing outer ring. By partially embedding the sensor in pre-drilled holes in the bearing housing, it can more accurately capture temperature changes caused by bearing overheating due to friction. These locations effectively reflect the state of the core heat-generating components during motor operation, providing temperature characteristics for diagnosing faults such as winding overload and bearing lubrication failure.

[0033] Step S2 involves denoising and filtering the vibration and temperature signals to improve signal quality and recognizability, resulting in preprocessed vibration and temperature signals.

[0034] Specifically, bandpass and lowpass filtering are used to filter the vibration electrical signal to remove high-frequency noise and low-frequency interference. Then, wavelet transform and mean filtering are used for denoising to remove high-frequency noise and random interference, resulting in a pre-processed vibration signal. The temperature electrical signal is also filtered to obtain a pre-processed temperature signal.

[0035] Step S3 involves performing spectrum and noise analysis on the preprocessed vibration signal to extract vibration characteristic parameters of the brushless DC motor related to faults such as bearing abnormality, shaft wear, and rotational deviation. Simultaneously, feature extraction is performed on the preprocessed temperature signal to calculate its time-domain and frequency-domain characteristics in order to obtain temperature characteristic parameters reflecting the state of the motor's heating components. The time-domain characteristics include mean, variance, temperature fluctuation amplitude, and temperature change rate, while the frequency-domain characteristics include the temperature spectrum extracted through Fourier transform.

[0036] Specifically, the preprocessed vibration signal is transformed from the time domain to the frequency domain using Fast Fourier Transform (FFT). The frequency components and energy distribution of the signal are analyzed, and frequency features related to faults such as bearing abnormalities, shaft wear, and rotational deviation are extracted, including fault characteristic frequencies and harmonic frequencies. A combination of power spectrum estimation and cepstral analysis is used to perform noise analysis on the preprocessed vibration signal, analyzing the noise components and characteristics, and extracting noise features such as noise power and noise spectrum characteristics. These frequency and noise features constitute the vibration characteristic parameters of the brushless DC motor.

[0037] Step S4: The temperature characteristic parameters and vibration characteristic parameters are fused in the feature layer, and the fused comprehensive characteristic parameters are matched with the pre-established fault characteristic database. Based on the matching results, the fault type and fault location of the brushless DC motor are determined.

[0038] Specifically, the fault characteristic database contains characteristic parameters of brushless DC motors under normal operating conditions and characteristic parameters under various typical fault conditions, and is established based on statistical analysis of experimental data and actual operating data.

[0039] Furthermore, the fused integrated feature parameters are matched and compared with the fault feature database. Based on the similarity and difference between the integrated feature parameters and the feature parameters in the fault feature database, the fault type and fault location of the brushless DC motor are determined.

[0040] In this embodiment, the brushless DC motor is judged to be operating normally by comparing the similarity between the comprehensive feature parameters and the normal state feature parameters in the fault feature database; the brushless DC motor is judged to be faulty and fault type by comparing the similarity between the comprehensive feature parameters and the various fault state feature parameters in the fault feature database, and the fault location is determined.

[0041] Step S5: Use artificial intelligence algorithms to learn the fault type and fault location, and update the fault feature database based on the learning results to improve the accuracy and reliability of fault diagnosis.

[0042] Specifically, artificial intelligence algorithms combining support vector machines (SVM) and deep learning (DL) are used to learn fault types and fault locations, and the fault feature database is updated based on the learning results, thereby improving the accuracy and reliability of fault diagnosis.

[0043] Example 2: The present invention also provides a brushless DC motor health management system, which can be implemented by executing the process steps of the brushless DC motor health management method. That is, those skilled in the art can understand the brushless DC motor health management method as a preferred embodiment of the brushless DC motor health management system.

[0044] Figure 2 This is a schematic diagram of the structure of a brushless DC motor health management system according to an embodiment of the present invention.

[0045] like Figure 2 As shown, the brushless DC motor health management system includes: The sensor module collects vibration and temperature signals from the brushless DC motor using various sensors installed on the motor.

[0046] The signal preprocessing module includes hardware filtering circuits and digital filtering algorithms to perform noise reduction and filtering preprocessing on vibration and temperature signals, thereby improving signal quality and recognizability, and obtaining preprocessed vibration and temperature signals.

[0047] The feature extraction module uses a digital signal processor (DSP) or field-programmable gate array (FPGA) to perform spectrum analysis and noise analysis on the pre-processed vibration signal, and extract vibration characteristic parameters of the brushless DC motor related to faults such as bearing abnormality, shaft wear, and rotational deviation; it also performs feature extraction on the pre-processed temperature signal to obtain temperature characteristic parameters reflecting the state of the motor's heating components.

[0048] The fault location and database module is deployed on the server platform. It fuses temperature characteristic parameters and vibration characteristic parameters at the feature layer, and matches the fused comprehensive characteristic parameters with the pre-established fault characteristic database. Based on the matching results, it determines the fault type and fault location of the brushless DC motor. The AI ​​learning module uses artificial intelligence algorithms to learn fault types and fault locations, and updates the fault feature database based on the learning results, thereby improving the accuracy and reliability of fault diagnosis.

[0049] The display module is used to display the operating status, fault type, and fault location of the brushless DC motor. The display module is implemented using an LCD or touch screen.

[0050] The alarm module is used to issue an alarm signal when a fault in the brushless DC motor is detected. The alarm module supports audible and visual alarms as well as remote notification.

[0051] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, enabling the system and its various devices, modules, and units to function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered both software modules implementing the method and structures within the hardware component.

[0052] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method of health management of a brushless DC motor, characterized by, The method comprises the following steps: Step S1, collecting vibration signals and temperature signals of the brushless DC motor through various sensors installed on the brushless DC motor; Step S2, denoising and filtering preprocessing of the vibration signals and the temperature signals to obtain preprocessed vibration signals and preprocessed temperature signals; Step S3, frequency spectrum analysis and noise analysis of the preprocessed vibration signals to extract vibration characteristic parameters of the brushless DC motor; feature extraction of the preprocessed temperature signals to obtain temperature characteristic parameters reflecting the state of the motor heating components; Step S4, fusion of the temperature characteristic parameters and the vibration characteristic parameters in the feature layer, matching of the fused comprehensive characteristic parameters with a pre-established fault feature database, and determination of the fault type and fault location of the brushless DC motor based on the matching result; Step S5, learning of the fault type and the fault location by using an artificial intelligence algorithm, and updating of the fault feature database based on the learning result.

2. The brushless DC motor health management method of claim 1, wherein, In step S1, the sensors include vibration sensors and temperature sensors, the vibration sensors convert the collected vibration analog signals into vibration electrical signals, and the temperature sensors convert the collected temperature analog signals into temperature electrical signals.

3. The brushless DC motor health management method of claim 2, wherein, In step S1, the vibration sensors are arranged on the bearing seat, and the temperature sensors are installed on the shell corresponding to the winding area.

4. The brushless DC motor health management method of claim 3, wherein, In step S2, band-pass filtering and low-pass filtering are used to screen the frequency band of the vibration electrical signals, and then wavelet transform and mean filtering are used for denoising processing to obtain the preprocessed vibration signals; The temperature electrical signals are filtered to obtain the preprocessed temperature signals.

5. The brushless DC motor health management method of claim 1, wherein, In step S3, the preprocessed vibration signals are converted from time domain to frequency domain by using fast Fourier transform to extract the frequency characteristics of the preprocessed vibration signals; power spectrum estimation and cepstrum analysis are combined to analyze the noise of the preprocessed vibration signals to extract the noise characteristics of the preprocessed vibration signals; the frequency characteristics and the noise characteristics constitute the vibration characteristic parameters of the brushless DC motor; The preprocessed temperature signals are feature-extracted to calculate the time domain characteristics and frequency domain characteristics of the temperature signals, the time domain characteristics include mean, variance, temperature fluctuation amplitude and temperature change rate, and the frequency domain characteristics include temperature spectrum extracted by Fourier transform.

6. The brushless DC motor health management method of claim 1, wherein, In step S4, the fault feature database contains characteristic parameters of the brushless DC motor in normal operating state and characteristic parameters in various typical fault states, and is established based on statistical analysis of experimental data and actual operating data.

7. The brushless DC motor health management method of claim 6, wherein, In step S4, the fused comprehensive characteristic parameters are compared with the fault feature database based on the similarity and difference between the comprehensive characteristic parameters and the characteristic parameters in the fault feature database to determine the fault type and fault location of the brushless DC motor.

8. The brushless DC motor health management method of claim 1, wherein, In the step S5, an artificial intelligence algorithm combining support vector machines and deep learning is used to learn the fault type and the fault location, and the fault feature database is updated based on the learning result.

9. A brushless DC motor health management system, characterized by, The application comprises: a sensor module for collecting vibration signals and temperature signals of the brushless direct current motor through various sensors installed on the brushless direct current motor; a signal preprocessing module including a hardware filtering circuit and a digital filtering algorithm for denoising and filtering preprocessing of the vibration signals and the temperature signals to obtain preprocessed vibration signals and preprocessed temperature signals; a feature extraction module for performing frequency spectrum analysis and noise analysis on the preprocessed vibration signals using a digital signal processor or a field programmable gate array to extract vibration characteristic parameters of the brushless direct current motor, and performing feature extraction on the preprocessed temperature signals to obtain temperature characteristic parameters reflecting the state of the motor heating components; a fault positioning and database module deployed on a server platform for fusing the temperature characteristic parameters and the vibration characteristic parameters at the feature layer, matching the fused comprehensive characteristic parameters with a pre-established fault feature database, and determining the fault type and the fault location of the brushless direct current motor based on the matching result; an AI learning module for learning the fault type and the fault location using an artificial intelligence algorithm, and updating the fault feature database based on the learning result.

10. The brushless DC motor health management system of claim 9, wherein, The brushless direct current motor health management system further comprises: a display module for displaying the running state, the fault type and the fault location of the brushless direct current motor, wherein the display module is realized by a liquid crystal display or a touch screen; an alarm module for issuing an alarm signal when the brushless direct current motor fault is detected, wherein the alarm module supports sound and light alarm and remote notification mode.

Images