WO2023138337A1 - Procédé et appareil de détection de défaillance de moteur - Google Patents

Procédé et appareil de détection de défaillance de moteur Download PDF

Info

Publication number
WO2023138337A1
WO2023138337A1 PCT/CN2022/143121 CN2022143121W WO2023138337A1 WO 2023138337 A1 WO2023138337 A1 WO 2023138337A1 CN 2022143121 W CN2022143121 W CN 2022143121W WO 2023138337 A1 WO2023138337 A1 WO 2023138337A1
Authority
WO
WIPO (PCT)
Prior art keywords
motor
data
fault diagnosis
operating data
motors
Prior art date
Application number
PCT/CN2022/143121
Other languages
English (en)
Chinese (zh)
Inventor
王萌
孟超
胡亨捷
刘凌
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2023138337A1 publication Critical patent/WO2023138337A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/28—Testing of electronic circuits, e.g. by signal tracer
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
    • Y04S10/52—Outage or fault management, e.g. fault detection or location

Definitions

  • the present application relates to the field of terminal technology, and in particular to a motor fault detection method and device.
  • the motor is the most common power source equipment for driving various machinery in the production process, and it often plays a very important role in today's modernized production process.
  • the scale of industrial production system has also become larger and larger, resulting in the continuous increase in the use of electrical equipment. If the motor suddenly breaks down or even stops running suddenly without warning, it will not only cause damage to the motor itself, but in severe cases, it may even threaten the safety of the entire production system and cause great losses. Therefore, in order to ensure the stable operation of the motor, many protection measures are often used during the use of the motor, such as stator current overcurrent, stator voltage overvoltage, stator voltage undervoltage, motor negative phase sequence protection, and motor differential protection.
  • the relay By setting various relays to monitor various parameter values of the motor, if some parameter values exceed the set value of the relay, the relay will send out an alarm, and if necessary, the operation of the asynchronous induction motor will be stopped directly by cutting off the control circuit of the motor to prevent further expansion of the scale of production accidents.
  • the protection settings of the asynchronous induction motor seem to be perfect, but in the actual operation process, the relay will only send out an alarm when the motor fails, and there is no early warning function.
  • the application provides a motor fault detection method, device, computer storage medium and computer program product, which can accurately detect motor faults in complex environments and realize early warning of motor faults.
  • the present application provides a motor fault detection method, the method comprising: acquiring first operating data of N different types of motors in different states in an ideal environment, N ⁇ 2; determining hyperparameters of a first neural network according to the first operating data; obtaining second operating data of N motors in an actual environment; using the second operating data to train the first neural network to obtain a fault diagnosis model; using the fault diagnosis model to perform fault diagnosis on N motors, and outputting fault diagnosis results.
  • the hyperparameters of the neural network are determined, and then the operating data of the motor obtained in the actual environment is used to train the neural network to obtain a fault diagnosis model, which realizes the automatic classification of different types of motor samples in the industrial park scene and fault diagnosis under a small number of fault samples, and realizes the self-optimization of the model with the accumulation of fault samples.
  • a fault diagnosis model which realizes the automatic classification of different types of motor samples in the industrial park scene and fault diagnosis under a small number of fault samples, and realizes the self-optimization of the model with the accumulation of fault samples.
  • determining the hyperparameters of the first neural network according to the first operating data specifically includes: respectively determining eigenvectors corresponding to the first operating data of each motor to obtain N first eigenvectors; respectively determining a first distance between every two eigenvectors in the N first eigenvectors; and determining the hyperparameters of the first neural network based on the smallest first distance among the obtained first distances.
  • using the fault diagnosis model to perform fault diagnosis on N motors specifically includes: acquiring third operating data of N motors in an actual environment; inputting the N third operating data into the fault diagnosis model to obtain a fault diagnosis result.
  • the target operation data includes one or more of the following: motor number, voltage data of the motor, current data of the motor, stator vibration data of the motor, or stator temperature data of the motor; wherein, the target operation data is one of the first operation data, the second operation data or the third operation data.
  • the present application provides a motor fault detection device, the device comprising: a model building module, which is used to obtain the first operating data of N different types of motors in different states in an ideal environment, N ⁇ 2; a model building module, which is also used to determine the hyperparameters of the first neural network according to the first operating data; a model building module, which is also used to obtain second operating data of N motors in an actual environment; a model building module, which is also used to use the second operating data to train the first neural network to obtain a fault diagnosis model; And output the fault diagnosis result.
  • the model building module is further configured to: respectively determine the eigenvectors corresponding to the first operating data of each motor to obtain N first eigenvectors; respectively determine the first distance between every two eigenvectors in the N first eigenvectors; determine the hyperparameters of the first neural network based on the smallest first distance among the obtained first distances.
  • the fault diagnosis module is further configured to: obtain third operating data of N motors in an actual environment; input the N third operating data into the fault diagnosis model to obtain a fault diagnosis result.
  • the target operation data includes one or more of the following: motor number, voltage data of the motor, current data of the motor, stator vibration data of the motor, or stator temperature data of the motor; wherein, the target operation data is one of the first operation data, the second operation data or the third operation data.
  • the present application provides a motor fault detection device, including: at least one memory for storing programs; at least one processor for executing the programs stored in the memory, and when the programs stored in the memory are executed, the processor is used for executing the method as provided in the first aspect.
  • the present application provides an electronic device, which includes at least one memory for storing programs and at least one processor for executing the programs stored in the memory. Wherein, when the program stored in the memory is executed, the processor is configured to execute the method as provided in the first aspect.
  • the present application provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program runs on an electronic device, the electronic device executes the method as provided in the first aspect.
  • the present application provides a computer program product, which, when the computer program product is run on an electronic device, causes the electronic device to execute the method as provided in the first aspect.
  • FIG. 1 is a schematic flow chart of a physical model-based fault detection method provided in an embodiment of the present application
  • FIG. 2 is a schematic flow diagram of a supervised fault detection method based on a data-driven model provided in an embodiment of the present application
  • Fig. 3 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • Fig. 4 is a schematic flow chart of a motor fault detection scheme provided by an embodiment of the present application.
  • FIG. 5 is a schematic flow chart of a motor fault detection method provided in an embodiment of the present application.
  • Fig. 6 is a schematic diagram of the hardware structure of a motor fault detection device provided by the embodiment of the present application.
  • Fig. 7 is a schematic diagram of a hardware structure of a motor fault detection device provided by an embodiment of the present application.
  • first and second and the like in the specification and claims herein are used to distinguish different objects, rather than to describe a specific order of objects.
  • first response message and the second response message are used to distinguish different response messages, rather than describing a specific order of the response messages.
  • words such as “exemplary” or “for example” are used as examples, illustrations or illustrations. Any embodiment or design scheme described as “exemplary” or “for example” in the embodiments of the present application shall not be interpreted as being more preferred or more advantageous than other embodiments or design schemes. Rather, the use of words such as “exemplary” or “such as” is intended to present related concepts in a concrete manner.
  • multiple means two or more, for example, a plurality of processing units refers to two or more processing units, etc.; a plurality of elements refers to two or more elements, etc.
  • Fig. 1 shows a schematic flowchart of a fault detection method based on a physical model.
  • the physical model of the motor is constructed, and the parameter changes of the model under different faults are clarified (for example, when a broken rotor bar fault occurs, the rotor magnetomotive force will be multiplied by sin2 ⁇ ).
  • the characteristics of the electrical signal under the model are obtained; then, during the operation of the motor, after the electrical signal (such as voltage, current, etc.) of the motor is obtained, the electrical signal of the motor is analyzed through the physical model of the motor, and finally the fault detection result is obtained, thereby realizing the detection of different motor faults.
  • Fig. 2 shows a schematic flowchart of a supervised fault detection method based on a data-driven model.
  • the fault detection method first collects characteristic parameters related to motor properties based on sensors, associates these characteristic parameters with useful information, and uses machine learning and data mining algorithms to perform model training through supervised learning to obtain a fault detection model.
  • this method requires high data quality and requires a large amount of data labeling.
  • it is difficult to guarantee the generalization of the motor model. Under different motors and different working conditions, it may be necessary to re-train the model (new data is required at this time).
  • the present application also provides another motor fault detection method.
  • This method is mainly to accumulate empirical knowledge through laboratory experiments, determine the hyperparameters of the data-driven model, and then use the data-driven method of self-organization learning to realize the automatic classification of different types of motor samples in the industrial park scene and the fault diagnosis under a small number of fault samples, and realize the self-optimization of the model with the accumulation of fault samples. In this way, the problems of low accuracy and high data cost of the current traditional motor fault detection method caused by various types of motors and complex working conditions in the industrial park are solved.
  • FIG. 3 shows an application scenario provided by the embodiment of the present application.
  • the motor group 100 includes at least two motors
  • the sensor group 200 includes at least one sensor.
  • one motor may correspond to one sensor, or multiple motors may correspond to one sensor.
  • the sensors in the sensor group 200 are mainly used to collect the operation data of the motors in the motor group 100 .
  • the data collected by the sensors in the sensor group 200 may include the motor number (such as the model of the motor, etc.), the voltage of the motor, the current of the motor, etc.; it may also include the vibration data of the motor stator, the temperature data of the motor stator, and the like.
  • the server 300 is mainly used to construct a fault diagnosis model based on the empirical knowledge of the laboratory experiment schedule and the data reported by the sensor group 200, and perform fault diagnosis on the motors in the motor group 100 based on the data reported by the sensor group 200, and output the fault diagnosis results.
  • the connection between the sensor group 200 and the server 300 can be established directly, or indirectly through an edge computing gateway (not shown in the figure). The details may be determined according to actual conditions, and are not limited here.
  • the server 300 can also be replaced with other devices or device clusters capable of realizing the same function, and the replaced solution is still within the protection scope of the present application.
  • Fig. 4 shows a schematic flowchart of a motor fault detection solution.
  • the scheme shown in FIG. 4 may be but not limited to be executed by the server 300 shown in FIG. 3 .
  • the scheme mainly includes four steps: data preprocessing, feature extraction, model construction, model use & model liberalization.
  • the data preprocessing mainly includes: eliminating abnormal values, missing values and/or repeated values, etc. in the original data set obtained from the sensor group 200, and completing the above data by using a data completion method.
  • Feature extraction is mainly to analyze the preprocessed signal data in the time domain and frequency domain, and extract the eigenvalues of the signal in different time periods/frequency bands to form feature vectors.
  • the model construction is mainly to analyze the characteristics of multiple motors under ideal conditions, and determine the characteristic distance vectors of different motors under normal operation as empirical knowledge; then use the above empirical knowledge to determine the hyperparameters of the self-organizing neural network (self-organizing map, SOM), and collect the operating data of the motors for a certain period of time on site to implement model training to obtain a fault diagnosis model.
  • Model usage & model self-optimization is mainly to apply the trained model to fault diagnosis, and based on the online data of the park, continuously accumulate fault modes to realize model self-optimization.
  • the core of the above scheme can be in the step of model construction, which is different from the traditional idea of directly using data to carry out data-driven modeling.
  • the motor fault detection scheme in Figure 4 mainly includes the following steps:
  • Step1 Data preprocessing.
  • the raw data is preprocessed mainly for the following common data abnormalities.
  • Missing data processing Select the value with the most occurrences in the data set to fill in the corresponding missing data.
  • Abnormal data processing Given a normal range band, if the voltage and current data is not within the band, it will be regarded as abnormal data and directly eliminated. In one example, missing data processing can be performed after removing outlier data.
  • the wavelet threshold denoising method can be used. Firstly, the original signal is decomposed by wavelet, and the appropriate threshold value of wavelet coefficient is selected, and the threshold value is compared with the high-frequency wavelet coefficient at each decomposition scale. If the wavelet coefficient is greater than the threshold value, the wavelet coefficient is retained; The thresholded wavelet coefficients are subjected to inverse wavelet transform to obtain the reconstructed signal after denoising.
  • the execution order of the above preprocessing can be selected according to actual needs, which is not limited here.
  • Step2 Feature extraction.
  • the feature extraction is mainly to divide the original data into segments according to the time window (eg, 1 minute segment), perform feature extraction on each segment of data, and finally splicing all the features to form a feature vector.
  • the time window eg, 1 minute segment
  • feature extraction may be performed based on an empirical mode decomposition (empirical mode decomposition, EMD) algorithm.
  • EMD empirical mode decomposition
  • the empirical mode decomposition (EMD) algorithm can be used to extract signal features; firstly, the EMD algorithm can be used to take all the extreme points of the signal, do cubic spline interpolation to obtain the upper and lower envelopes, and calculate their average value; then, remove the average value of the envelope from the original signal, and use the remaining part as the signal to be processed again. Repeat the above operations until the rest of the signal meets the intrinsic mode functions (IMF) screening conditions, and use it as the first IMF component.
  • IMF intrinsic mode functions
  • the remaining part of the original signal after subtracting the IMF component can be used as the original signal again, and the loop operation is performed until the remaining signal is monotonous.
  • the EMD algorithm decomposes the signal into a series of IMF components and the sum of the residual function, and calculates the energy value of each IMF component.
  • feature splicing is performed to form a feature vector.
  • features such as voltage average value, current average value, voltage peak value, current peak value, voltage IMF value of each layer, and current IMF value of each layer can be spliced to form a feature vector.
  • the experience knowledge is firstly summarized in the laboratory environment (data is easy to collect), then the hyperparameters of the SOM are determined using the above experience knowledge, and finally the SOM is trained using the field data of the motor in the park to obtain the training model (ie, the fault diagnosis model).
  • the data of different types of motors in different states are collected and feature extracted, and the distance calculation is performed on the above feature vectors to determine the resolution points that can distinguish different states.
  • the eigenvector of motor 1 is [210 3.7 232 4.1 1.2 2.3 3.4] under normal conditions
  • the eigenvector of motor 2 is [211 4.7234 5.1 2.2 2.3 5.4] under normal conditions
  • the Euclidean distance between the two eigenvectors is 3.
  • Calculate the distance between the eigenvectors in every two different situations by the above method find the minimum value of the distance, for example, the minimum distance is 1, and use the minimum distance value as the resolution point.
  • the exemplary distinguishing point can be understood as a critical point between normal and abnormal.
  • the data can be collected in the actual environment of the industrial park. Considering that the motors will be running in normal working conditions at the beginning, the data can be trained through the neural network to obtain the clustering centers of the multi-motors in the normal state and obtain the fault diagnosis model.
  • Step4 Model usage and self-optimization.
  • the data obtained after feature extraction may be input into the fault diagnosis model to obtain the fault diagnosis result.
  • the data input into the fault diagnosis model deviates from the cluster center of the model, it can be determined that the motor is faulty.
  • the model can also be adjusted with real-time data to achieve self-optimization of the model.
  • the characteristics of the motors in the industrial park will change with the working conditions and other reasons.
  • the parameters of the fault diagnosis model can be further updated every day.
  • the self-organizing learning method is used to complete the establishment of a multi-motor cooperative fault diagnosis model, which solves the problem that it is difficult to realize migration diagnosis through a single model due to the differences of multi-motors in the park, and the traditional physical model method cannot solve this problem.
  • empirical knowledge is acquired in a laboratory environment and applied to the hyperparameter setting of SOM; this idea, compared with traditional artificial intelligence supervised learning methods, greatly reduces the cost of data acquisition; and in the use of SOM technology, it is not purely borrowed, and empirical knowledge is effectively integrated to achieve accurate identification of motor faults.
  • the fault diagnosis model established by this scheme can realize the construction of a model, which can not only deal with the coexistence of different types of motors, but also effectively improve the modeling efficiency for park scenarios; in the initial stage, it is only necessary to conduct basic analysis in the laboratory to obtain empirical knowledge, which greatly reduces the cost of data acquisition, and the above-mentioned data-driven method of integrating empirical models can effectively improve the accuracy of fault diagnosis.
  • a motor fault detection method provided in the embodiment of the present application is introduced. It can be understood that this method is another expression of the motor fault detection solution described above, and the two are combined. This method is proposed based on the motor fault detection scheme described above, and part or all of the content of the method may refer to the description in the motor fault detection scheme above.
  • FIG. 5 is a schematic flowchart of a motor fault detection method provided by an embodiment of the present application. It can be understood that the method can be executed by any device, device, platform, or device cluster that has computing and processing capabilities. As shown in Figure 5, the motor fault detection method includes:
  • N motors of different types may be controlled to operate in different states, so as to obtain the first operating data of the N motors.
  • the first operating data may include one or more of the following: motor number, motor voltage data, motor current data, motor stator vibration data, or motor stator temperature data.
  • the ideal environment may be, but not limited to, a laboratory environment.
  • hyperparameters of the first neural network may be determined according to the acquired first operating data.
  • the first neural network may be, but not limited to, a self-organizing neural network SOM.
  • the eigenvectors corresponding to the first operating data of each motor may be respectively determined first, so as to obtain N first eigenvectors. Then, the first distance between every two eigenvectors among the N first eigenvectors is respectively determined. Finally, based on the smallest first distance among the obtained first distances, the hyperparameters of the first neural network are determined. Exemplarily, but not limited to, feature extraction may be performed on each first operating data by using an EMD algorithm, so as to obtain a feature vector corresponding to the first operating data of each motor.
  • the first distance may be, but not limited to, a Euclidean distance. Exemplarily, the first distance can be understood as empirical knowledge.
  • the operation of the N motors can be controlled in an actual environment, so as to obtain the second operating data of the N motors.
  • the second operating data may include one or more of the following: motor number, motor voltage data, motor current data, motor stator vibration data, or motor stator temperature data.
  • the actual environment may be, but not limited to, the aforementioned environment in the industrial park.
  • the acquired second operating data may be used to train the first neural network to obtain a fault diagnosis model.
  • neural network training algorithms such as Gradient descent, Newton's method, Conjugate gradient, Quasi-Newton method, and Levenberg-Marquardt algorithm can be used to train the first neural network.
  • the fault diagnosis model can be used to perform fault diagnosis on N motors and output a fault diagnosis result.
  • the third operating data of the N motors in the actual environment can be obtained first; then, the N third operating data can be input into the fault diagnosis model to obtain the fault diagnosis result.
  • the third operating data may include one or more of the following: motor number, motor voltage data, motor current data, motor stator vibration data, or motor stator temperature data.
  • the method accumulates empirical knowledge in an ideal environment, determines the hyperparameters of the neural network, and then trains the neural network with the motor operating data obtained in the actual environment to obtain a fault diagnosis model. It realizes the automatic classification of different types of motor samples in the industrial park scene and the fault diagnosis under a small number of fault samples, and realizes the self-optimization of the model with the accumulation of fault samples. In this way, the problems of low accuracy and high data cost of the current traditional motor fault detection method caused by various types of motors and complex working conditions in the industrial park are solved.
  • sequence numbers of the steps in the above embodiments do not mean the order of execution, and the execution order of each process should be determined by its functions and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.
  • the steps in the foregoing embodiments may be selectively executed according to actual conditions, may be partially executed, or may be completely executed, which is not limited here.
  • FIG. 6 is a schematic structural diagram of a motor fault detection device provided by an embodiment of the present application.
  • the motor fault detection device 600 includes: a model building module 601 and a fault diagnosis module 602 .
  • the model building module 601 can be used to obtain the first operating data of N different types of motors in different states in an ideal environment, N ⁇ 2; and according to the first operating data, determine the hyperparameters of the first neural network; and obtain the second operating data of N motors in the actual environment; and use the second operating data to train the first neural network to obtain a fault diagnosis model.
  • the fault diagnosis module 602 can be used to perform fault diagnosis on N motors using the fault diagnosis model, and output fault diagnosis results.
  • the model building module 601 is further configured to: respectively determine the eigenvectors corresponding to the first operating data of each motor to obtain N first eigenvectors; respectively determine the first distance between every two eigenvectors in the N first eigenvectors; determine the hyperparameters of the first neural network based on the smallest first distance among the acquired first distances.
  • the fault diagnosis module 602 is further configured to: obtain third operating data of N motors in an actual environment; input the N third operating data into the fault diagnosis model to obtain a fault diagnosis result.
  • the target operation data includes one or more of the following: motor number, voltage data of the motor, current data of the motor, stator vibration data of the motor, or stator temperature data of the motor; wherein, the target operation data is one of the first operation data, the second operation data or the third operation data.
  • the above-mentioned device is used to execute the method in the above-mentioned embodiment, and the corresponding program modules in the device have similar implementation principles and technical effects to the description in the above-mentioned method.
  • the working process of the device can refer to the corresponding process in the above-mentioned method, and will not be repeated here.
  • FIG. 7 is a schematic structural diagram of a motor fault detection device provided by an embodiment of the present application.
  • the motor fault detection device 700 includes one or more processors 701 and an interface circuit 702.
  • the motor fault detection device 700 may also include a bus 703 . in:
  • the processor 701 may be an integrated circuit chip with signal processing capability. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in the processor 701 or instructions in the form of software.
  • the above-mentioned processor 701 may be a general-purpose processor, a neural network processor (Neural Network Processing Unit, NPU), a digital communicator (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.
  • NPU neural network processor
  • DSP digital communicator
  • ASIC application-specific integrated circuit
  • FPGA field programmable gate array
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • the processor 701 may perform some or all of the steps performed by the electronic device 100 in the above-mentioned embodiments.
  • the processor 701 may perform some or all of the steps performed by the electronic device 200 in the above-mentioned embodiments.
  • the processor 701 may perform some or all of the steps performed by the vehicle 300 in the above-mentioned embodiments.
  • the interface circuit 702 can be used for sending or receiving data, instructions or information.
  • the processor 701 can use the data, instructions or other information received by the interface circuit 702 to process, and can send the processing completion information through the interface circuit 702 .
  • the interface circuit 702 can be used to receive messages sent by the vehicle 300 or the electronic device 200 , or send messages to the vehicle 300 or the electronic device 200 .
  • the interface circuit 702 can be used to receive messages sent by the vehicle 300 or the electronic device 70 , or send messages to the vehicle 300 or the electronic device 70 .
  • the interface circuit 702 can be used to receive messages sent by the electronic device 100 or the electronic device 200 , or send messages to the electronic device 100 or the electronic device 200 .
  • the motor fault detection device 700 further includes a memory, which may include a read-only memory and a random access memory, and provides operation instructions and data to the processor.
  • a portion of the memory may also include non-volatile random access memory (NVRAM).
  • NVRAM non-volatile random access memory
  • the memory may be coupled with the processor 701 .
  • the memory stores executable software modules or data structures
  • the processor 701 can execute corresponding operations by calling operation instructions stored in the memory (the operation instructions can be stored in the operating system).
  • the interface circuit 702 may be used to output the execution result of the processor 701 .
  • processor 701 and the interface circuit 702 can be realized by hardware design, software design, or a combination of software and hardware, which is not limited here.
  • the processor in the embodiments of the present application may be a central processing unit (central processing unit, CPU), and may also be other general-purpose processors, digital signal processors (digital signal processors, DSP), application specific integrated circuits (application specific integrated circuits, ASICs), field programmable gate arrays (field programmable gate arrays, FPGAs) or other programmable logic devices, transistor logic devices, hardware components or other any combination.
  • a general-purpose processor can be a microprocessor, or any conventional processor.
  • the method steps in the embodiments of the present application may be implemented by means of hardware, or may be implemented by means of a processor executing software instructions.
  • the software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (random access memory, RAM), flash memory, read-only memory (read-only memory, ROM), programmable read-only memory (programmable rom, PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically erasable programmable read-only memory (electrically EPROM, EEPROM), registers, hard disk, mobile hard disk, CD-ROM, or any other form of storage medium known in the art.
  • An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium.
  • the storage medium may also be a component of the processor.
  • the processor and storage medium can be located in the ASIC.
  • all or part of them may be implemented by software, hardware, firmware or any combination thereof.
  • software When implemented using software, it may be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present application will be generated in whole or in part.
  • the computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in or transmitted via a computer-readable storage medium.
  • the computer instructions may be transmitted from one web site, computer, server, or data center to another web site, computer, server, or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center integrated with one or more available media.
  • the available medium may be a magnetic medium (such as a floppy disk, a hard disk, or a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (solid state disk, SSD)), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Control Of Electric Motors In General (AREA)

Abstract

Un procédé de détection de défaillance de moteur consistant à : acquérir des premières données de fonctionnement de N types différents de moteurs dans différents états dans un environnement idéal ; déterminer un hyperparamètre d'un premier réseau de neurones en fonction des premières données de fonctionnement ; acquérir des secondes données de fonctionnement des N moteurs dans un environnement réel ; entraîner le premier réseau de neurones à l'aide des secondes données de fonctionnement, pour obtenir un modèle de diagnostic de défaillance ; effectuer un diagnostic de défaillance sur les N moteurs à l'aide du modèle de diagnostic de défaillance, et délivrer un résultat de diagnostic de défaillance. De cette manière, l'hyperparamètre du réseau de neurones est déterminé au moyen de connaissances empiriques formées dans un environnement idéal, puis le réseau de neurones est entraîné à l'aide de données de fonctionnement de moteur obtenues dans un environnement réel, pour obtenir le modèle de diagnostic de défaillance, ce qui permet de réaliser une classification automatique de différents types d'échantillons de moteur dans un scénario de parc industriel et un diagnostic de défaillance à l'aide d'un petit nombre d'échantillons de défaillance, et de résoudre les problèmes associés à la faible précision de détection d'une défaillance d'un moteur et aux coûts élevés des données en raison des divers types de moteur et des conditions de travail complexes dans un parc industriel.
PCT/CN2022/143121 2022-01-18 2022-12-29 Procédé et appareil de détection de défaillance de moteur WO2023138337A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210053329.9A CN116500411A (zh) 2022-01-18 2022-01-18 一种电机故障检测方法及装置
CN202210053329.9 2022-01-18

Publications (1)

Publication Number Publication Date
WO2023138337A1 true WO2023138337A1 (fr) 2023-07-27

Family

ID=87321784

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/143121 WO2023138337A1 (fr) 2022-01-18 2022-12-29 Procédé et appareil de détection de défaillance de moteur

Country Status (2)

Country Link
CN (1) CN116500411A (fr)
WO (1) WO2023138337A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116800588A (zh) * 2023-08-28 2023-09-22 深圳市华曦达科技股份有限公司 网通产品网络优化方法和装置
CN116980279A (zh) * 2023-09-25 2023-10-31 之江实验室 一种可编程网元设备的故障诊断系统及故障诊断方法
CN117033950A (zh) * 2023-10-08 2023-11-10 华中科技大学 一种gis隔离开关机械故障在线诊断方法及装置
CN117170304A (zh) * 2023-11-03 2023-12-05 傲拓科技股份有限公司 一种基于工业物联网的plc远程监测控制方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102135593A (zh) * 2010-12-28 2011-07-27 太原理工大学 大电机绝缘状态在线诊断评估方法
DE102014212572A1 (de) * 2014-06-30 2015-12-31 Elmos Semiconductor Aktiengesellschaft Verfahren zur Erlangung eines Hinweises, insbesondere Anfangshinweises auf eine mögliche fehlerhafte Lastbedingung eines mehrphasigen Elektromotors
CN110531266A (zh) * 2019-10-14 2019-12-03 国网山东省电力公司电力科学研究院 一种同步电机励磁绕组匝间短路故障预警方法
CN112052796A (zh) * 2020-09-07 2020-12-08 电子科技大学 一种基于深度学习的永磁同步电机故障诊断方法
CN113762486A (zh) * 2021-11-11 2021-12-07 中国南方电网有限责任公司超高压输电公司广州局 换流阀故障诊断模型的构建方法、装置和计算机设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102135593A (zh) * 2010-12-28 2011-07-27 太原理工大学 大电机绝缘状态在线诊断评估方法
DE102014212572A1 (de) * 2014-06-30 2015-12-31 Elmos Semiconductor Aktiengesellschaft Verfahren zur Erlangung eines Hinweises, insbesondere Anfangshinweises auf eine mögliche fehlerhafte Lastbedingung eines mehrphasigen Elektromotors
CN110531266A (zh) * 2019-10-14 2019-12-03 国网山东省电力公司电力科学研究院 一种同步电机励磁绕组匝间短路故障预警方法
CN112052796A (zh) * 2020-09-07 2020-12-08 电子科技大学 一种基于深度学习的永磁同步电机故障诊断方法
CN113762486A (zh) * 2021-11-11 2021-12-07 中国南方电网有限责任公司超高压输电公司广州局 换流阀故障诊断模型的构建方法、装置和计算机设备

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116800588A (zh) * 2023-08-28 2023-09-22 深圳市华曦达科技股份有限公司 网通产品网络优化方法和装置
CN116800588B (zh) * 2023-08-28 2023-12-22 深圳市华曦达科技股份有限公司 网通产品网络优化方法和装置
CN116980279A (zh) * 2023-09-25 2023-10-31 之江实验室 一种可编程网元设备的故障诊断系统及故障诊断方法
CN116980279B (zh) * 2023-09-25 2023-12-12 之江实验室 一种可编程网元设备的故障诊断系统及故障诊断方法
CN117033950A (zh) * 2023-10-08 2023-11-10 华中科技大学 一种gis隔离开关机械故障在线诊断方法及装置
CN117033950B (zh) * 2023-10-08 2023-12-26 华中科技大学 一种gis隔离开关机械故障在线诊断方法及装置
CN117170304A (zh) * 2023-11-03 2023-12-05 傲拓科技股份有限公司 一种基于工业物联网的plc远程监测控制方法及系统
CN117170304B (zh) * 2023-11-03 2024-01-05 傲拓科技股份有限公司 一种基于工业物联网的plc远程监测控制方法及系统

Also Published As

Publication number Publication date
CN116500411A (zh) 2023-07-28

Similar Documents

Publication Publication Date Title
WO2023138337A1 (fr) Procédé et appareil de détection de défaillance de moteur
WO2016101690A1 (fr) Procédé de nettoyage de données de surveillance d'état basé sur une analyse de séquence temporelle de dispositif de transmission et de transformation d'énergie
CN109297689B (zh) 一种引入权重因子的大型水力机械智能诊断方法
WO2019080367A1 (fr) Procédé d'évaluation de l'état de santé d'un dispositif mécanique
WO2023093177A1 (fr) Procédé et appareil de diagnostic de défaillance de dispositif, dispositif électronique et support de stockage
Vodenčarević et al. Identifying behavior models for process plants
CN111431819A (zh) 一种基于序列化的协议流特征的网络流量分类方法和装置
CN114266197A (zh) 一种水电站设备故障诊断方法
Vamsi et al. Random forest based real time fault monitoring system for industries
CN114993669A (zh) 多传感器信息融合的传动系统故障诊断方法及系统
CN111862951A (zh) 语音端点检测方法及装置、存储介质、电子设备
CN112565187A (zh) 基于逻辑回归的电网攻击检测方法、系统、设备及介质
CN115800272A (zh) 基于拓扑识别的电网故障分析方法、系统、终端及介质
CN110188837A (zh) 一种基于模糊神经的mvb网络故障诊断方法
CN117076955A (zh) 一种高压变频器的故障检测方法及系统
CN116992376A (zh) 用于检测电网中的异常的装置、系统和方法
CN110187206B (zh) 一种复杂工况下非高斯过程的悬浮系统的故障检测方法
CN112949714A (zh) 一种基于随机森林的故障可能性预估方法
CN113012107B (zh) 电网缺陷检测方法及系统
CN113111585A (zh) 一种智能机柜故障预测方法、系统及智能机柜
CN116467592A (zh) 一种基于深度学习的生产设备故障智能监测方法及系统
CN115959549A (zh) 基于数字孪生的扶梯故障诊断方法
CN114157486B (zh) 通信流量数据异常检测方法、装置、电子设备及存储介质
CN115276228A (zh) 一种功率数据同步上送的方法、系统、存储和计算设备
CN114881162A (zh) 计量自动化主站的故障预测方法、装置、设备及介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22921757

Country of ref document: EP

Kind code of ref document: A1