Asynchronous motor short circuit diagnosis method, device and equipment and storage medium

Abstract

The invention discloses an asynchronous motor short circuit diagnosis method, device and equipment and a storage medium, and relates to the technical field of fault diagnosis. The method comprises the following steps: obtaining stator three-phase current data and removing a direct current component in the stator three-phase current data to obtain target three-phase current data; performing coordinate conversion on the target three-phase current data to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and generating a vector diagram according to the current components; extracting image contour features in the vector diagram, and determining a vector ratio of images in the vector diagram according to the image contour features; and inputting the vector ratio of the image in the vector diagram into a pre-trained fault detection model, and determining the fault severity value of the asynchronous motor corresponding to the vector ratio of the image in the vector diagram according to the relationship between the vector ratio in the fault detection model and the fault severity value. According to the embodiment of the invention, the severity of the short-circuit fault of the asynchronous motor can be accurately quantified, and the quantitative diagnosis of the short-circuit fault of the asynchronous motor is realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of fault diagnosis, and particularly relates to a short-circuit diagnosis method, device and equipment of an asynchronous motor and a storage medium. BACKGROUND

[0002] As an important power equipment, asynchronous motors are widely used in various machines and automation systems. However, as the running time increases, the motor may have different types of faults, among which the inter-turn short-circuit fault is one of the most common and most influential faults on the safety and operating efficiency of the equipment. The inter-turn short-circuit fault not only causes the performance of the motor to decline, but also may cause serious safety hazards. Therefore, it is particularly important to diagnose the fault in a timely and effective manner.

[0003] In the prior art, when diagnosing the short-circuit fault of an asynchronous motor, the time sequence characteristics are often converted into an image form, the fault diagnosis of the asynchronous motor is realized by comparing the differences between the normal image features and the short-circuit fault image features, and the fault severity is determined by the difference size. However, there is not a simple linear relationship between the image features and the fault severity, and it is difficult to measure the severity of the fault by using intuitive numerical indicators. That is, the prior art can only reflect the qualitative information of the fault of the asynchronous motor, and cannot provide an accurate numerical mapping of the severity of the short-circuit fault. Therefore, the prior art is difficult to quantify the severity of the short-circuit fault of the asynchronous motor. SUMMARY

[0004] The embodiments of the application provide a short-circuit diagnosis method, device, equipment and storage medium of an asynchronous motor to solve the problem that the prior art is difficult to quantify the severity of the short-circuit fault of the asynchronous motor.

[0005] In a first aspect, the embodiments of the application provide a short-circuit diagnosis method of an asynchronous motor, which comprises the following steps.

[0006] Obtaining stator three-phase current data of the asynchronous motor;

[0007] Removing a direct current component in the stator three-phase current data to obtain target three-phase current data;

[0008] Performing coordinate conversion on the target three-phase current data to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and generating a vector diagram according to the current components, the two-phase rotating coordinate system being synchronous with the rotor of the asynchronous motor;

[0009] Extracting image contour features in the vector diagram, and determining a vector ratio of the image in the vector diagram according to the image contour features;

[0010] The vector ratio of the image in the vector diagram is input into the pre-trained fault detection model, and the fault severity value corresponding to the vector ratio of the image in the vector diagram is determined through the relationship between the vector ratio and the fault severity value in the fault detection model.

[0011] In a second aspect, the embodiments of the present application provide a device for short circuit diagnosis of an asynchronous motor, which comprises:

[0012] An acquisition module is configured to acquire stator three-phase current data of the asynchronous motor.

[0013] A removal module is configured to remove a direct current component in the stator three-phase current data to obtain target three-phase current data.

[0014] A conversion module is configured to perform coordinate conversion on the target three-phase current data to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and generate a vector diagram according to the current components, the two-phase rotating coordinate system being synchronous with a rotor of the asynchronous motor.

[0015] An extraction module is configured to extract an image contour feature in the vector diagram, and determine a vector ratio of the image in the vector diagram according to the image contour feature.

[0016] A determination module is configured to input the vector ratio of the image in the vector diagram into a pre-trained fault detection model, and determine a fault severity value of the asynchronous motor corresponding to the vector ratio of the image in the vector diagram through a relationship between the vector ratio and the fault severity value in the fault detection model.

[0017] In a third aspect, the embodiments of the present application provide a terminal device, which comprises a processor and a memory storing computer program instructions; and the processor implements the method for short circuit diagnosis of the asynchronous motor according to the first aspect when executing the computer program instructions.

[0018] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores computer program instructions; and the computer program instructions are executed by a processor to implement the method for short circuit diagnosis of the asynchronous motor according to the first aspect.

[0019] In a fifth aspect, the embodiments of the present application provide a computer program product, and instructions in the computer program product are executed by a processor of an electronic device to enable the electronic device to perform the method for short circuit diagnosis of the asynchronous motor according to the first aspect.

[0020] An embodiment of the present application provides a method for diagnosing asynchronous motor short circuits. The method determines an image vector ratio from a vector diagram, quantitatively reflecting changes in the current vector, and obtains a corresponding fault severity value based on the numerical mapping relationship between the vector ratio and the fault severity in a fault detection model, thereby accurately quantifying the fault severity of the asynchronous motor short circuit. The vector diagram is obtained using the equivalent DC current component in a rotating coordinate system, which simplifies the mathematical model of the motor current and reduces data processing complexity. Furthermore, the equivalent DC current component is generated using the AC component rather than the complete stator three-phase current data, making the fault-related frequency components more prominent. Therefore, the embodiment of the present application enables quantitative diagnosis of asynchronous motor short circuit faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1 1 is a flow chart of a method for diagnosing short circuit of an asynchronous motor provided in an embodiment of the present application;

[0023] Figure 2 is a vector diagram of a healthy motor provided by an embodiment of the present application;

[0024] Figure 3 This is a vector diagram of a motor with a short-circuit ratio of 12% provided in an embodiment of the present application;

[0025] Figure 4 This is a vector diagram of a motor with a short-circuit ratio of 17% provided in an embodiment of the present application;

[0026] Figure 5 is a schematic diagram of a confusion matrix of a fault classification result provided in an embodiment of the present application;

[0027] Figure 6 Schematic diagram of the structure of the device for diagnosing short circuit of an asynchronous motor provided in an embodiment of the present application;

[0028] Figure 7 It is a structural diagram of the terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] The features and exemplary embodiments of the various aspects of the present application will be described in detail below with reference to the drawings. For the purpose of clarity, technical terms may be defined herein before being consistently used throughout the specification. Embodiments described herein are intended to explain the principles of the application and to enable others skilled in the art to most readily practice the application. The application will be described with regard to the following figures and detailed description.

[0030] It should be noted that the relational terms herein, such as first and second, and the like, are used solely to distinguish one from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0031] An asynchronous motor is an alternating current motor that works based on the law of electromagnetic induction, mainly composed of a stator and a rotor. The stator is the fixed part of the motor, which includes stator windings and a stator core where the stator windings are placed. The rotor is the rotating part of the motor, which includes rotor conductors and a rotor core where the rotor conductors are placed. When the stator windings of the asynchronous motor are connected to three-phase alternating current, a rotating magnetic field is generated. This rotating magnetic field cuts the rotor conductors, inducing an electromotive force and current in the rotor conductors. The magnetic field generated by these induced currents interacts with the rotating magnetic field generated by the stator, producing an electromagnetic torque that drives the rotor to rotate. Because the rotor speed always has a certain speed difference with the speed of the stator rotating magnetic field, it is called an asynchronous motor.

[0032] In modern industry, asynchronous motors are widely used as important power equipment in various types of machinery and automation systems. However, as the running time increases, the motor may experience different types of faults, among which the inter-turn short circuit fault is one of the most common and most influential faults on the safety and efficiency of the equipment. Inter-turn short circuit faults not only cause the performance of the motor to decline, but also can cause serious safety hazards. Therefore, it is particularly important to diagnose this fault in a timely and effective manner.

[0033] Most of the existing asynchronous motor inter-turn short-circuit fault diagnosis methods based on stator current signal cannot establish a numerical mapping relationship between fault severity and fault feature. In order to overcome this difficulty, the existing method proposes a graphical feature of a planar figure drawn by an equation as a fault feature. Although this feature is defined and proved to be positively correlated with fault severity, the numerical mapping relationship between them has not been revealed, and further research is needed to achieve quantitative diagnosis. Therefore, based on the above analysis, it can be seen that how to extract a fault feature that can be used for quantitative diagnosis from the stator current signal is still a difficult problem to be solved.

[0034] In order to solve the prior art, the embodiment of the present application provides an asynchronous motor short-circuit diagnosis method, which determines an image vector ratio from a vector diagram, quantifies the change of the current vector, and obtains the corresponding fault severity value according to the numerical mapping relationship between the vector ratio and the fault severity in the fault detection model, thereby accurately quantifying the fault severity of the asynchronous motor short-circuit. The vector diagram is obtained by an equivalent direct current component in a rotating coordinate system, which simplifies the mathematical model of the motor current and reduces the data processing complexity. Moreover, the equivalent direct current component is generated by alternating current components rather than complete stator three-phase current data, so that the frequency components related to the fault are more prominent. Therefore, the embodiment of the present application realizes quantitative diagnosis of the asynchronous motor short-circuit fault.

[0035] The asynchronous motor short-circuit diagnosis method provided by the embodiment of the present application will be described below with reference to the accompanying drawings.

[0036] Figure 1 A flowchart of the asynchronous motor short-circuit diagnosis method provided by an embodiment of the present application is shown. As shown in Figure 1 The method can include the following steps: S101 to S105.

[0037] S101, obtaining stator three-phase current data of an asynchronous motor.

[0038] The stator three-phase current is an alternating current flowing into three windings of the stator of the asynchronous motor, and the size and direction change with time. The three sets of alternating currents have the same frequency and amplitude, and are sequentially phase-shifted by 120 degrees, and are usually marked as phase A, phase B and phase C.

[0039] The stator three-phase current is a continuously changing signal during the operation of the asynchronous motor, which can reflect the running state of the asynchronous motor in real time and intuitively. When the asynchronous motor has a short-circuit fault, the stator three-phase current will change abnormally. By monitoring the three-phase current, these abnormal changes can be directly captured, so that the occurrence of the fault can be quickly and accurately judged.

[0040] In some embodiments, the stator winding of the asynchronous motor is connected with a current sensor, and the current sensor is used to sample the three-phase current of the stator. For example, a current clamp can be clamped on the wire of the stator winding of the asynchronous motor, and the three-phase current of the stator is sampled by using the current clamp.

[0041] In some embodiments, when the three-phase current of the stator is sampled, the output end of the current sensor is connected to the input channel of the data acquisition card, and the corresponding sampling frequency and sampling unit are set by the driving software of the data acquisition card according to different strategy requirements.

[0042] In some embodiments, when the sampling frequency is set, the sampling frequency is higher than the Nyquist frequency determined by the highest frequency of the signal, that is, the sampling frequency needs to be higher than twice the highest frequency of the current signal.

[0043] In an ideal state, when the sampling frequency is equal to twice the highest frequency, the signal can be accurately reconstructed, and signal aliasing can be prevented. However, in actual application, the signal often contains noise and unexpected high-frequency components, so the highest frequency of the signal cannot always be accurately determined. Therefore, the sampling frequency is selected to be higher than twice the highest frequency, which provides an additional margin to further reduce the risk of signal aliasing.

[0044] S102, removing the direct current component in the three-phase current data of the stator to obtain target three-phase current data.

[0045] In some embodiments, a high-pass filter is connected between the asynchronous motor and the current sensor, or between the current sensor and the data acquisition card. By setting the cutoff frequency of the high-pass filter, the direct current component in the three-phase current data is removed, and the alternating current signal is retained while the direct current component is filtered out.

[0046] In some embodiments, after sampling the three-phase current of the stator of the asynchronous motor, the preset algorithm in the digital high-pass filter is used to filter the direct current signal in the digital domain, and the direct current component in the three-phase current data of the stator is filtered out.

[0047] In an ideal alternating current signal, the direct current component of the current should be zero, but in the actual operation process of the asynchronous motor, due to power interference, asymmetry of the internal structure of the motor, or load imbalance, etc., the obtained three-phase current data of the stator often contains a direct current component. The existence of the direct current component will make the collected current data deviate from the actual alternating current signal value, and may be confused with other low-frequency signal components, thereby interfering with the identification of the fault characteristics in the alternating current signal. Therefore, the target three-phase current data obtained by removing the direct current component is crucial for analyzing the operating state of the motor.

[0048] The target three-phase current data obtained after the DC component removal processing can reduce the proportion of the DC component in the signal and avoid confusion with the low-frequency signal in the alternating current signal, so that the characteristics of the alternating current signal and the fault characteristics are more prominent, thereby providing accurate input for the subsequent analysis steps.

[0049] In S103, the target three-phase current data is subjected to coordinate conversion to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and a vector diagram is generated according to the current components, and the two-phase rotating coordinate system rotates synchronously with the rotor of the asynchronous motor.

[0050] The two-phase rotating coordinate system is a coordinate system that rotates synchronously with the rotor of the motor, and is usually composed of a direct axis (d-axis) that is the same as the direction of the rotor magnetic field and a quadrature axis (q-axis) that is perpendicular to the direct axis, and the original coordinate system of the target three-phase current data is a stationary three-phase coordinate system.

[0051] Converting the target three-phase current data into the two-phase rotating coordinate system can obtain the contribution of each current component of the target three-phase current data to the torque, and the current components on the direct axis and the quadrature axis correspond to the current components in the direction of the rotor magnetic field and perpendicular to the direction of the magnetic field, respectively. Since the coordinate axes rotate synchronously with the rotor, the direct-axis and quadrature-axis current components can be regarded as direct-current signals, thereby simplifying the complex alternating-current signal into a direct-current signal that is easy to analyze and reducing the computational complexity.

[0052] In some embodiments, the present application uses Park transformation to project the target three-phase current data onto the direct axis and the quadrature axis of the two-phase rotating coordinate system that rotates with the rotor.

[0053] In S104, image contour features in the vector diagram are extracted, and a vector ratio of the image in the vector diagram is determined according to the image contour features.

[0054] The different vectors in the vector diagram reflect the size and direction of different current components, and the vector ratio refers to the proportional relationship between different vectors in the vector diagram, i.e., the relative relationship between different current components.

[0055] In some embodiments, the present application uses the related functions of the machine vision function library opencv in python to extract the image contour features.

[0056] In some embodiments, extracting the image contour features in the vector diagram can include:

[0057] Converting the vector diagram into a grayscale image and using the Canny edge detection algorithm to highlight the contour edges in the grayscale image;

[0058] The image contour in the gray image is found using the cv2.findContours function in the opencv library, and the image contour feature is obtained according to the image contour. The image contour is a closed shape or curve, which can be represented by a set of points.

[0059] In some embodiments, determining the vector ratio of the image in the vector diagram can be to determine the length ratio between vectors or the included angle between vectors in the vector diagram.

[0060] By extracting the image contour feature of the vector diagram and determining the vector ratio, the complex current data can be represented in an intuitive graphical form, making the running state of the motor more easily understood and analyzed.

[0061] S105, input the vector ratio of the image in the vector diagram into the pre-trained fault detection model, and determine the fault severity value of the asynchronous motor corresponding to the vector ratio of the image in the vector diagram through the relationship between the vector ratio in the fault detection model and the fault severity value.

[0062] The contour feature of the vector diagram and the corresponding vector ratio are relatively sensitive to the change of the running state of the motor. For example, when the motor has a short circuit or an open circuit fault, the contour feature and the vector ratio of the vector diagram will change significantly. Based on the change value of the vector ratio, the quantitative analysis of the fault type and the severity can be quickly and accurately realized.

[0063] The embodiment of the present application determines the image vector ratio from the vector diagram, which quantifies the change of the current vector, and obtains the corresponding fault severity value according to the numerical mapping relationship between the vector ratio and the fault severity in the fault detection model, thereby accurately quantifying the fault severity of the asynchronous motor short circuit. The vector diagram is obtained by rotating the equivalent direct current component in the coordinate system, which simplifies the mathematical model of the motor current and reduces the data processing complexity. Moreover, the equivalent direct current component is generated by the alternating component rather than the complete stator three-phase current data, so that the frequency component related to the fault is more prominent. Therefore, the embodiment of the present application realizes the quantitative diagnosis of the asynchronous motor short circuit fault.

[0064] In some embodiments, removing the direct current component from the stator three-phase current data to obtain the target three-phase current data can include:

[0065] Calculating the average value of each phase current data in the three-phase current data;

[0066] Calculating the difference between each phase current data and the average value of each phase current data to obtain the target three-phase current data.

[0067] Since the direct current component is constant, the direct current component is usually represented as the average value of the current signal. By calculating the average value of each phase current, the direct current component in the phase current can be obtained. Subtracting the average value from the original current data can filter out the direct current component and obtain the target three-phase current data containing only alternating current components.

[0068] In some embodiments, the average value of each phase current data in the three-phase current data is calculated, and the formula can include (1)-(3):

[0069]

[0070] wherein I ma is the average value of the a-phase current data, t is the sampling time, t1 is the start time of sampling, t2 is the end time of sampling, I a is the sampled a-phase initial current data, I mb is the average value of the b-phase current data, I b is the sampled b-phase initial current data, I mc is the average value of the c-phase current data, I c is the sampled c-phase initial current data.

[0071] In some embodiments, the difference between each phase current data and the average value of each phase current data is calculated to obtain the target three-phase current data, and the formula can include (4)-(6):

[0072] I A = I a -I ma (4)

[0073] I B = I b -I mb (5)

[0074] I C = I c -I mc (6)

[0075] wherein I A is the a-phase current data after removing the direct current component, I B is the b-phase current data after removing the direct current component, and I C is the c-phase current data after removing the direct current component.

[0076] In some embodiments, the target three-phase current data is subjected to coordinate conversion to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and a vector diagram is generated according to the current components, which can include:

[0077] The target three-phase current data is multiplied by the preset transformation matrix to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system.

[0078] A phase angle and a modulus between the two current components are calculated, and a vector diagram is generated according to the phase angle and the modulus, the phase angle being a polar angle of the vector diagram and the modulus being a polar radius of the vector diagram.

[0079] The three-phase current data is converted into two current components in a two-phase rotating coordinate system, which reduces the data dimension and makes the data easier to process and analyze, and a vector diagram is generated through the phase angle and the modulus, thereby providing quantitative analysis data for motor performance and fault analysis.

[0080] In some embodiments, the preset transformation matrix of the present application is shown in formula (7):

[0081]

[0082] wherein C is the preset transformation matrix.

[0083] The target three-phase current data is multiplied by the preset transformation matrix to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, as shown in formulas (8)-(10):

[0084]

[0085] wherein I d is a component of the target three-phase current data on the d-axis, I q is a component of the target three-phase current data on the q-axis, and I 0 is a zero sequence component. α β

[0086] In some embodiments, the calculation formula of the phase angle is shown in formula (11):

[0087]

[0088] wherein θ represents a phase angle between the d-axis and q-axis currents.

[0089] In some embodiments, the modulus formula between the two current components is shown in formula (12):

[0090]

[0091] wherein M is a modulus of the d-axis and q-axis currents.

[0092] In some embodiments, generating the vector diagram according to the phase angle and the modulus can include:

[0093] ​​A polar coordinate system is established with the d-axis and q-axis as coordinate axes, each current component corresponds to a coordinate value, and each point in the vector diagram represents the current state at a specific time, and the length and direction of the vector represent the current size and phase. By analyzing the vector diagram, the running state of the motor can be intuitively understood, such as the magnetic field strength and the torque size.

[0094] In some embodiments, the application can determine the running state and fault condition of the motor in advance through the shape and trend of the vector diagram. When the asynchronous motor is running normally, the vector diagram presents a stable rotating circle or ellipse, and if the vector length abnormally increases or the direction suddenly changes, it may indicate that the motor has a fault. For example, abnormal d-axis component may be related to excitation system failure, and abnormal q-axis component may indicate torque fluctuation or mechanical failure.

[0095] In some embodiments, the image contour features include the maximum distance between two points on the image contour of the vector diagram, the zeroth moment of the image contour, and the first moment of the image contour; determining the vector ratio of the image in the vector diagram according to the image contour features can include:

[0096] According to the relationship between the zeroth moment, the first moment and the centroid, the centroid of the image contour corresponding to the zeroth moment and the first moment of the image contour is determined;

[0097] The target point closest to the centroid of the image contour on the image contour is obtained, and the distance between the centroid of the image contour and the target point is calculated to obtain the minimum inscribed circle radius;

[0098] According to the relationship between the maximum distance between two points on the image contour, the minimum inscribed circle radius and the vector ratio, the vector ratio of the image in the vector diagram is determined.

[0099] The embodiments of the application consider various image contour features such as the maximum distance between two points on the image contour of the vector diagram, the zeroth moment and the first moment of the image contour, comprehensively describe the image features of the vector diagram from different angles, and can more accurately reflect the running state of the motor.

[0100] In some embodiments, the application uses the maximum distance between two points on the image contour of the Euclidean distance vector diagram, and the calculation formula is shown in formula (13):

[0101]

[0102] Where x and y are the horizontal and vertical coordinates of the points on the image contour, and d is the Euclidean distance between the two points.

[0103] In some embodiments, according to the relationship between the zeroth moment, the first moment and the centroid, the centroid of the image contour corresponding to the zeroth moment and the first moment of the image contour is determined, and the calculation formula is shown in formula (14)-(15):

[0104]

[0105] wherein cx is the horizontal coordinate of the centroid, cy is the vertical coordinate of the centroid, M 00 is the zeroth moment of the image, M 01 , M 10 is the first moment of the image.

[0106] In some embodiments, the zeroth moment and the first moment of the image are obtained by a cv2.moments() function of an opencv library.

[0107] In some embodiments, determining the vector ratio of the image in the vector diagram according to the relationship between the maximum distance between two points on the image contour, the minimum incircle radius and the vector ratio can include: obtaining the vector ratio PRD (Park Distance Ratio) by dividing the maximum distance between two points on the image contour by the minimum incircle radius.

[0108] In some embodiments, before inputting the vector ratio of the image in the vector diagram into the pre-trained fault detection model and determining the fault severity value of the asynchronous motor corresponding to the vector ratio of the image in the vector diagram through the relationship between the vector ratio and the fault severity value in the fault detection model, the method can further include:

[0109] subtracting a preset intercept from the vector ratio and dividing by a preset slope to obtain the fault severity value;

[0110] establishing the relationship between the vector ratio and the fault severity value.

[0111] The embodiment of the present application uses a linear function as the relationship model between the vector ratio and the fault severity value, and the vector ratio and the fault severity value are positively correlated. Using a linear model function is easy to analyze and understand and will not have the problem of inconsistent monotonicity in multiple intervals of a multiple function.

[0112] In some embodiments, the relationship between the vector ratio and the fault severity value is as shown in formula (16), which can include:

[0113] PRD = m (FD) + b (16)

[0114] wherein m is a preset slope and b is a preset intercept.

[0115] In some embodiments, before inputting the vector ratio of the image in the vector diagram into the pre-trained fault detection model and determining the fault severity value of the asynchronous motor corresponding to the vector ratio of the image in the vector diagram through the relationship between the vector ratio and the fault severity value in the fault detection model, the method can further include:

[0116] obtaining a first training sample, the first training sample including a plurality of vector ratio samples and corresponding fault severity sample values;

[0117] inputting the vector ratio sample into the initial fault detection model, determining a predicted fault severity value corresponding to the vector ratio sample through a relationship between the vector ratio and the fault severity value in the initial fault detection model;

[0118] determining a loss function value of the initial fault detection model according to the predicted fault severity value and the fault severity sample value, and training model parameters of the initial fault detection model according to the loss function value until a training stop condition is met, to obtain a pre-trained fault detection model.

[0119] The embodiments of the application train the initial fault detection model by using the first training sample containing a plurality of vector ratio samples and corresponding fault severity sample values, so that the model can learn the mapping relationship between the vector ratio and the fault severity, improve the accuracy and reliability of the fault detection model, and better adapt to the actual motor fault diagnosis demand.

[0120] In some embodiments, the step of training the initial fault detection model to obtain the pre-trained fault detection model can include:

[0121] inputting a point pair composed of 10 groups to less than 20 groups of vector ratio samples and corresponding fault severity sample values into the initial fault detection model, the point pair can be (x1, y1), (x2, y2), …, (xn, yn), where x is the vector ratio sample, y is the corresponding fault severity sample value, and the predicted fault severity value of the output initial fault detection model is shown in formula (17):

[0122] y′ i = mx i +b (17)

[0123] where y′ i is the predicted fault severity value corresponding to the i-th vector ratio sample, x i is the i-th vector ratio sample.

[0124] Determine the residual corresponding to each vector ratio sample according to formula (18):

[0125] σ i =y i -y′ i =y i -(mx i +b) (18)

[0126] where σ i is the residual value, and y i is the fault severity sample value corresponding to the i-th vector ratio sample.

[0127] A sum of squares of all errors is added according to formula (19) to construct a sum of squares of errors S, and a loss function value of the initial fault detection model is obtained:

[0128]

[0129] Wherein, S is the loss function value.

[0130] Partial derivatives of the preset slope and the preset intercept of the initial fault detection model are calculated based on the loss function value, and the partial derivative calculation formulas of the preset slope and the preset intercept are shown in formulas (20)-(21):

[0131]

[0132]

[0133] Wherein, is the partial derivative of the preset slope, is the partial derivative of the preset intercept.

[0134] According to the loss function value, the values of m and b corresponding to the initial fault model are found, so that S is minimized.

[0135] The partial derivatives of the preset slope and the preset intercept are set to 0, that is, the minimum value of the loss function value is found, the preset slope and the preset intercept when the loss function value is minimum are obtained, and the preset slope and the preset intercept of the preset initial fault detection model are updated to obtain the pre-trained fault detection model, and the calculation formulas are shown in formulas (22)-(23):

[0136]

[0137] Wherein, m is the preset slope, and b is the preset intercept.

[0138] In some embodiments, after extracting the image contour features in the vector diagram and determining the vector ratio of the image in the vector diagram according to the image contour features, the method can further include:

[0139] The vector ratio is input into a plurality of sub-models of the pre-trained classification model, and a plurality of prediction classification results corresponding to the plurality of sub-models are determined according to the vector ratio and preset model parameters of the sub-models;

[0140] The target classification result is determined according to the number of different prediction classification results in the plurality of prediction classification results.

[0141] The embodiment of the application inputs the vector ratio into a plurality of sub-models of a pre-trained classification model after determining the vector ratio, further classifies the fault through the classification model, and determines a target classification result according to a majority prediction result. The multi-model collaborative classification manner can give full play to the advantages of each sub-model, improve the accuracy and robustness of fault classification, and reduce diagnostic errors caused by possible misjudgments of a single model.

[0142] In some embodiments, the classification model is a random forest model.

[0143] In some embodiments, before the vector ratio is input into the plurality of sub-models of the pre-trained classification model, the method can further include:

[0144] obtaining a second training sample, the second training sample including a plurality of vector ratio samples and corresponding fault classification labels;

[0145] randomly selecting a preset number of vector ratio samples multiple times and inputting the vector ratio samples into a plurality of initial sub-models of an initial classification model, for each initial sub-model, performing multiple binary classifications on the input vector ratio samples by selecting a plurality of division thresholds, until the fault classification labels of the vector ratio samples in each classification are the same, and obtaining a plurality of sub-models including model parameters of the plurality of division thresholds.

[0146] In the embodiment of the application, when training the sub-models of the classification model, the sub-models with multiple division threshold model parameters are obtained by multiple binary classifications until the fault classification labels of the samples in the classification are the same. This training manner can make the sub-models better learn the relationship between the vector ratio and the fault classification, and improve the classification performance and accuracy of the sub-models.

[0147] In some embodiments, training the initial classification model to obtain the pre-trained classification model can include:

[0148] inputting a document including vector ratio samples and corresponding fault classification labels, and reading the vector ratio samples and the corresponding fault classification labels in the document using a pd.read_excel function.

[0149] detecting whether there are invalid values (Not a Number, NaN) or infinite values (Infinity, Inf) in the vector ratio samples and the corresponding fault classification labels, and filtering out the invalid values and the infinite values.

[0150] randomly classifying the vector ratio and the corresponding fault classification labels, so that each fault state has 250 training samples and 250 test samples.

[0151] The training samples corresponding to the plurality of fault states are input into an initial classification model. A total of 100 decision tree sub-models are used, and each decision tree is trained independently using an autonomous bagging strategy. A total of T sub-training sets Dt of size n are generated from the original data set formed by the training samples using a sampling method with replacement, and each sub-training set is used to train a decision tree.

[0152] For each decision tree, the feature space is recursively divided to construct the decision tree. Specifically, let the sample set contained in the current decision tree node be Dv, and select a division point from the candidate features such that the impurity index of a certain classification node after classification is minimized. In this embodiment, the Gini coefficient is used as the impurity index, and the calculation formula of the Gini coefficient for a certain classification node after classification is shown in formula (24):

[0153]

[0154] wherein Gini(D v ) is the Gini coefficient, is the probability that the sample in the node belongs to the kth class.

[0155] The division feature and the division point are selected such that the weighted sum of the impurity indexes of the child nodes is minimized, and the calculation formula is shown in formula (25):

[0156]

[0157] wherein θ * is the division point corresponding to the division feature, D L is the left child node after division, and D R is the right child node after division.

[0158] In one example, a 1.1-kw motor with a rated speed of 2850 RPM is taken as an example, and the motor with a 5%, 10%, 12%, 15%, and 17% turn-to-turn short circuit fault and a healthy motor of this type are measured and tested. The three-phase current is sampled using a current clamp, and the sampling frequency is 20480. The three-phase stator current data obtained is operated according to the above steps, the vector diagram of the healthy motor is shown in FIG. 5, the vector diagram of the motor with a short circuit ratio of 12% is shown in FIG. 6, and the vector diagram of the motor with a short circuit ratio of 17% is shown in FIG. 7. Figure 2 Figure 3 Figure 4

[0159] ​​​The relationship table of the specific fault severity value and the vector ratio value is shown in Table 1. Table 2 is a relationship table of the input vector ratio and the predicted fault degree. It can be seen that the error between the predicted fault degree obtained by using the present application and the actual fault severity value is small, which shows that the fault severity of the asynchronous motor turn-to-turn short circuit can be quantitatively diagnosed by using the present application, and the quantitative diagnosis of the asynchronous motor turn-to-turn short circuit fault is realized.

[0160] Table 1 Relationship table of fault severity value and vector ratio data

[0161] Fault degree Health 5％ 10％ 12％ 15％ 17％ Vector ratio 2.14 7.27 7.44 7.58 8.47 9.51

[0162] Table 2 Relationship table of vector ratio and predicted fault degree

[0163]

[0164] On the basis of the obtained vector ratio, the vector ratio is input into the classification model of the present application, and the obtained fault classification result is shown in Table 3, and the corresponding output confusion matrix is shown in Table 4. Figure 5 The experimental results show that the classification of the present method has a classification accuracy of more than 95% for each type of fault, which shows that the technical scheme of the present application has practicality and can realize the quantitative diagnosis of the asynchronous motor turn-to-turn short circuit fault.

[0165] Table 3 Fault classification result

[0166]

[0167] Figure 6 An asynchronous motor short circuit diagnosis device 600 provided by an embodiment of the present application is shown, which can include:

[0168] An acquisition module 601 is configured to acquire stator three-phase current data of an asynchronous motor.

[0169] A removal module 602 is configured to remove a direct current component in the stator three-phase current data to obtain target three-phase current data.

[0170] A conversion module 603 is configured to perform coordinate conversion on the target three-phase current data to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and generate a vector diagram according to the current components, and the two-phase rotating coordinate system rotates synchronously with a rotor of the asynchronous motor.

[0171] An extraction module 604 is configured to extract image contour features in the vector diagram, and determine a vector ratio of the image in the vector diagram according to the image contour features.

[0172] The determining module 605 is configured to input the vector ratio of the image in the vector diagram into a pre-trained fault detection model, and determine a fault severity value of the asynchronous motor corresponding to the vector ratio of the image in the vector diagram according to a relationship between the vector ratio and the fault severity value in the fault detection model.

[0173] In some embodiments, the asynchronous motor short circuit diagnosis apparatus 600 can further include:

[0174] The calculating module is configured to calculate an average value of each-phase current data in the three-phase current data.

[0175] The calculating module is further configured to calculate a difference value between each-phase current data and the average value of each-phase current data to obtain target three-phase current data.

[0176] In some embodiments, the calculating module is further configured to multiply the target three-phase current data by a preset transformation matrix to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system.

[0177] The calculating module is further configured to calculate a phase angle and a modulus value between the two current components, and generate a vector diagram according to the phase angle and the modulus value, the phase angle being a polar angle of the vector diagram and the modulus value being a polar radius of the vector diagram.

[0178] In some embodiments, the determining module 605 is further configured to determine a centroid of the image contour according to a relationship among the zeroth moment, the first moment and the centroid of the image contour.

[0179] The calculating module is further configured to obtain a target point on the image contour closest to the centroid of the image contour, and calculate a distance between the centroid of the image contour and the target point to obtain a minimum inscribed circle radius.

[0180] The determining module 605 is further configured to determine the vector ratio of the image in the vector diagram according to a relationship among a maximum distance between two points on the image contour, the minimum inscribed circle radius and the vector ratio.

[0181] In some embodiments, the asynchronous motor short circuit diagnosis apparatus 600 can further include:

[0182] The calculating module is further configured to subtract a preset intercept from the vector ratio and divide the result by a preset slope to obtain the fault severity value.

[0183] The establishing module is configured to establish a relationship between the vector ratio and the fault severity value.

[0184] In some embodiments, the asynchronous motor short circuit diagnosis apparatus 600 can further include:

[0185] The obtaining module 601 is further configured to obtain a first training sample, the first training sample including a plurality of vector ratio samples and corresponding fault severity sample values.

[0186] The determination module 605 is further configured to input the vector ratio sample into the initial fault detection model, and determine a predicted fault severity value corresponding to the vector ratio sample by a relationship between the vector ratio and the fault severity value in the initial fault detection model.

[0187] The training module is configured to determine a loss function value of the initial fault detection model according to the predicted fault severity value and the fault severity sample value, and train model parameters of the initial fault detection model according to the loss function value until a training stop condition is met, to obtain a pre-trained fault detection model.

[0188] In some embodiments, the determination module 605 is further configured to input the vector ratio into a plurality of sub-models of the pre-trained classification model, and determine a plurality of predicted classification results corresponding to the plurality of sub-models according to the vector ratio and preset model parameters of the sub-models.

[0189] The determination module 605 is further configured to determine the target classification result according to a number of different predicted classification results in the plurality of predicted classification results.

[0190] In some embodiments, the device 600 for diagnosing short circuit of an asynchronous motor can further include:

[0191] The acquisition module 601 is further configured to acquire a second training sample, the second training sample including a plurality of vector ratio samples and corresponding fault classification labels.

[0192] The classification module is configured to randomly select a preset number of vector ratio samples multiple times and input the selected vector ratio samples into a plurality of initial sub-models of an initial classification model, for each initial sub-model, perform multiple binary classifications on the input vector ratio samples by selecting a plurality of division thresholds, until the fault classification labels of the vector ratio samples in each classification are the same, and obtain a plurality of sub-models including model parameters of the plurality of division thresholds.

[0193] Figure 6 Each module in the device shown can implement Figure 1 each step in the method and achieve the corresponding technical effects, which will not be described herein for brevity.

[0194] Figure 7 A hardware structure schematic diagram of a terminal device provided by an embodiment of the present application is shown.

[0195] The terminal device can include a processor 701 and a memory 702 having computer program instructions stored therein.

[0196] Specifically, the processor 701 can include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or can be configured to implement one or more integrated circuits that implement the embodiments of the present application.

[0197] The memory 702 can include a mass storage for data or instructions. By way of example and not limitation, the memory 702 can include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive or a combination of two or more of these. In one example, the memory 702 can include removable or non-removable (or fixed) media, or the memory 702 is non-volatile solid state memory. The memory 702 can be internal or external to the integrated gateway disaster recovery device.

[0198] In one example, the memory 702 can include read-only memory (ROM), random access memory (RAM), a disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software that, when executed (e.g., by one or more processors), is operable to perform operations described with reference to the method for asynchronous motor short circuit diagnosis according to the present disclosure.

[0199] The processor 701 implements the method for asynchronous motor short circuit diagnosis in the embodiments by reading and executing computer program instructions stored in the memory 702. Figure 1

[0200] In one example, the terminal device can also include a communication interface 703 and a bus 704. Wherein, as shown, the processor 701, the memory 702, the communication interface 703 are connected through the bus 704 and complete the communication between each other. Figure 7

[0201] The communication interface 703 is mainly used to realize the communication between each module, device, unit and / or equipment in the embodiments of the present application.

[0202] ​​Bus 704 includes a hardware, software, or both that couples components of terminal device to each other. As an example and not by way of limitation, bus can include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand™ interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association local (VLB) bus, or another suitable bus or a combination of two or more of these. Where appropriate, bus 704 can include one or more buses. Although this application describes and shows a particular bus, this application contemplates any suitable bus or interconnect.

[0203] In addition, in combination with the method for diagnosing short circuit of asynchronous motor in the above-mentioned embodiments, the embodiments of the present application can provide a computer storage medium for implementation. The computer storage medium has computer program instructions stored thereon; the computer program instructions are executed by a processor to implement any of the methods for diagnosing short circuit of asynchronous motor in the above-mentioned embodiments.

[0204] The embodiments of the present application also provide a computer program product, which includes a computer program, and the computer program is executed by a processor to implement any of the methods for diagnosing short circuit of asynchronous motor in the above-mentioned embodiments.

[0205] It needs to be clear that the present application is not limited to the specific configurations and processes described above and shown in the drawings. For the sake of brevity, detailed descriptions of well-known methods are omitted here. In the above-mentioned embodiments, several specific steps are described and shown as examples. However, the method processes of the present application are not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between steps, after understanding the spirit of the present application.

[0206] The functions indicated in the structural block diagrams described above can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, and the like. When implemented in software, the elements of the present application are program or text segments used to perform the required tasks. The program or text segments can be stored in a machine-readable medium or transmitted through a data signal carried in a carrier wave over a transmission medium or communication link. The "machine-readable medium" can include any medium capable of storing or transmitting information. Examples of the machine-readable medium include electronic circuits, semiconductor memory devices, read-only memory (ROM), flash memory, erasable read-only memory (EROM), floppy disks, compact discs (CD-ROM), optical disks, hard disks, optical fiber media, radio frequency (RF) links, and the like. The text segments can be downloaded via a computer network such as the Internet, an intranet, and the like.

[0207] It is also necessary to note that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in an order different from the embodiments, or several steps can be performed simultaneously.

[0208] The above describes the aspects of the present application with reference to the flowcharts and / or block diagrams of the methods, apparatuses (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each block in the flowcharts and / or block diagrams, and the combination of the blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus enable the implementation of the functions / actions specified in one or more blocks of the flowcharts and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each block in the block diagrams and / or flowcharts, and the combination of the blocks in the block diagrams and / or flowcharts, can also be implemented by special hardware that performs the specified functions or actions, or can be implemented by a combination of special hardware and computer instructions.

[0209] The above merely describes a specific implementation of the present application. Those skilled in the art can clearly understand the specific working processes of the system, modules and units described above for the convenience and brevity of description, and can refer to the corresponding processes in the foregoing method embodiments, which will not be described herein again. It should be understood that the protection scope of the present application is not limited to this, and any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered within the protection scope of the present application.

Claims

1. A method for diagnosing short circuit of an asynchronous motor, characterized in that: The method comprises: Obtain the stator three-phase current data of the asynchronous motor; removing a DC component from the stator three-phase current data to obtain target three-phase current data; performing coordinate transformation on the target three-phase current data to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and generating a vector diagram based on the current components, wherein the two-phase rotating coordinate system rotates synchronously with the rotor of the asynchronous motor; extracting image contour features from the vector image, and determining a vector ratio of the image in the vector image based on the image contour features; The vector ratio of the images in the vector diagram is input into a pre-trained fault detection model, and the fault severity value of the asynchronous motor corresponding to the vector ratio of the images in the vector diagram is determined based on the relationship between the vector ratio and the fault severity value in the fault detection model.

2. The method for diagnosing short circuit of an asynchronous motor according to claim 1, characterized in that: The removing of the DC component from the stator three-phase current data to obtain target three-phase current data includes: Calculating an average value of each phase current data in the three-phase current data; The difference between each phase current data and the average value of each phase current data is calculated to obtain the target three-phase current data.

3. The method for diagnosing short circuit of an asynchronous motor according to claim 1, characterized in that: The coordinate transformation of the target three-phase current data to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and generating a vector diagram according to the current components, includes: Multiplying the target three-phase current data by a preset transformation matrix to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system; The phase angle and the modulus between the two current components are calculated, and a vector diagram is generated according to the phase angle and the modulus, wherein the phase angle is the polar angle of the vector diagram, and the modulus is the polar diameter of the vector diagram.

4. The method for diagnosing short circuit of an asynchronous motor according to claim 1, characterized in that: The image contour features include the maximum distance between two points on the image contour of the vector graph, the zero-order moment of the image contour, and the first-order moment of the image contour; and determining the vector ratio of the image in the vector graph based on the image contour features includes: Determine the centroid of the image contour corresponding to the zero-order moment of the image contour and the first-order moment of the image contour according to the relationship among the zero-order moment, the first-order moment and the centroid; Obtaining a target point on the image contour that is closest to the centroid of the image contour, and calculating the distance between the centroid of the image contour and the target point to obtain a minimum inscribed circle radius; The vector ratio of the image in the vector graph is determined according to the relationship between the maximum distance between two points on the image contour, the minimum inscribed circle radius and the vector ratio.

5. The method for diagnosing short circuit of an asynchronous motor according to claim 1, characterized in that: Before inputting the vector ratio of the images in the vector diagram into a pre-trained fault detection model and determining the fault severity value of the asynchronous motor corresponding to the vector ratio of the images in the vector diagram based on the relationship between the vector ratio and the fault severity value in the fault detection model, the method further includes: Subtract a preset intercept from the vector ratio and divide it by a preset slope to obtain a fault severity value; The relationship between the vector ratio and the fault severity value is established.

6. The method for diagnosing short circuit of an asynchronous motor according to claim 1, characterized in that: Before inputting the vector ratio of the images in the vector diagram into a pre-trained fault detection model and determining the fault severity value of the asynchronous motor corresponding to the vector ratio of the images in the vector diagram based on the relationship between the vector ratio and the fault severity value in the fault detection model, the method further includes: Acquire a first training sample, wherein the first training sample includes a plurality of vector ratio samples and corresponding fault severity sample values; Inputting the vector ratio sample into an initial fault detection model, and determining a predicted fault severity value corresponding to the vector ratio sample based on a relationship between the vector ratio and the fault severity value in the initial fault detection model; The loss function value of the initial fault detection model is determined according to the predicted fault severity value and the fault severity sample value, and the model parameters of the initial fault detection model are trained according to the loss function value until the training stop condition is met to obtain a pre-trained fault detection model.

7. The method for diagnosing short circuit of an asynchronous motor according to claim 1, characterized in that: After extracting the image contour features in the vector image and determining the vector ratio of the image in the vector image according to the image contour features, the method further includes: Inputting the vector ratio into multiple sub-models of a pre-trained classification model, and determining multiple predicted classification results corresponding to the multiple sub-models based on the vector ratio and preset model parameters of the sub-models; A target classification result is determined according to the number of different predicted classification results in the multiple predicted classification results.

8. The method for diagnosing short circuit of an asynchronous motor according to claim 7, characterized in that: Before inputting the vector ratio into a plurality of sub-models of a pre-trained classification model, the method further comprises: Acquire a second training sample, where the second training sample includes a plurality of vector ratio samples and corresponding fault classification labels; A preset number of vector ratio samples are randomly selected multiple times and input into multiple initial sub-models of the initial classification model respectively. For each initial sub-model, the input vector ratio samples are binary classified multiple times by selecting multiple partitioning thresholds until the fault classification labels of the vector ratio samples in each classification are the same, thereby obtaining multiple sub-models including model parameters of multiple partitioning thresholds.

9. A device for diagnosing short circuit of an asynchronous motor, characterized in that: The device comprises: An acquisition module is used to obtain the stator three-phase current data of the asynchronous motor; a removal module, configured to remove a DC component from the stator three-phase current data to obtain target three-phase current data; a conversion module, configured to perform coordinate conversion on the target three-phase current data to obtain two current components of the target three-phase current data in a two-phase rotating coordinate system, and generate a vector diagram based on the current components, wherein the two-phase rotating coordinate system rotates synchronously with the rotor of the asynchronous motor; an extraction module, configured to extract image contour features from the vector image, and determine a vector ratio of the image in the vector image based on the image contour features; A determination module is used to input the vector ratio of the image in the vector diagram into a pre-trained fault detection model, and determine the fault severity value of the asynchronous motor corresponding to the vector ratio of the image in the vector diagram through the relationship between the vector ratio and the fault severity value in the fault detection model.

10. A terminal device, characterized in that: The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the method for diagnosing short circuit of an asynchronous motor according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method for diagnosing short circuit of an asynchronous motor according to any one of claims 1 to 8 is implemented.

12. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the method for diagnosing short circuit of an asynchronous motor according to any one of claims 1 to 8.

Images