Intelligent diagnosis system and method for vehicle water pump brushless motor

Abstract

The invention relates to the technical field of motor testing, and discloses an intelligent diagnosis system and method for a brushless motor of a vehicle water pump, and the method comprises the steps: collecting an actually-measured first vibration mean value, and obtaining an estimated second vibration mean value of a brushless motor to be tested at the r + v rotating speed based on the actually-measured first vibration mean value; acquiring an actually measured second vibration mean value of the brushless motor to be measured at the r + v rotating speed, and judging whether the brushless motor to be measured has abnormal performance according to the estimated second vibration mean value and the actually measured second vibration mean value; the process is repeated until abnormal performance occurs or r is equal to R, circulation is ended, the temperature reference value, the temperature mean value and the electrical measurement value are input into a first machine learning model, and an electrical correction value is obtained; intercepting an abnormal wave band in the actual electrical oscillogram, and inputting the abnormal wave band into a second machine learning model in an image form to obtain a fault type of the brushless motor to be tested; according to the invention, the fault type of the brushless motor can be diagnosed with high accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of motor testing, and more specifically, to an intelligent diagnosis system and method for a brushless motor of a vehicle water pump. Background Art

[0002] With the rapid development of the automobile industry, automotive electronic water pumps are increasingly used in engine cooling systems. Compared with traditional mechanical water pumps, electronic water pumps have the advantages of high efficiency, low energy consumption, and flexible control, and have become an important part of modern automobile cooling systems. As the core component of electronic water pumps, brushless motors are widely used in electronic water pumps due to their high efficiency, low noise, and long life. However, during the manufacturing and assembly process of brushless motors, problems such as winding short circuit, open circuit, rotor imbalance, and control circuit failure may occur. If these problems are not effectively detected and resolved before leaving the factory, the electronic water pump may experience performance degradation, shortened life, or even failure during actual use, thereby affecting the performance and safety of the entire vehicle. Therefore, how to perform effective diagnosis and testing of brushless motors before leaving the factory has become a focus of attention in the industry.

[0003] At present, the traditional motor detection method mainly relies on manual detection and simple instrument measurement, which has the defects of low detection efficiency, low accuracy, and susceptibility to human factors. Although there are some intelligent improved technical documents, for example, the patent with publication number CN116430225A discloses a test device for a brushless motor drive system. This method simulates the occurrence of faults by setting a test board, thereby realizing the test of the brushless motor drive system. However, the above method and the prior art are studied and applied in practice, and it is found that the above method and the prior art have at least the following defects: (1) The detection parameters and methods are single, and it is impossible to identify abnormal changes in the brushless motor based on the mutual influence or change performance of multiple parameters; (2) It is not possible to detect the electrical parameters of the brushless motor and correct the electrical parameters based on the identification of abnormal changes, making it difficult to accurately detect the fault type of the brushless motor based on the corrected electrical parameters. Furthermore, it is impossible to guarantee the performance of the electronic water pump during actual use. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent diagnosis system and method for a brushless motor of a vehicle water pump.

[0005] To achieve the above object, the present invention provides the following technical solutions: An intelligent diagnostic system for a brushless motor of a vehicle water pump, the system comprising: A data acquisition module is used to collect a first vibration mean value measured during the operation of the brushless motor to be tested at an rth speed within a predetermined test time range, and obtain an estimated second vibration mean value of the brushless motor to be tested at an r+vth speed based on the first vibration mean value measured, where r and v are integers greater than zero; The abnormality detection module is used to collect the measured second vibration mean value of the brushless motor to be tested at the r+vth speed, and judge whether the brushless motor to be tested has abnormal performance according to the estimated second vibration mean value and the measured second vibration mean value. If it does, the temperature mean value and electrical measurement value of the brushless motor to be tested at the rth speed are collected; if it does not, v=v+d is set, and the data acquisition module is triggered, where d is an integer greater than zero; A data correction module is used to repeat the above data acquisition module to the abnormality detection module until the brushless motor to be tested shows abnormal performance, or when r=R, the loop is terminated, a temperature reference value is obtained, and the temperature reference value, the temperature mean and the electrical measurement value are input into a preconfigured first machine learning model to obtain an electrical correction value of the brushless motor to be tested within a predetermined test time range, where R is a set maximum speed of the brushless motor to be tested; The intelligent diagnosis module is used to construct an actual electrical waveform diagram according to the electrical correction value, intercept the abnormal band in the actual electrical waveform diagram, input the abnormal band into a preconfigured second machine learning model in the form of an image, and obtain the fault type of the brushless motor to be tested.

[0006] Further, the step of obtaining an estimated second vibration mean value of the brushless motor to be tested at the r+vth speed includes: Calculate the speed difference v between the r+vth speed and the rth speed, and call the preset proportional coefficient; Based on the preset proportional coefficient, the speed difference v and the measured first vibration mean value, a proportional conversion is performed to obtain an estimated second vibration mean value of the brushless motor to be tested at the r+vth speed.

[0007] Further, the step of determining whether the brushless motor to be tested has abnormal performance includes: Calculating an absolute difference between an estimated second vibration mean and an actually measured second vibration mean, and marking the absolute difference between the estimated second vibration mean and the actually measured second vibration mean as a vibration difference; The vibration difference is compared with a preset vibration difference threshold value. If the vibration difference is greater than or equal to the preset vibration difference threshold value, it is determined that the brushless motor to be tested has abnormal performance; if the vibration difference is less than the preset vibration difference threshold value, it is determined that the brushless motor to be tested has no abnormal performance; The electrical measurement is one of a current measurement, a voltage measurement, or a resistance measurement.

[0008] Further, when r=R, it is determined that the brushless motor to be tested has no fault.

[0009] Furthermore, the generation logic of the preconfigured first machine learning model is as follows: Acquire historical electrical correction training data, and divide the historical electrical correction training data into an electrical correction training set and an electrical correction test set; the historical electrical correction training data includes a plurality of electrical correction feature data and their corresponding electrical correction values; Wherein, the electrical correction characteristic data includes the temperature mean value and electrical measurement value of the brushless motor to be tested; Constructing a regression network, using electrical correction feature data in the electrical correction training set as input data of the regression network, and using electrical correction values ​​in the electrical correction training set as output data of the regression network, training the regression network, and obtaining an initial regression network; The initial regression network is model verified using the electrical correction test set, and the initial regression network that is less than or equal to a preset test error threshold is output as the preconfigured first machine learning model.

[0010] Furthermore, the logic for obtaining the electrical correction value in the historical electrical correction training data is as follows: a1: at the same rotation speed, obtaining a temperature reference value of the first test brushless motor, collecting an electrical measurement value of the first test brushless motor at the temperature reference value, and marking the electrical measurement value of the first test brushless motor at the temperature reference value as a first electrical measurement value; a2: At the same speed, collect the temperature average t of the second test brushless motor, and collect the electrical measurement value of the second test brushless motor at the temperature average t, mark the electrical measurement value of the second test brushless motor at the temperature average t as the second electrical measurement value, and t is an integer greater than zero; a3: Calculate the difference between the temperature reference value and the temperature mean t, and mark the difference between the temperature reference value and the temperature mean t as the temperature difference; a4: Calculate the difference between the first electrical measurement value and the second electrical measurement value, mark the difference between the first electrical measurement value and the second electrical measurement value as an electrical deviation, compare the electrical deviation with a preset electrical deviation threshold, and if the electrical deviation is greater than the preset electrical deviation threshold, sum the second electrical measurement value and the electrical deviation, mark the sum of the second electrical measurement value and the electrical deviation as an electrical correction value, associate and bind the corresponding temperature difference with the electrical correction value, obtain the electrical correction value under a certain temperature difference, and set t=t+k, or set t=tk, and return to step a2; if the electrical deviation is less than or equal to the preset electrical deviation threshold, set t=t+k, or set t=tk and return to step a2, where k is an integer greater than zero; a5: Repeat the above steps a2 to a4 until t=MaxT, or t=MinT, then end the cycle and obtain the electrical correction value at each temperature difference.

[0011] Furthermore, the intercepting of abnormal bands in the actual electrical waveform diagram includes: Retrieve the standard electrical waveform diagram of the brushless motor to be tested; Using the actual electrical waveform as a matching image and the standard electrical waveform as a template image; Set the step size to 1, and perform cross-difference calculation between the matching image and the template image by sliding window to obtain the difference degree of each overlapping window; The overlapping windows whose difference is less than or equal to the preset difference threshold are regarded as similar regions, and the overlapping windows whose difference is greater than the preset difference threshold are regarded as different regions; A difference region is segmented from the actual electrical waveform diagram, and the electrical waveform in the segmented difference region is used as an abnormal band.

[0012] Furthermore, the generation logic of the preconfigured second machine learning model is as follows: Acquire historical fault diagnosis training data, and divide the historical fault diagnosis training data into a fault diagnosis training set and a fault diagnosis test set, wherein the historical fault diagnosis training data includes abnormal bands in the form of multiple images and their corresponding fault type annotation labels; Construct a classifier, use the abnormal bands in the image form in the fault diagnosis training set as the input data of the classifier, and use the fault type annotation labels in the fault diagnosis training set as the output data of the classifier, train the classifier, and obtain an initial classification network; The initial classification network is model verified using the fault diagnosis test set, and an initial classification network with a test accuracy greater than or equal to a preset threshold is output as a preconfigured second machine learning model.

[0013] An intelligent diagnosis method for a brushless motor of a vehicle water pump, the method comprising: S201: within a predetermined test time range, collecting a first vibration mean value measured during the operation of the brushless motor to be tested at an rth speed, and obtaining an estimated second vibration mean value of the brushless motor to be tested at an r+vth speed based on the first vibration mean value measured, where r and v are integers greater than zero; S202: collecting the measured second vibration mean value of the brushless motor to be tested at the r+vth speed, judging whether the brushless motor to be tested has abnormal performance according to the estimated second vibration mean value and the measured second vibration mean value, if so, collecting the temperature mean value and electrical measurement value of the brushless motor to be tested at the rth speed; if not, setting v=v+d, and returning to step S201, d is an integer greater than zero; S203: Repeat the above steps S201 to S202 until the brushless motor to be tested shows abnormal performance, or when r=R, the loop is terminated, the temperature reference value is obtained, and the temperature reference value, the temperature mean and the electrical measurement value are input into a pre-configured first machine learning model to obtain the electrical correction value of the brushless motor to be tested within a predetermined test time range, where R is the set maximum speed of the brushless motor to be tested; S204: construct an actual electrical waveform diagram according to the electrical correction value, intercept the abnormal band in the actual electrical waveform diagram, input the abnormal band into the preconfigured second machine learning model in the form of an image, and obtain the fault type of the brushless motor to be tested.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present application discloses an intelligent diagnosis system and method for a brushless motor of a vehicle water pump, comprising: collecting a measured first vibration mean value, and obtaining an estimated second vibration mean value of the brushless motor to be tested at an r+vth speed based on the measured first vibration mean value; collecting the measured second vibration mean value of the brushless motor to be tested at an r+vth speed, and judging whether the brushless motor to be tested has abnormal performance according to the estimated second vibration mean value and the measured second vibration mean value; repeating the above process until abnormal performance occurs, or when r=R, terminating the cycle, inputting a temperature reference value, a temperature mean value, and an electrical measurement value into a first machine learning model, and obtaining an electrical correction value; intercepting the actual electrical The abnormal band in the waveform diagram is input into the second machine learning model in the form of an image to obtain the fault type of the brushless motor to be tested; based on the above process, the present invention can analyze whether the brushless motor has nonlinear changes according to the mutual influence or change performance among multiple parameters, and then can identify the abnormal changes of the brushless motor; in addition, on the basis of identifying the abnormal changes, the electrical parameters of the brushless motor are detected and corrected, which is beneficial to accurately detect the fault type of the brushless motor according to the corrected electrical parameters, and further, it is beneficial to ensure the performance of the electronic water pump during actual use. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A module structure diagram of the intelligent diagnosis system for the brushless motor of a vehicle water pump provided by the present invention; Figure 2 The present invention provides a flow chart of the intelligent diagnosis method for the brushless motor of a vehicle water pump. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example 1

[0017] See also Figure 1 As shown, this embodiment discloses an intelligent diagnosis system for a brushless motor of a vehicle water pump, the system comprising: The data acquisition module 101 is used to collect the measured first vibration mean value during the operation of the brushless motor to be tested at the rth speed within a predetermined test time range, and obtain the estimated second vibration mean value of the brushless motor to be tested at the r+vth speed based on the measured first vibration mean value, where r and v are integers greater than zero; It should be noted that: the specific duration of the predetermined test time range is artificially set by the technicians according to experiments or experience; the measured first vibration mean value is determined by summing the vibration values ​​at each moment within the predetermined test time range and dividing the sum by the specific duration of the predetermined test time range; the vibration value at each moment is collected in real time by a preset vibration sensor; In implementation, the step of obtaining the estimated second vibration mean value of the brushless motor to be tested at the r+vth speed includes: Calculate the speed difference v between the r+vth speed and the rth speed, and call the preset proportional coefficient; Based on the preset proportional coefficient, the speed difference v and the measured first vibration mean value, a proportional conversion is performed to obtain an estimated second vibration mean value of the brushless motor to be tested at the r+vth speed; It should be noted that: the preset proportional coefficient is pre-stored in the system database, and the preset proportional coefficient is obtained by the technicians through the motor test experiment. The preset proportional coefficient reflects the trend of the vibration mean value or vibration value of the brushless motor to be tested changing with the speed; further explanation is that for a normal brushless motor to be tested, its vibration mean value or vibration value will change with the change of the motor speed, and this change trend is linear, and once a nonlinear change occurs, it means that the brushless motor to be tested has abnormal characteristics; For example, assuming that when r=2000RPM, the measured first vibration mean of the brushless motor to be tested is 0.20mm / s, and assuming that the speed difference v=500RPM, and according to the preset proportional coefficient, it can be known that for every increase of 500RPM, the vibration mean increases by approximately 10%, then the preset proportional coefficient is 1.1. Therefore, after proportional conversion by the preset proportional coefficient, the speed difference v and the measured first vibration mean, it can be known that the estimated second vibration mean=0.20mm / s×1.1=0.22mm / s.

[0018] The abnormality detection module 102 is used to collect the measured second vibration mean value of the brushless motor to be tested at the r+vth speed, and judge whether the brushless motor to be tested has abnormal performance according to the estimated second vibration mean value and the measured second vibration mean value. If it does, the temperature mean value and electrical measurement value of the brushless motor to be tested at the rth speed are collected; if it does not, v=v+d is set, and the data acquisition module 101 is triggered, where d is an integer greater than zero; It should be noted that the logic and principle for obtaining the measured second vibration mean value are the same as those for the measured first vibration mean value. For details, please refer to the relevant part above, and no further elaboration will be given. In implementation, the step of determining whether the brushless motor to be tested exhibits abnormal performance includes: Calculating an absolute difference between an estimated second vibration mean and an actually measured second vibration mean, and marking the absolute difference between the estimated second vibration mean and the actually measured second vibration mean as a vibration difference; The vibration difference is compared with a preset vibration difference threshold value. If the vibration difference is greater than or equal to the preset vibration difference threshold value, it is determined that the brushless motor to be tested has abnormal performance; if the vibration difference is less than the preset vibration difference threshold value, it is determined that the brushless motor to be tested has no abnormal performance; Specifically, the electrical measurement value is one of a current measurement value, a voltage measurement value or a resistance measurement value; It should be noted that the temperature average value and electrical measurement value of the brushless motor to be tested are acquired through preset sensing equipment, and the preset sensing equipment includes but is not limited to temperature sensors, current sensors, voltage sensors, resistance sensors, etc.

[0019] The data correction module 103 is used to repeat the above data acquisition module 101 to the abnormality detection module 102 until the brushless motor to be tested shows abnormal performance, or when r=R, the loop is terminated, the temperature reference value is obtained, and the temperature reference value, the temperature mean and the electrical measurement value are input into a preconfigured first machine learning model to obtain the electrical correction value of the brushless motor to be tested within a predetermined test time range, where R is the set maximum speed of the brushless motor to be tested; It should be noted that: the temperature reference value is pre-stored in the system database; the set maximum speed R of the brushless motor to be tested refers to the maximum speed limited when the brushless motor to be tested leaves the factory, which is set in advance according to the factory speed standards of different models of brushless motors to be tested; In a specific implementation, when r=R, it is determined that the brushless motor to be tested does not have a fault; It can be understood that: when r=R, the absolute difference between the estimated second vibration mean value of the brushless motor to be tested at the Rv speed and the measured second vibration mean value of the brushless motor to be tested at the R speed is calculated, that is, the vibration difference is obtained. At this time, the vibration difference is compared with the preset vibration difference threshold value, and it is found that after multiple cycles, the vibration difference is still less than the preset vibration difference threshold value, which further indicates that the brushless motor to be tested has no abnormal performance, that is, the brushless motor to be tested has no fault; Specifically, the generation logic of the preconfigured first machine learning model is as follows: Acquire historical electrical correction training data, and divide the historical electrical correction training data into an electrical correction training set and an electrical correction test set; the historical electrical correction training data includes a plurality of electrical correction feature data and their corresponding electrical correction values; Wherein, the electrical correction characteristic data includes the temperature mean value and electrical measurement value of the brushless motor to be tested; It should be understood that the electrical correction feature data in the historical electrical correction training data is collected by various sensing devices, and the details can be referred to the relevant part of the above description, so it will not be repeated here; The logic for obtaining the electrical correction value in the historical electrical correction training data is as follows: a1: at the same rotation speed, obtaining a temperature reference value of the first test brushless motor, collecting an electrical measurement value of the first test brushless motor at the temperature reference value, and marking the electrical measurement value of the first test brushless motor at the temperature reference value as a first electrical measurement value; a2: At the same speed, collect the temperature average t of the second test brushless motor, and collect the electrical measurement value of the second test brushless motor at the temperature average t, mark the electrical measurement value of the second test brushless motor at the temperature average t as the second electrical measurement value, and t is an integer greater than zero; It should be noted that: the first test brushless motor, the second test brushless motor and the brushless motor to be tested are consistent in terms of model and specifications; and in the test scenario, the first test brushless motor and the second test brushless motor have the same parameters except for the temperature parameter; a3: Calculate the difference between the temperature reference value and the temperature mean t, and mark the difference between the temperature reference value and the temperature mean t as the temperature difference; a4: Calculate the difference between the first electrical measurement value and the second electrical measurement value, mark the difference between the first electrical measurement value and the second electrical measurement value as an electrical deviation, compare the electrical deviation with a preset electrical deviation threshold, and if the electrical deviation is greater than the preset electrical deviation threshold, sum the second electrical measurement value and the electrical deviation, mark the sum of the second electrical measurement value and the electrical deviation as an electrical correction value, associate and bind the corresponding temperature difference with the electrical correction value, obtain the electrical correction value under a certain temperature difference, and set t=t+k, or set t=tk, and return to step a2; if the electrical deviation is less than or equal to the preset electrical deviation threshold, set t=t+k, or set t=tk and return to step a2, where k is an integer greater than zero; It should be noted that: although the electrical measurement value may be a current measurement value, a voltage measurement value or a resistance measurement value, when the selected electrical measurement value is one of the current measurement value, the voltage measurement value or the resistance measurement value, the electrical deviation is the same as the attribute of the selected electrical measurement value; for example, when the selected electrical measurement value is the current measurement value, the electrical deviation is the current attribute, that is, the current deviation; a5: Repeat the above steps a2 to a4 until t=MaxT, or t=MinT, then end the cycle and obtain the electrical correction value at each temperature difference; It should be noted that: [MinT, MaxT] is a temperature range. It is understandable that during testing or actual operation, the brushless motor is set to a safe operating temperature range. Therefore, when the electrical parameters are corrected according to temperature changes, they should also be within a reasonable change range. The temperature range [MinT, MaxT] is set by the technicians in advance by experiments; Constructing a regression network, using electrical correction feature data in the electrical correction training set as input data of the regression network, and using electrical correction values ​​in the electrical correction training set as output data of the regression network, training the regression network, and obtaining an initial regression network; Using the electrical correction test set to perform model verification on the initial regression network, outputting an initial regression network that is less than or equal to a preset test error threshold as a preconfigured first machine learning model; It should be noted that: the regression network is a specific one of the regression algorithm models such as decision tree, random forest, polynomial regression, linear regression, support vector machine or neural network; It should be understood that the temperature of the brushless motor will affect the measurement accuracy of the electrical parameters. Therefore, the present invention improves the accuracy of the measurement of the electrical parameters of the brushless motor by introducing a compensation mechanism, which can help improve the accuracy of subsequent fault diagnosis of the brushless motor.

[0020] The intelligent diagnosis module 104 is used to construct an actual electrical waveform diagram according to the electrical correction value, intercept the abnormal band in the actual electrical waveform diagram, input the abnormal band into a preconfigured second machine learning model in the form of an image, and obtain the fault type of the brushless motor to be tested; It should be noted that: the actual electrical waveform diagram is constructed by inputting the electrical correction value of the brushless motor to be tested within a predetermined test time range into a data analysis software (such as origin, etc.), that is, the time is used as the horizontal axis and the electrical correction value is used as the horizontal axis to construct the actual electrical waveform diagram; In implementation, the intercepting of abnormal bands in the actual electrical waveform diagram includes: Retrieve the standard electrical waveform diagram of the brushless motor to be tested; It should be noted that: the standard electrical waveform diagram is pre-stored in the system database, and the standard electrical waveform diagram is an electrical waveform diagram of the brushless motor of the type to be tested in a non-fault condition or state, which reflects the electrical waveform of the brushless motor of the type to be tested in a normal state; Using the actual electrical waveform as a matching image and the standard electrical waveform as a template image; Set the step size to 1, and perform cross-difference calculation between the matching image and the template image by sliding window to obtain the difference degree of each overlapping window; It should be understood that the sliding window method is a method of image processing, and its basic idea is to use one image as a template and another image as a matching object, place the matching object on the template, and calculate the similarity or difference of each overlapping part by sliding the window of the matching object on the template, so as to obtain the similar area and the difference area of ​​the two images; the difference of each overlapping window is calculated by using a difference algorithm, and the difference algorithm includes but is not limited to the Euclidean distance algorithm, etc. The overlapping windows whose difference is less than or equal to the preset difference threshold are regarded as similar regions, and the overlapping windows whose difference is greater than the preset difference threshold are regarded as different regions; Segment a difference region from the actual electrical waveform diagram, and use the electrical waveform in the segmented difference region as an abnormal band; Specifically, the generation logic of the preconfigured second machine learning model is as follows: Acquire historical fault diagnosis training data, and divide the historical fault diagnosis training data into a fault diagnosis training set and a fault diagnosis test set, wherein the historical fault diagnosis training data includes abnormal bands in the form of multiple images and their corresponding fault type annotation labels; It should be noted that: the abnormal bands in the image form in the historical fault diagnosis training data are obtained by segmenting the actual electrical waveform diagram, and the details are described in the relevant part above; the fault type labeling labels in the historical fault diagnosis training data are formed by technicians or AI data annotators manually labeling the abnormal bands in each image form, and the fault types include but are not limited to short circuit, rotor eccentricity, bearing defect and stator looseness, etc.; Construct a classifier, use the abnormal bands in the image form in the fault diagnosis training set as the input data of the classifier, and use the fault type annotation labels in the fault diagnosis training set as the output data of the classifier, train the classifier, and obtain an initial classification network; Using the fault diagnosis test set to perform model verification on the initial classification network, outputting an initial classification network with a value greater than or equal to a preset test accuracy threshold as a preconfigured second machine learning model; It should be noted that: the classifier is a specific one of the classification algorithm models such as decision tree, logistic regression, random forest, naive Bayes, support vector machine or neural network; By comparing the difference between the estimated second vibration mean value and the measured second vibration mean value of the brushless motor, the present invention can analyze whether the brushless motor has nonlinear changes, and then determine whether the brushless motor has abnormal performance; on the basis of determining that the brushless motor has abnormal performance, by measuring the electrical parameters of the brushless motor and correcting the electrical parameters, important data support is provided for subsequent brushless motor fault diagnosis, and it is beneficial to ensure the accuracy of subsequent fault diagnosis; finally, by identifying abnormal bands through intelligent models, the present invention can quickly and accurately diagnose the fault type of the brushless motor, and is suitable for large-scale brushless motor production scenarios. Example 2

[0021] See also Figure 2 As shown, based on the same inventive concept, this embodiment discloses an intelligent diagnosis method for a brushless motor of a vehicle water pump. For details not provided in this embodiment, please refer to the relevant parts of the above embodiment 1. The method includes: S201: within a predetermined test time range, collecting a first vibration mean value measured during the operation of the brushless motor to be tested at an rth speed, and obtaining an estimated second vibration mean value of the brushless motor to be tested at an r+vth speed based on the first vibration mean value measured, where r and v are integers greater than zero; S202: collecting the measured second vibration mean value of the brushless motor to be tested at the r+vth speed, judging whether the brushless motor to be tested has abnormal performance according to the estimated second vibration mean value and the measured second vibration mean value, if so, collecting the temperature mean value and electrical measurement value of the brushless motor to be tested at the rth speed; if not, setting v=v+d, and returning to step S201, d is an integer greater than zero; S203: Repeat the above steps S201 to S202 until the brushless motor to be tested shows abnormal performance, or when r=R, the loop is terminated, the temperature reference value is obtained, and the temperature reference value, the temperature mean and the electrical measurement value are input into a pre-configured first machine learning model to obtain the electrical correction value of the brushless motor to be tested within a predetermined test time range, where R is the set maximum speed of the brushless motor to be tested; S204: construct an actual electrical waveform diagram according to the electrical correction value, intercept the abnormal band in the actual electrical waveform diagram, input the abnormal band into the preconfigured second machine learning model in the form of an image, and obtain the fault type of the brushless motor to be tested.

[0022] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments 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 or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired network or a wireless network. 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 data center that includes one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD) or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0023] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0024] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0025] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0026] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0027] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0028] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0029] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent diagnostic system for a brushless motor of a vehicle water pump, characterized in that: The system comprises: A data acquisition module is used to collect a first vibration mean value measured during the operation of the brushless motor to be tested at an rth speed within a predetermined test time range, and obtain an estimated second vibration mean value of the brushless motor to be tested at an r+vth speed based on the first vibration mean value measured, where r and v are integers greater than zero; The abnormality detection module is used to collect the measured second vibration mean value of the brushless motor to be tested at the r+vth speed, and judge whether the brushless motor to be tested has abnormal performance according to the estimated second vibration mean value and the measured second vibration mean value. If it does, the temperature mean value and electrical measurement value of the brushless motor to be tested at the rth speed are collected; if it does not, v=v+d is set, and the data acquisition module is triggered, where d is an integer greater than zero; A data correction module is used to repeat the above data acquisition module to the abnormality detection module until the brushless motor to be tested shows abnormal performance, or when r=R, the loop is terminated, a temperature reference value is obtained, and the temperature reference value, the temperature mean and the electrical measurement value are input into a preconfigured first machine learning model to obtain an electrical correction value of the brushless motor to be tested within a predetermined test time range, where R is a set maximum speed of the brushless motor to be tested; The intelligent diagnosis module is used to construct an actual electrical waveform diagram according to the electrical correction value, intercept the abnormal band in the actual electrical waveform diagram, input the abnormal band into a preconfigured second machine learning model in the form of an image, and obtain the fault type of the brushless motor to be tested.

2. The intelligent diagnosis system for brushless motor of vehicle water pump according to claim 1 is characterized in that: The step of obtaining an estimated second vibration mean value of the brushless motor to be tested at the r+vth speed includes: Calculate the speed difference v between the r+vth speed and the rth speed, and call the preset proportional coefficient; Based on the preset proportional coefficient, the speed difference v and the measured first vibration mean value, a proportional conversion is performed to obtain an estimated second vibration mean value of the brushless motor to be tested at the r+vth speed.

3. The intelligent diagnosis system for brushless motor of vehicle water pump according to claim 2, characterized in that: The step of determining whether the brushless motor to be tested has abnormal performance includes: Calculating an absolute difference between an estimated second vibration mean and an actually measured second vibration mean, and marking the absolute difference between the estimated second vibration mean and the actually measured second vibration mean as a vibration difference; The vibration difference is compared with a preset vibration difference threshold value. If the vibration difference is greater than or equal to the preset vibration difference threshold value, it is determined that the brushless motor to be tested has abnormal performance; if the vibration difference is less than the preset vibration difference threshold value, it is determined that the brushless motor to be tested has no abnormal performance; The electrical measurement is one of a current measurement, a voltage measurement, or a resistance measurement.

4. The intelligent diagnosis system for brushless motor of vehicle water pump according to claim 3 is characterized in that: When r=R, it is determined that the brushless motor to be tested has no fault.

5. The intelligent diagnosis system for brushless motor of vehicle water pump according to claim 4, characterized in that: The generation logic of the preconfigured first machine learning model is as follows: Acquire historical electrical correction training data, and divide the historical electrical correction training data into an electrical correction training set and an electrical correction test set; the historical electrical correction training data includes a plurality of electrical correction feature data and their corresponding electrical correction values; Wherein, the electrical correction characteristic data includes the temperature mean value and electrical measurement value of the brushless motor to be tested; Constructing a regression network, using electrical correction feature data in the electrical correction training set as input data of the regression network, and using electrical correction values ​​in the electrical correction training set as output data of the regression network, training the regression network, and obtaining an initial regression network; The initial regression network is model verified using the electrical correction test set, and the initial regression network that is less than or equal to a preset test error threshold is output as the preconfigured first machine learning model.

6. The intelligent diagnosis system for brushless motor of vehicle water pump according to claim 5, characterized in that: The logic for obtaining the electrical correction value in the historical electrical correction training data is as follows: a1: at the same rotation speed, obtaining a temperature reference value of the first test brushless motor, collecting an electrical measurement value of the first test brushless motor at the temperature reference value, and marking the electrical measurement value of the first test brushless motor at the temperature reference value as a first electrical measurement value; a2: At the same speed, collect the temperature average t of the second test brushless motor, and collect the electrical measurement value of the second test brushless motor at the temperature average t, mark the electrical measurement value of the second test brushless motor at the temperature average t as the second electrical measurement value, and t is an integer greater than zero; a3: Calculate the difference between the temperature reference value and the temperature mean t, and mark the difference between the temperature reference value and the temperature mean t as the temperature difference; a4: Calculate the difference between the first electrical measurement value and the second electrical measurement value, mark the difference between the first electrical measurement value and the second electrical measurement value as an electrical deviation, compare the electrical deviation with a preset electrical deviation threshold, and if the electrical deviation is greater than the preset electrical deviation threshold, sum the second electrical measurement value and the electrical deviation, mark the sum of the second electrical measurement value and the electrical deviation as an electrical correction value, associate and bind the corresponding temperature difference with the electrical correction value, obtain the electrical correction value under a certain temperature difference, and set t=t+k, or set t=tk, and return to step a2; if the electrical deviation is less than or equal to the preset electrical deviation threshold, set t=t+k, or set t=tk and return to step a2, where k is an integer greater than zero; a5: Repeat the above steps a2 to a4 until t=MaxT, or t=MinT, then end the cycle and obtain the electrical correction value at each temperature difference.

7. The intelligent diagnosis system for brushless motor of vehicle water pump according to claim 6, characterized in that: The method of intercepting the abnormal wave band in the actual electrical waveform diagram comprises: Retrieve the standard electrical waveform diagram of the brushless motor to be tested; Using the actual electrical waveform as a matching image and the standard electrical waveform as a template image; Set the step size to 1, and perform cross-difference calculation between the matching image and the template image by sliding window to obtain the difference degree of each overlapping window; The overlapping windows whose difference is less than or equal to the preset difference threshold are regarded as similar regions, and the overlapping windows whose difference is greater than the preset difference threshold are regarded as different regions; A difference region is segmented from the actual electrical waveform diagram, and the electrical waveform in the segmented difference region is used as an abnormal band.

8. The intelligent diagnosis system for brushless motor of vehicle water pump according to claim 7, characterized in that: The generation logic of the preconfigured second machine learning model is as follows: Acquire historical fault diagnosis training data, and divide the historical fault diagnosis training data into a fault diagnosis training set and a fault diagnosis test set, wherein the historical fault diagnosis training data includes abnormal bands in the form of multiple images and their corresponding fault type annotation labels; Construct a classifier, use the abnormal bands in the image form in the fault diagnosis training set as the input data of the classifier, and use the fault type annotation labels in the fault diagnosis training set as the output data of the classifier, train the classifier, and obtain an initial classification network; The initial classification network is model verified using the fault diagnosis test set, and an initial classification network with a test accuracy greater than or equal to a preset threshold is output as a preconfigured second machine learning model.

9. An intelligent diagnosis method for a brushless motor of a vehicle water pump, characterized in that: The method comprises: S201: within a predetermined test time range, collecting a first vibration mean value measured during the operation of the brushless motor to be tested at an rth speed, and obtaining an estimated second vibration mean value of the brushless motor to be tested at an r+vth speed based on the first vibration mean value measured, where r and v are integers greater than zero; S202: collecting the measured second vibration mean value of the brushless motor to be tested at the r+vth speed, judging whether the brushless motor to be tested has abnormal performance according to the estimated second vibration mean value and the measured second vibration mean value, if so, collecting the temperature mean value and electrical measurement value of the brushless motor to be tested at the rth speed; if not, setting v=v+d, and returning to step S201, d is an integer greater than zero; S203: Repeat the above steps S201 to S202 until the brushless motor to be tested shows abnormal performance, or when r=R, the loop is terminated, the temperature reference value is obtained, and the temperature reference value, the temperature mean and the electrical measurement value are input into a pre-configured first machine learning model to obtain the electrical correction value of the brushless motor to be tested within a predetermined test time range, where R is the set maximum speed of the brushless motor to be tested; S204: construct an actual electrical waveform diagram according to the electrical correction value, intercept the abnormal band in the actual electrical waveform diagram, input the abnormal band into the preconfigured second machine learning model in the form of an image, and obtain the fault type of the brushless motor to be tested.

Images