A method for detecting and diagnosing defects in a motor production line

Abstract

The invention discloses a method for detecting and diagnosing defects in a motor production line, which belongs to the technical field of production quality optimization, and specifically comprises the following steps: collecting the rotation speed of a motor when the motor is running, and recording the weight of an object carried by the motor each time; collecting the noise when the motor is working, and filtering the environmental noise to obtain the motor noise; dividing the weight of the object to be carried by the motor into several levels, performing linear fitting on the relationship between the motor noise and the rotation speed and the relationship between the motor noise and the load at different levels, and calculating the fitting degree of the noise and the rotation speed after the linear fitting, as well as the fitting degree of the noise and the load; when the fitting degree of the noise and the load is high, calculating the maximum load of the motor by adjusting the load, and when the fitting degree of the noise and the rotation speed is high, calculating the maximum rotation speed of the motor by adjusting the motor rotation speed; judging the motor whose difference between the maximum load and the maximum rotation speed and the set value is greater than a preset threshold as unqualified.

Description

Technical Field

[0001] The present invention relates to the technical field of production quality optimization, and in particular to a method for detecting and diagnosing defects in a motor production line. Background Art

[0002] With the widespread application of motors in industry and daily life, the stability and reliability of their performance have become particularly important. As the core power source of many devices, the quality of motors directly affects the working efficiency and safe operation of the entire equipment. Therefore, the quality inspection of motors during the production process has become a key step to ensure the performance of motors.

[0003] Traditional motor detection methods mostly rely on manual detection, which is not only inefficient but also easily affected by human factors, making it difficult to ensure the accuracy and consistency of the detection results. In addition, traditional detection methods can often only detect certain specific parameters of the motor, such as resistance, insulation, etc., but cannot fully evaluate the performance of the motor under actual working conditions, such as the noise level under different loads and speeds.

[0004] The noise and load of the motor are one of the important indicators for evaluating its performance. Unqualified noise levels will affect the operating experience of the motor. Therefore, the detection of motor noise and the analysis of related factors will help improve the quality of motor equipment and the production level and quality inspection level of the production line. Summary of the invention

[0005] The purpose of the present invention is to provide a motor production line defect detection and diagnosis method to solve the following technical problems:

[0006] The detection of motor noise and the analysis of related factors will help improve the quality of motor equipment and the production and quality inspection levels of production lines.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A method for detecting and diagnosing defects in a motor production line comprises the following steps:

[0009] Carry out load detection on the motors in the production line, collect the speed of the motors during operation, and record the load weight of the motors each time they are tested;

[0010] The noise of the motor when it is working is collected, and the environmental noise is filtered to obtain the motor noise;

[0011] The weight of the object loaded by the motor is divided into several levels. Linear fitting is performed on the relationship between the motor noise and the speed and the relationship between the motor noise and the load at different levels. The fitting degree R of the noise and the speed after linear fitting is calculated. a 2 , and the goodness of fit R of noise and loadb 2 ;

[0012] When the noise and load fit R b 2 Greater than the fitting degree R of noise and speed a 2 When the weight level is reached, gradually increase the weight of the load until the motor noise is greater than the set standard, obtain the weight of the current load object, and set the load object as the maximum load of the motor;

[0013] When the noise and speed fit R a 2 Greater than the noise-load fit R b 2 When the weight level reaches the limit, gradually increase the motor speed until the motor noise is greater than the set standard, obtain the current motor speed, and set the speed as the maximum speed of the motor;

[0014] The maximum load and maximum speed of each motor tested are compared with the set values, and the motors whose difference with the set values ​​is greater than a preset threshold are judged as unqualified.

[0015] As a further solution of the present invention: the calculation process of the degree of fit is:

[0016] A rectangular coordinate system is established, and linear fitting is performed using the logarithm of the speed or the logarithm of the load as the independent variable and the noise decibel as the dependent variable. After linear fitting, the coordinate values ​​of each point of the linear fitting function are obtained. A number of coordinate points are selected, and the calculation formula for the fitting degree is:

[0017] ;

[0018] Where j is a or b, n is the number of times the load or speed is adjusted at each weight level, i∈[1,n], y i is the ordinate value of the actual collected noise corresponding to any coordinate point in the linear fitting function, is the value of the function ordinate corresponding to any coordinate point in the linear fitting function, It is the average value of the actual noise collected.

[0019] As a further solution of the present invention: when there is a degree of fit R in any weight level a 2 and the fitness R b 2 When the difference is greater than the preset threshold A and less than the preset threshold B, the weight level is divided into several sub-weight levels again, and the relationship between the motor noise and the speed and the relationship between the motor noise and the load under different sub-weight levels are linearly fitted again.

[0020] As a further embodiment of the present invention: when there is a degree of fit R in any weight class or sub-weight class a 2 and the fitness R b 2 When the difference is less than the preset threshold value A, the load object with noise greater than the set value is screened out, and a prompt is given to reduce the motor speed when loading the load object.

[0021] As a further solution of the present invention: the process of filtering the environmental noise is:

[0022] Based on the deep learning model, the model is trained to identify and separate the motor operation noise and environmental noise, and the noise is captured separately through several microphones in different directions, and the noise audio in the opposite direction of the motor position is eliminated.

[0023] As a further solution of the present invention: when the fitting degree R of the noise and the speed is detected a 2 , and the goodness of fit R of noise and load b 2 When both are less than the preset degree of fit, the motor is tested.

[0024] As a further solution of the present invention: collecting quality inspection data of motors in the production line at fixed intervals, marking unqualified speed and unqualified load as key defects, and sampling a number of qualified motors off the motor production line, with a sampling ratio of K, where K is the sum of unqualified rates of key defects;

[0025] Qualified motors are tested for the key defects. Each qualified motor is tested for only one key defect, and the sampling ratio for each key defect is allocated according to the unqualified rate ratio;

[0026] After every m fixed cycles, where m is a positive integer, the proportion k of qualified motors that still have key defects in the total number of inspections is counted, and a proportion sequence k1, k2, ..., km is generated according to the time sequence. If the proportion mean k0>αK, α is the preset target value, α∈(0,1), then the inspection proportion for the next m fixed cycles will still be K; if the proportion mean k0 is less than or equal to αK, then the inspection proportion for the next m fixed cycles will be αK.

[0027] Beneficial effects of the present invention:

[0028] The present invention realizes a comprehensive evaluation of motor performance and improves the accuracy and reliability of motor quality detection through steps such as load detection, noise collection and filtering, and linear fitting. By linearly fitting the relationship between motor noise, speed, and load, the maximum load and maximum speed of the motor can be more accurately judged, thereby avoiding motor damage caused by overload or overspeed. The environmental noise filtering process based on the deep learning model can effectively identify and separate motor operation noise from environmental noise, improve the purity of noise data, and thus improve the accuracy of the test results. The quality inspection data of the motors in the production line is collected at fixed intervals, and key defects are selected as key defects for detection, which helps to improve the quality inspection level of the motor production line in a targeted manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The present invention will be further described below in conjunction with the accompanying drawings.

[0030] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0031] 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.

[0032] See also Figure 1 As shown, the present invention is a method for detecting and diagnosing defects in a motor production line, comprising:

[0033] Load testing: Load testing is performed on the motor on the production line. This step involves collecting the motor's operating data under different loads, including key parameters such as speed, and recording the load weight borne by the motor during each test. This data will be used for subsequent analysis to evaluate the motor's performance under different load conditions.

[0034] Noise collection and filtering: At the same time, the noise generated by the motor when it is working is collected. In order to obtain accurate motor noise data, the environmental noise needs to be effectively filtered to obtain a pure motor noise signal. This step is crucial for subsequent analysis because noise is an important indicator for evaluating motor performance.

[0035] Linear fitting analysis: The collected motor noise data is correlated with its corresponding load weight and speed. Specifically, the weight of the object carried by the motor is divided into several levels, that is, several intervals, and then the relationship between the motor noise and the speed change, and the relationship between the motor noise and the load change in these different intervals are linearly fitted. The interval range cannot be too large. For example, the interval [100kg, 300kg] starts with 100kg, keeps the motor as the base speed, and gradually increases the load weight in the interval until it reaches 300kg. Each time the noise value is increased, the noise value is recorded; similarly, keep 100kg unchanged, gradually increase the motor speed, and record the noise value each time it is increased. The number of times the load is increased is the same as the number of times the speed is increased. Through this fitting, the degree of fit between noise and speed and the degree of fit between noise and load can be calculated.

[0036] Determination of maximum load and maximum speed: Based on the above fitting results, when the noise-to-load fit is greater than the noise-to-speed fit, gradually increase the weight of the load until the motor noise exceeds the set standard value. At this time, record the weight of the current load object and set it as the maximum load of the motor. On the contrary, if the noise-to-speed fit is greater than the noise-to-load fit, gradually increase the motor speed until the motor noise also exceeds the set standard. At this time, record the current speed and set it as the maximum speed of the motor.

[0037] Qualification determination: Finally, the maximum load and maximum speed of each motor are compared with the preset standard values. If the difference between the maximum load or maximum speed of a motor and the set value is greater than the preset threshold, the motor will be judged as unqualified and requires further inspection or repair.

[0038] In another preferred embodiment of the present invention, the calculation process of the degree of fit is:

[0039] A rectangular coordinate system is established, and linear fitting is performed using the logarithm of the speed or the logarithm of the load as the independent variable and the noise decibel as the dependent variable. After linear fitting, the coordinate values ​​of each point of the linear fitting function are obtained. A number of coordinate points are selected, and the calculation formula for the fitting degree is:

[0040] ;

[0041] Where j is a or b, n is the number of times the load or speed is adjusted at each weight level, i∈[1,n], y i is the ordinate value of the actual collected noise corresponding to any coordinate point in the linear fitting function, is the value of the function ordinate corresponding to any coordinate point in the linear fitting function, It is the average value of the actual noise collected.

[0042] In another preferred embodiment of the present invention, when there is a degree of fit R in any weight class a 2 and the fitness R b 2 When the difference is greater than the preset threshold A and less than the preset threshold B, this indicates that the linear fit between the relationship between the motor noise and the speed and the relationship between the noise and the load at this weight level is quite different, but has not yet reached the level where immediate corrective measures need to be taken. At this point, in order to improve the accuracy and meticulousness of the diagnosis, the weight level can be further refined, that is, divided into several sub-weight levels. Next, for these different sub-weight levels, the linear fit analysis of the relationship between the motor noise and the speed, and the relationship between the motor noise and the load is re-performed.

[0043] This detailed analysis can more accurately capture the performance trend of the motor under different load conditions, thus providing a more accurate basis for the subsequent determination of the maximum load and maximum speed. At the same time, it also helps to discover subtle problems that may be overlooked in the initial analysis, further improving the effectiveness and reliability of the detection and diagnosis method.

[0044] In another preferred embodiment of the present invention, when there is a degree of fit R in any weight class or sub-weight class a 2 and the fitness R b 2 When the difference is less than the preset threshold A, this usually means that at this level or sub-level, the linear fit between the relationship between motor noise and speed and the relationship between noise and load is relatively close, indicating that the performance of the motor under this load condition is relatively stable. However, if a load object with noise greater than the set standard is still detected in this case, this may be due to the special properties of the load object or the abnormal performance of the motor at a specific working point.

[0045] At this time, the system will screen out these load objects with excessive noise and give prompts, suggesting that the motor speed should be appropriately reduced when loading these objects. Such prompts not only help protect the motor from overload damage and extend its service life, but also ensure the stability and reliability of the motor in actual work. In this way, the embodiments of the present invention not only improve the defect detection efficiency on the motor production line, but also further enhance the safety and reliability of the motor during use.

[0046] In another preferred embodiment of the present invention, the process of environmental noise filtering uses advanced deep learning technology. First, a deep learning model is trained to identify and separate the noise generated by the motor running from the background noise in the environment. This model can learn and understand the characteristics of the motor noise, so as to accurately extract the working sound of the motor from the complex environmental noise.

[0047] In order to improve the accuracy and comprehensiveness of noise collection, the present invention also uses several microphones in different directions to capture noise. These microphones are arranged at different positions around the motor to record noise data in all directions. In this way, it can ensure that the noise signal of the motor is captured from multiple angles, providing more abundant data support for subsequent noise analysis.

[0048] After collecting noise data from multiple directions, the present invention further adopts an innovative processing method, which is to remove the noise audio in the opposite direction of the motor position. This is because the noise in the opposite direction of the motor position often contains more environmental noise components and less working noise of the motor itself. By removing this part of the noise, the accuracy and purity of the motor noise data can be further improved.

[0049] In another preferred embodiment of the present invention, when it is detected that the fit between noise and speed and the fit between noise and load are both less than a preset fit threshold, this indicates that under the current test conditions, the relationship between noise and speed and load of the motor does not conform to the expected linear relationship. This may mean that there is some potential defect or abnormality in the motor, which requires further detection and diagnosis.

[0050] Therefore, when this happens, the present invention will perform a more detailed inspection of the motor. This may include a careful inspection of the various components of the motor to find possible faults or wear; testing the electrical parameters of the motor to ensure that they meet the design specifications; and may even include disassembling and reassembling the motor to gain a deeper understanding of its internal structure and working principle.

[0051] In another preferred embodiment of the present invention, in order to more efficiently monitor and manage the quality of the motor production line, a random inspection mechanism based on data drive and dynamic adjustment is adopted.

[0052] Regular quality inspection data collection:

[0053] At regular intervals (e.g., every shift, every day, or every week), the quality inspection data of all motors in the production line is automatically collected.

[0054] These data include but are not limited to key performance indicators such as motor speed, load, temperature, and vibration.

[0055] Key defect marking:

[0056] In the quality inspection data, special attention is paid to the two types of defects, unqualified speed and unqualified load, and they are marked as key defects.

[0057] These two types of defects usually have a significant impact on the performance and life of the motor, so they require special attention.

[0058] Qualified motor sampling inspection:

[0059] From the qualified motors coming off the motor production line, a certain proportion K is sampled for inspection.

[0060] This ratio K is determined based on the sum of the failure rates of key defects to ensure the targetedness and effectiveness of the random inspection.

[0061] Key defect detection:

[0062] For each qualified motor selected for random inspection, specific inspection items will be arranged according to the corresponding key defect type. It is worth noting that each motor is only inspected for one key defect, which not only reduces unnecessary duplication of work, but also ensures the pertinence and effectiveness of the inspection process. At the same time, the specific sampling ratio of each key defect will also be dynamically allocated according to its proportion in the total problem rate to ensure the rational use of resources.

[0063] Data statistics and analysis:

[0064] After every m fixed cycles (m is a positive integer), the proportion of qualified motors that still have key defects to the total number of random inspections is counted and recorded as k.

[0065] Generate a proportional sequence k1, k2, ..., km according to the time sequence in order to analyze the quality change trend of the production process.

[0066] Adjustment of sampling ratio:

[0067] Calculate the mean value of the proportion k0 and compare it with the preset target value αK, where α is a decimal value between 0 and 1, representing the target failure rate after quality inspection.

[0068] If k0 is greater than αK, it means that the current sampling ratio is high enough to effectively find key defects, so the sampling ratio for the next m fixed cycles remains at K.

[0069] If k0 is less than or equal to αK, it means that the current sampling ratio may be too high or too low and needs to be adjusted. At this time, the sampling ratio of the next m fixed cycles is adjusted to αK to allocate detection resources more reasonably.

[0070] In order to continuously optimize this process, the present invention will count the proportion k of key defects still existing in all qualified motors participating in the random inspection after every m fixed cycles, and record the results of these n times in chronological order to form a proportion sequence k1, k2, ..., km. By averaging this sequence to obtain k0, if it is found that k0 is greater than the preset target value αK (where α is a preset threshold), it means that the current random inspection strategy may not be strict or effective enough, and it is necessary to continue to maintain a high random inspection ratio K; on the contrary, if k0 is less than or equal to αK, it means that the existing random inspection intensity is sufficient to deal with the current quality problems, so the random inspection ratio in the next m cycles can be appropriately reduced to αK, so as to achieve the purpose of cost saving without affecting the effect of product quality control.

[0071] 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 application 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 may be 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 by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains 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.

[0072] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein 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 this application.

[0073] 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 modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0074] In the several embodiments provided in the present application, 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 modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules 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 modules, which can be electrical, mechanical or other forms.

[0075] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0076] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0077] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0078] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0079] 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. A method for detecting and diagnosing defects in a motor production line, characterized in that: The following steps are involved: Carry out load detection on the motors in the production line, collect the speed of the motors during operation, and record the load weight of the motors each time they are tested; The noise of the motor when it is working is collected, and the environmental noise is filtered to obtain the motor noise; The weight of the object loaded by the motor is divided into several levels. Linear fitting is performed on the relationship between the motor noise and the speed and the relationship between the motor noise and the load at different levels. The fitting degree R of the noise and the speed after linear fitting is calculated. a 2 , and the goodness of fit R of noise and load b 2 ; When the noise and load fit R b 2 Greater than the fitting degree R of noise and speed a 2 When the weight level is reached, gradually increase the weight of the load until the motor noise is greater than the set standard, obtain the weight of the current load object, and set the load object as the maximum load of the motor; When the noise and speed fit R a 2 Greater than the noise-load fit R b 2 When the weight level reaches the limit, gradually increase the motor speed until the motor noise is greater than the set standard, obtain the current motor speed, and set the speed as the maximum speed of the motor; The maximum load and maximum speed of each motor tested are compared with the set values, and the motors whose difference with the set values ​​is greater than a preset threshold are judged as unqualified; The calculation process of the goodness of fit is: A rectangular coordinate system is established, and linear fitting is performed using the logarithm of the speed or the logarithm of the load as the independent variable and the noise decibel as the dependent variable. After linear fitting, the coordinate values ​​of each point of the linear fitting function are obtained. A number of coordinate points are selected, and the calculation formula for the fitting degree is: ； Where j is a or b, n is the number of times the load or speed is adjusted at each weight level, i∈[1,n], y i is the ordinate value of the actual collected noise corresponding to any coordinate point in the linear fitting function, is the value of the function ordinate corresponding to any coordinate point in the linear fitting function, It is the average value of the actual noise collected.

2. A motor production line defect detection and diagnosis method according to claim 1, characterized in that: When there is a goodness of fit R in any weight level a 2 and the fitness R b 2 When the difference is greater than the preset threshold A and less than the preset threshold B, the weight level is divided into several sub-weight levels again, and the relationship between the motor noise and the speed and the relationship between the motor noise and the load under different sub-weight levels are linearly fitted again.

3. A motor production line defect detection and diagnosis method according to claim 2, characterized in that: When there is a good fit R in any weight class or sub-weight class a 2 and the fitness R b 2 When the difference is less than the preset threshold value A, the load object with noise greater than the set value is screened out, and a prompt is given to reduce the motor speed when loading the load object.

4. A motor production line defect detection and diagnosis method according to claim 1, characterized in that: The process of filtering the environmental noise is as follows: Based on the deep learning model, the model is trained to identify and separate the motor operation noise and environmental noise, and the noise is captured separately through several microphones in different directions, and the noise audio in the opposite direction of the motor position is eliminated.

5. The motor production line defect detection and diagnosis method according to claim 1, characterized in that: When the noise and speed fit R is detected a 2 , and the goodness of fit R of noise and load b 2 When both are less than the preset degree of fit, the motor is tested.

6. A motor production line defect detection and diagnosis method according to claim 1, characterized in that: The quality inspection data of the motors in the production line are collected at fixed intervals, and unqualified speed and unqualified load are marked as key defects. Several qualified motors off the motor production line are randomly inspected, and the sampling ratio is K, where K is the sum of the unqualified rates of key defects; Qualified motors are tested for the key defects. Each qualified motor is tested for only one key defect, and the sampling ratio for each key defect is allocated according to the unqualified rate ratio; After every m fixed cycles, where m is a positive integer, the proportion k of qualified motors that still have key defects in the total number of inspections is counted, and a proportion sequence k1, k2, ..., km is generated according to the time sequence. If the proportion mean k0>αK, α is the preset target value, α∈(0,1), then the inspection proportion for the next m fixed cycles will still be K; if the proportion mean k0 is less than or equal to αK, then the inspection proportion for the next m fixed cycles will be αK.

Images