Connection process quality detection method, device and system

Abstract

The invention discloses a method, device and system for detecting the quality of a connection process, and the method comprises the steps: obtaining pressure data which is used for reflecting the pressure between a plate and a rivet when the riveting operation is executed on the plate, determining the qualification probability that the riveting operation is qualified according to the pressure data and a neural network model, and carrying out the detection of the quality of the connection process. Therefore, the operation quality of the riveting operation is determined based on the qualification probability, the qualification probability of the riveting operation can be determined through the neural network model, the situation of manual sampling inspection is reduced, the situations of connection defect missing inspection and the like are reduced, and the quality of self-piercing riveting is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of vehicle manufacturing, and in particular, relates to a method, device and system for detecting the quality of a connection process. Background Art

[0002] Self-piercing riveting is a cold connection technology that uses special rivets (π-shaped nails) to penetrate the top sheet. Under the action of the riveting die, the hollow structure at the tail of the rivet will expand and pierce into the bottom sheet to form a strong riveting point, thereby connecting two or more metal sheets together.

[0003] At present, in order to ensure the quality of self-piercing riveting, it is necessary to test the connection quality of self-piercing riveting through manual sampling, so as to recycle the plates with connection defects. However, due to the low frequency of manual sampling and the reliance on the professional level of sampling personnel in assessing connection defects, connection defects may be missed, which reduces the quality of self-piercing riveting. Summary of the invention

[0004] The embodiments of the present application provide a method, device and system for detecting the quality of connection process, aiming to solve the problem of missed detection of connection defects when detecting the quality of self-piercing rivets by manual sampling, thereby reducing the quality problem of self-piercing rivets.

[0005] In a first aspect, an embodiment of the present application provides a method for detecting connection process quality, the method comprising:

[0006] Acquiring pressure data; wherein the pressure data is used to reflect the pressure between the sheet and the rivet when the sheet is riveted;

[0007] According to the pressure data and the neural network model, the probability that the riveting operation is a qualified operation is determined, so as to determine the operation quality of the riveting operation based on the qualified probability; the neural network model is obtained by training a plurality of sample data, the sample data comprising positive sample data and negative sample data, the positive sample data comprising the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is a qualified operation, and the negative sample data comprising the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is an unqualified operation.

[0008] In a possible implementation manner of the first aspect above, the method further includes:

[0009] Determining a target operating station where the sheet material is located when performing the riveting operation on the sheet material;

[0010] Wherein, determining the probability that the riveting operation is a qualified operation according to the pressure data and the neural network model includes:

[0011] The pressure data is input into the neural network model corresponding to the target operating station to obtain the probability that the riveting operation is a qualified operation; wherein different operating stations correspond to different neural network models, and the neural network model corresponding to any operating station is used to determine the probability that the riveting operation performed on the sheet material at the operating station is a qualified operation.

[0012] In a possible implementation of the first aspect, before inputting the pressure data into the neural network model corresponding to the target operation station to obtain the probability that the riveting operation is a qualified operation, the method further includes:

[0013] Determine a target event topic corresponding to the riveting operation, and record the pressure data of the riveting operation in the target event topic;

[0014] The step of inputting the pressure data into the neural network model corresponding to the target operation station to obtain the probability that the riveting operation is a qualified operation includes:

[0015] Determine a target thread that subscribes to the target event topic; wherein the target thread is a thread that transmits data to a neural network model corresponding to the target operation station, and different threads are used to transmit different data to corresponding neural network models;

[0016] The pressure data in the target event topic is transmitted to the neural network model corresponding to the target operation station through the target thread to obtain the qualified probability that the riveting operation is a qualified operation.

[0017] In a possible implementation manner of the first aspect, before determining the probability that the riveting operation is a qualified operation according to the pressure data and the neural network model, the method further includes:

[0018] generating a data sequence corresponding to the unqualified operation according to the pressure data of the unqualified operation;

[0019] The first data segment in the data sequence is processed based on a first random number to obtain a processed first data segment; wherein the first random number belongs to a first random number range,

[0020] Divide a second data segment in the data sequence into at least two data segments; wherein the average value of the second data segment is greater than the average value of the first data segment,

[0021] Processing the corresponding data segment based on the second random number corresponding to each of the at least two data segments to obtain a processed second data segment; wherein the second random number belongs to a second random number range, and the second random number range is larger than the first random number range;

[0022] Generate processed negative sample data according to the processed first data segment and the processed second data segment, so as to train the neural network model based on the processed negative sample data.

[0023] In a possible implementation manner of the first aspect, before processing the first data segment in the data sequence based on the first random number to obtain the processed first data segment, the method further includes:

[0024] Determine a target abnormal time period corresponding to the unqualified operation; wherein the target abnormal time period is a concentrated time period during which abnormalities occur in the unqualified operation;

[0025] Based on the target abnormal time period, the data sequence is divided into a first data segment and a second data segment; wherein the second data segment is a concentrated time period when abnormalities occur.

[0026] In a possible implementation manner of the first aspect above, the method further includes:

[0027] Determine the pressure change trend of the riveting operation according to the pressure data; wherein the pressure change trend is used to reflect the pressure change between the sheet material and the rivet when the riveting operation is performed on the sheet material within a time period;

[0028] Wherein, determining the operation quality of the riveting operation based on the qualified probability includes:

[0029] The operation quality of the riveting operation is determined according to the pressure variation trend and the qualified probability.

[0030] In a possible implementation manner of the first aspect above, the method further includes:

[0031] Determining whether the riveting operation is a target riveting operation according to the execution position of the riveting operation on the sheet material; wherein the target riveting operation is a riveting operation having a preset change feature in the pressure change trend;

[0032] In a case where the riveting operation is a target riveting operation, using the neural network model to identify a target feature in the pressure change trend;

[0033] Determining the operation quality of the riveting operation according to the pressure change trend and the probability includes:

[0034] The operation quality of the riveting operation is determined according to the target feature, the pressure change trend and the qualified probability.

[0035] In a possible implementation manner of the first aspect above, the method further includes:

[0036] Displaying the pressure variation trend;

[0037] Wherein, determining the operation quality of the riveting operation based on the qualified probability includes:

[0038] detecting an evaluation operation for the pressure change trend;

[0039] In response to the evaluating operation, determining quality evaluation information for the riveting operation;

[0040] The operation quality of the riveting operation is determined according to the quality assessment information and the pass probability.

[0041] In a second aspect, a device for detecting connection process quality is provided, the device comprising:

[0042] An acquisition module, used for acquiring pressure data; wherein the pressure data is used for reflecting the pressure between the sheet material and the rivet when the riveting operation is performed on the sheet material;

[0043] A determination module is used to determine the probability that the riveting operation is a qualified operation based on the pressure data and the neural network model, so as to determine the operation quality of the riveting operation based on the qualified probability; the neural network model is trained by multiple sample data, and the sample data includes positive sample data and negative sample data, the positive sample data includes the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is a qualified operation, and the negative sample data includes the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is an unqualified operation.

[0044] In a third aspect, an embodiment of the present application provides a connection process quality detection system, the system comprising:

[0045] A data acquisition module, used to collect pressure data between the rivet and the sheet material during the riveting operation performed at each operating station; wherein the riveting operation is an operation of riveting the sheet material;

[0046] The quality detection module is used to distribute the pressure data to the neural network model corresponding to the operation station, obtain the output result of the neural network model, and generate the operation quality of the riveting operation performed at the operation station based on the output result.

[0047] In a fourth aspect, an embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for detecting the connection process quality provided in the first aspect above is implemented.

[0048] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for detecting the quality of the connection process provided in the first aspect above is implemented.

[0049] In a sixth aspect, an embodiment of the present application provides a computer program product, including a computer program. When the computer program is run on a computer, the computer executes the method for detecting the connection process quality provided in the first aspect.

[0050] It can be understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0051] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0052] In an embodiment of the present application, pressure data is obtained, and the pressure data is used to reflect the pressure between the sheet material and the rivet when the riveting operation is performed on the sheet material. According to the pressure data and the neural network model, the probability that the riveting operation is a qualified operation is determined, so as to determine the operation quality of the riveting operation based on the qualified probability, and then the qualified probability of the riveting operation can be determined through the neural network model, thereby reducing the situation of manual sampling, thereby reducing the missed detection of connection defects, etc., and improving the quality of self-piercing riveting. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of the steps of a method for detecting the quality of a connection process provided by an embodiment of the present application;

[0054] Figure 2 It is a structural schematic diagram of a neural network model provided by an embodiment of the present application;

[0055] Figure 3 It is a flowchart of another method for detecting the quality of a connection process provided by an embodiment of the present application;

[0056] Figure 4 is a schematic diagram of a pressure change trend provided by an embodiment of the present application;

[0057] Figure 5 It is a structural schematic diagram of a connection process quality detection system provided by an embodiment of the present application;

[0058] Figure 6 It is a flow chart of a method for detecting the quality of a connection process provided by an embodiment of the present application;

[0059] Figure 7 It is a structural schematic diagram of a connection process quality detection device provided by an embodiment of the present application;

[0060] Figure 8 It is a structural block diagram of a terminal device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0062] Self-piercing riveting is a cold connection technology. Special rivets (π-shaped nails) are inserted into and penetrate the top layer of the sheet. Under the action of the riveting die, the hollow structure at the tail of the rivet will expand and penetrate into the bottom layer of the sheet to form a strong riveting point, thereby connecting two or more metal sheets together. It is widely used in the field of vehicle manufacturing, especially in the connection process of automobile bodies, such as the connection of body frames, doors, hoods, trunks, seats and other parts.

[0063] However, in the actual riveting process, the riveting process, riveting equipment, and the quality of the rivets may cause connection defects, such as rivet deviation from the riveting position, rivet distortion, rivet cracks, and other connection defects. Specifically, the connection defects may also include defects that affect the connection strength and defects that affect the appearance of the connection. Defects that affect the connection strength may include cracks in the sheet and / or rivets after the riveting operation, deformation of the sheet and / or rivets after the riveting operation and the size change after deformation is greater than a preset ratio, and the distance that the rivet deviates from the riveting position is greater than a preset distance. Defects that affect the appearance of the connection may include scratches on the rivet or sheet, deformation of the sheet and / or rivet after the riveting operation and the size change after deformation is less than or equal to a preset ratio, and the distance that the rivet deviates from the riveting position is less than or equal to a preset distance.

[0064] Among them, defects that affect the connection strength will affect the connection strength between each sheet, making the connected sheets easily affected by external forces and disconnected, that is, after two or more sheets are connected together, they are easily affected by external forces and the riveted parts between each sheet become loose or broken, causing the connection between each sheet to be disconnected. Defects that affect the appearance of the connection will not affect the connection strength between each sheet or the impact is less than a predetermined range, making it difficult for the connected sheets to be disconnected by external forces, and only affect the appearance of the connected sheets.

[0065] At present, in order to ensure the quality of self-piercing riveting, it is necessary to test the connection quality of self-piercing riveting by manual sampling, so as to recycle the plates with connection defects. Usually, the manual sampling method is to extract one or more plates from all the plates that have undergone self-piercing riveting, and manually chisel the riveted parts of the extracted plates, so that the sampling personnel can observe the cross-section of the chiseled parts to evaluate the riveted connection quality of the riveted parts, so that when the riveted connection quality of the plate is evaluated as unqualified, it can be determined that the riveted plates in this batch have riveted quality problems, and the plates in this batch can be recycled for further testing and processing to ensure the connection quality of self-piercing riveting.

[0066] However, due to the low frequency of manual sampling and the reliance on the professional level of the sampling personnel when evaluating the quality of riveted connections, there is a possibility that connection defects are missed, which reduces the quality of self-piercing riveting.

[0067] Based on this, the present application provides a method for detecting the quality of a connection process, by acquiring pressure data, which is used to reflect the pressure between the sheet and the rivet when the riveting operation is performed on the sheet. According to the pressure data and the neural network model, the probability that the riveting operation is a qualified operation is determined, so as to determine the operation quality of the riveting operation based on the qualified probability, and then the qualified probability of the riveting operation can be determined through the neural network model, thereby reducing the situation of manual sampling, thereby reducing the situation of missed detection of connection defects, etc., and improving the quality of self-piercing riveting.

[0068] See also Figure 1 , Figure 1 A flow chart of a method for detecting the quality of a connection process provided by an embodiment of the present application is shown, which may specifically include the following steps:

[0069] Step 101, obtaining pressure data.

[0070] Among them, the pressure data can be used to reflect the pressure between the sheet and the rivet when the riveting operation is performed on the sheet. The riveting operation can be an operation of riveting the sheet, and the pressure between the rivet and the sheet corresponds to the displacement of the rivet inserted into the sheet. Different displacements can correspond to different pressures. The sheet can be a plate that needs to be riveted.

[0071] In practical applications, riveting equipment can be used to perform riveting operations on different sheet materials. Specifically, the riveting equipment can be used to apply pressure to the rivet so that the rivet penetrates the top sheet material, and under the action of the riveting die, the hollow structure at the tail of the rivet will expand and penetrate the sheet material to form a firm riveting point, thereby completing the riveting operation on the sheet material. The riveting equipment can record the pressure between the rivet and the sheet material during the riveting process, as well as the displacement of the rivet inserted into the sheet material, and then establish a corresponding relationship between the pressure between the rivet and the sheet material and the displacement of the rivet inserted into the sheet material, so that the pressure data of each riveting operation can be recorded.

[0072] In a specific implementation, the pressure data recorded in the riveting equipment can be obtained through the terminal device.

[0073] Step 102, determining the probability that the riveting operation is a qualified operation according to the pressure data and the neural network model, so as to determine the operation quality of the riveting operation based on the qualified probability.

[0074] Among them, the neural network model can be used to determine the probability of the riveting operation being a qualified operation according to the pressure data, and the qualified riveting connection quality can indicate that there is no connection defect that affects the connection strength at the riveted part. The neural network model can be obtained by training with a plurality of sample data, and can be obtained by training with sample data of each operating station. The operating station can be a station for performing a riveting operation on a sheet material, and each operating station can be deployed with a riveting device for performing a riveting operation. The sample data can be the pressure data between the rivet and the sheet material when the riveting operation is performed, and the sample data can include positive sample data and negative sample data. The positive sample data can include the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is a qualified operation, and can specifically include at least one pressure data of a qualified operation, and the negative sample data can include the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is an unqualified operation, and can specifically include at least one pressure data corresponding to an unqualified operation. A qualified operation may be a riveting operation with qualified riveting connection quality, and an unqualified operation may be a riveting operation with unqualified riveting connection quality. A qualified riveting connection quality may indicate that there are no connection defects that affect the connection strength at the riveting part, and an unqualified riveting connection quality may indicate that there are connection defects that affect the connection strength at the riveting part. The operation quality may be information for evaluating the quality of the riveting operation, and specifically may be information for evaluating whether the riveting operation will produce connection defects that affect the connection strength.

[0075] After obtaining the pressure data of the riveting operation, the operating station that performs the riveting operation can be determined, as well as the neural network model corresponding to the operating station, and the pressure data of the riveting operation can be input into the neural network model. The neural network model can then output the corresponding qualified probability based on the input pressure data, that is, the qualified probability that the riveting operation is a qualified operation, thereby being able to evaluate the operational quality of the riveting operation based on the qualified probability that the riveting operation is a qualified operation.

[0076] In practical applications, the qualified probability output by the neural network model can be compared with a pre-set qualified probability threshold, and then when the qualified probability output by the neural network model is greater than or equal to the qualified probability threshold, the riveting operation can be determined to be a qualified operation, and the operating quality of the riveting operation can be determined to be that which will not produce connection defects that affect the connection strength, that is, the riveting operation does not have a riveting quality problem, and when the qualified probability output by the neural network model is less than the qualified probability threshold, it can be determined that the riveting operation is an unqualified operation, and the operating quality of the riveting operation can be determined to be that which will produce connection defects that affect the connection strength, that is, the riveting operation has a riveting quality problem.

[0077] In a specific implementation, before determining the neural network model, a neural network model for determining the probability that the riveting operation is a qualified operation can be pre-constructed. Specifically, the constructed neural network model can be a convolutional neural network model.

[0078] See also Figure 2 , Figure 2 A schematic diagram of the structure of a neural network model provided by an embodiment of the present application is shown. Figure 2 As shown, the neural network model 2 is composed of two convolutional layers 21, two maximum pooling layers 22 and a fully connected layer 23, and each convolutional layer 21 is connected to a maximum pooling layer 22. Among them, the convolutional layer 21 can be used to perform convolution operations on the input data to identify the features in the input data, the maximum pooling layer 22 can be used to remove the features with low correlation in the input data, and the fully connected layer 23 can output the corresponding probability information based on the features with high correlation, that is, output the qualified probability that the riveting operation is a qualified operation.

[0079] After constructing the neural network model, positive sample data and negative sample data can be collected, and then the neural network model can be trained using the positive sample data and the negative sample data, and the model parameters of the neural network model can be updated based on the training results to perform an operation to determine the probability that the riveting operation is a qualified operation based on the trained neural network model.

[0080] In practical applications, after the sample data is collected, the sample data can be preprocessed, that is, each sample data is sorted from small to large according to the displacement corresponding to the pressure data to obtain a data sequence. It should be understood that since the riveting equipment will record the pressure between the rivet and the sheet material and the displacement of the rivet inserted into the sheet material when performing the riveting operation, it is possible to establish a corresponding relationship between the pressure between the rivet and the sheet material and the displacement of the rivet inserted into the sheet material, that is, each pressure data can correspond to the displacement of the rivet inserted into the sheet material, and for the sample data, the pressure data can be sorted in order from small to large based on the displacement corresponding to each pressure data of the sample data to obtain a data sequence for each sample data.

[0081] After obtaining the data sequence, the data sequence can be divided into at least two data segments, and a sample matrix can be constructed based on the divided data segments, that is, a matrix composed of each data segment is obtained, and then the sample matrix can be input into the neural network model to train the neural network model.

[0082] In practical applications, after obtaining the data sequence of each sample data, that is, after obtaining the data sequence of each positive sample data and the data sequence of each negative sample data, for each data sequence, the data sequence can be divided according to a predetermined number of data to obtain at least two data segments, and a sample matrix is ​​generated based on each data segment, thereby obtaining a sample matrix corresponding to each data sequence.

[0083] In a specific implementation, since the sample data may be pressure data between the rivet and the sheet material when performing a riveting operation, the sample matrix corresponding to each sample data may be represented as a sample matrix corresponding to each riveting operation.

[0084] For example, data sequence a may consist of 256 pressure data, and the predetermined number of data may be 16. Data sequence a may then be divided according to the 16 data numbers to obtain 16 data segments, and each data segment may include 16 pressure data. The 16 data segments may be stacked to generate a 16*16 sample matrix.

[0085] After obtaining the sample matrix corresponding to each data sequence, for each sample matrix, the sample matrix can be input into the neural network model, and through the operations of the convolution layer, the maximum pooling layer and the fully connected layer in the neural network model in turn, the probability that the riveting operation corresponding to the sample matrix is ​​a qualified operation is output, or the probability that the riveting operation corresponding to the sample matrix is ​​an unqualified operation is output, and then the model parameters of the neural network model can be adjusted based on the output probability, so that the trained neural network model can determine the qualified probability that the riveting operation is a qualified operation.

[0086] In practical applications, since the position where each operating station performs the riveting operation may be different, and the different positions where the riveting operation is performed may result in different sheet thicknesses, different sheet materials, different sheet structures, etc., which will lead to different pressures between the sheet and the rivet when performing the riveting operation at different positions, that is, the pressures of qualified operations corresponding to different operating stations will also be different. In addition, the riveting operation may include pre-positioning, holding, puncturing, extrusion, forming and other operations, and the pressures between the rivet and the sheet are different in each operation of the riveting operation performed at different positions. If the same neural network model is used to determine the qualified probability of the riveting operation of all operating stations, there will be a situation where the accuracy is low or even cannot be determined. Then, a corresponding neural network model can be constructed for each operating station, and different operating stations can correspond to different neural network models. For each operating station, positive sample data and negative sample data in the operating station can be collected, so as to train the neural network model corresponding to the operating station based on the positive sample data and negative sample data of the operating station, thereby obtaining a neural network model corresponding to each operating station and trained.

[0087] In an embodiment of the present application, before step 102, the following steps may also be included:

[0088] According to the pressure data of the unqualified operation, a data sequence corresponding to the unqualified operation is generated, the first data segment in the data sequence is processed based on the first random number to obtain the processed first data segment, the second data segment in the data sequence is divided into at least two data segments, the corresponding data segment is processed based on the second random number corresponding to each data segment in the at least two data segments to obtain the processed second data segment, and processed negative sample data is generated according to the processed first data segment and the processed second data segment, so as to train the neural network model based on the processed negative sample data.

[0089] Among them, the first random number may belong to a first random number range, the second random number may belong to a second random number range, the first random number range and the second random number range may be ranges for generating random numbers, and the first random number range is different from the second random number range, the first random number range and the second random number range may be determined based on the value of sample data in the operating station, the second random number range is larger than the first random number range, and the average value of the second data segment is larger than the average value of the first data segment.

[0090] In practical applications, after preprocessing the sample data, data enhancement operations can also be performed on the negative sample data.

[0091] Since the sample data can be obtained from actual production, the pressure data in each riveting operation will be recorded, and it will be determined whether each riveting operation is a qualified operation. If the riveting operation is a qualified operation, the pressure data of the riveting operation is determined to be positive sample data, and if the riveting operation is an unqualified operation, the pressure data of the riveting operation is determined to be negative sample data. It should be understood that since the number of negative sample data is small in actual production, usually accounting for one thousandth of the total number of samples, in order to improve the training efficiency when training the neural network model and improve the accuracy of the trained neural network model, it is necessary to perform data enhancement operations on the negative sample data.

[0092] Specifically, since the negative sample data may include pressure data corresponding to at least one unqualified operation, for each unqualified operation, the pressure data in the unqualified operation may be sorted according to the displacement size corresponding to the pressure data to generate a data sequence corresponding to the unqualified operation, and then the data sequence may be divided into a first data segment and a second data segment.

[0093] In an embodiment of the present application, before processing the first data segment in the data sequence based on the first random number, the following steps may also be included:

[0094] A target abnormal time period corresponding to the unqualified operation is determined, and based on the target abnormal time period, the data sequence is divided into a first data segment and a second data segment.

[0095] The second data segment may be a concentrated period of time when anomalies occur, the first data segment may be a data segment in the data sequence other than the second data segment, and the target abnormal period may be a concentrated period of time when anomalies occur in unqualified operations.

[0096] After the data sequence is obtained, the target abnormal period in the unqualified operation can be determined.

[0097] In practical applications, the pressure data corresponding to the unqualified operations of each operating station in the past period of time can be obtained in advance, and for each operating station, the time period when the abnormality occurs in each unqualified operation corresponding to the operating station can be determined and counted, and then according to the time period when the abnormality occurs in each unqualified operation, the concentrated time period when the abnormality occurs when the riveting operation is performed in the operating station can be determined, that is, the time period with the largest proportion of abnormalities among all unqualified operations is determined as the target abnormal time period, and then the target abnormal time period corresponding to each operating station can be determined.

[0098] In a specific implementation, the operating station corresponding to each negative sample data can be determined, and the target abnormal time period corresponding to the operating station can be determined, that is, the target abnormal time period corresponding to the unqualified operation, and then the data segment in the target abnormal time period can be determined from the unqualified operation as the second data segment, and the data segment not in the target abnormal time period can be determined as the first data segment.

[0099] Exemplarily, since the riveting operation may include operations such as pre-positioning, holding, puncturing, extruding, and forming, that is, the riveting operation may include the pressure data between the rivet and the sheet material in each operation, thus for the unqualified operations performed by each operating station in the past period of time, the abnormal operation in each unqualified operation of the operating station can be determined, and then the abnormal operations can be counted from the operations in which the abnormality occurs in each unqualified operation, that is, the operation with the largest number of abnormalities is determined as the abnormal operation, and the time period in which the abnormal operation is located is determined to be the concentrated time period in which abnormalities occur when performing the riveting operation in the operating station.

[0100] For example, operating station a has performed 100 unqualified operations in the past 30 days, among which the puncture operation has 90 abnormal times, the extrusion operation has 8 abnormal times, and the molding operation has 2 abnormal times. It can be determined that the operation with the most abnormal times is the puncture operation, and the puncture operation is determined to be the abnormal operation of operating station a, and the time period of the puncture operation is determined to be the target abnormal time period corresponding to operating station a.

[0101] After the first data segment and the second data segment are determined, a first random number range corresponding to the first data segment and a second random number range corresponding to the second data segment may be determined.

[0102] In practical applications, the first random number range can be determined based on the maximum value of all data in the first data segment, and similarly, the second random number range can be determined based on the maximum value in the second data segment. Generally, at least two first random number ranges can be predefined, and the corresponding relationship between each first random number range and any value can be determined, so that after determining the maximum value of all data in the first data segment, the corresponding first random number range can be determined based on the maximum value of all data in the first data segment, and the second random number can be determined in the same manner.

[0103] After determining the first random number range and the second random number range, a first random number may be randomly determined from the first random number range, and the first data segment may be processed based on the first random number to obtain a processed first data segment.

[0104] In practical applications, the first random number may be added to each data in the first data segment to obtain a processed first data segment.

[0105] Exemplarily, the first random number range may be [-2, 2], and the first random number 1 may be randomly selected from the first random number range, and then each data in the first data segment may be added based on the first random number, that is, 1 is added to each data in the first data segment to obtain the processed first data segment.

[0106] After determining the first random number range and the second random number range, the second data segment may be divided into at least two data segments.

[0107] In practical applications, the second data segment can be divided according to a preset number of data to obtain at least two data segments. Typically, the number of data can be set to 20, that is, the second data segment is divided according to the number of 20 data to obtain at least two data segments, and each data segment includes 20 data.

[0108] After determining at least two data segments, a second random number corresponding to each data segment may be randomly determined from a second random number range, and the corresponding data segment may be processed based on each second random number to obtain a processed second data segment.

[0109] In a specific implementation, each data in the data segment corresponding to the second random number pair may be added to obtain a processed second data segment.

[0110] In one embodiment of the present application, the second random number corresponding to each data segment can be determined in a predetermined order. For example, for the first data segment, the second random number corresponding to the first data segment can be randomly determined from the second random number range, and for the second data segment, the random number range corresponding to the second data segment can be determined based on the second random number corresponding to the first data segment and the second random number range, and the second random number corresponding to the second data segment can be determined from the random number range corresponding to the second data segment, and the second random number corresponding to the second data segment can be determined, and so on, the second random numbers corresponding to the second data segment to the last data segment can be obtained.

[0111] Exemplarily, the second random number range may be [a, b], and the second random number corresponding to the first data segment may be k1, then the random number range corresponding to the second data segment may be [k1+a, k1+b], and the second random number corresponding to the second data segment may be k2, then the random number range corresponding to the third data segment may be [k2+a, k2+b], and similarly, the random number range and the corresponding second random number corresponding to the third data segment to the last data segment may be obtained.

[0112] After obtaining the processed first data segment and the second data segment, processed negative sample data can be generated according to the processed first data segment and the second data segment, so as to train the neural network model based on the processed negative sample data.

[0113] In an embodiment of the present application, the following steps may also be included:

[0114] Determine the target operating station where the sheet is located when performing riveting operations on the sheet.

[0115] The target operating station may be an operating station for riveting sheet materials, and the operating station may be a station for performing riveting operations on sheet materials.

[0116] After obtaining the pressure data of the riveting operation, the operation station where the riveting operation is performed can be determined, which is the target operation station.

[0117] In an embodiment of the present application, step 102 may be performed as follows:

[0118] The pressure data is input into the neural network model corresponding to the target operation station to obtain the probability that the riveting operation is a qualified operation.

[0119] Among them, different operating stations can correspond to different neural network models.

[0120] After the target operation station is determined, the neural network model corresponding to the target operation station can be determined, that is, the neural network model corresponding to the target operation station is determined from all neural network models.

[0121] In practical applications, since a corresponding neural network model can be pre-constructed for each operating station, after the target operating station is determined, the neural network model corresponding to the target operating station can be determined.

[0122] Exemplarily, after constructing the neural network model corresponding to each operation station, a mapping relationship between each station identification and the neural network model can be established according to the station identification of the operation station, so that after determining the target operation station of the riveting operation, the neural network model corresponding to the target operation station can be determined and retrieved based on the station identification of the target operation station and the mapping relationship between each station identification and the neural network model.

[0123] After determining the corresponding neural network model, the pressure data of the riveting operation can be input into the corresponding neural network model, and after the operation of the corresponding neural network model, the qualified probability of the riveting operation being a qualified operation can be obtained to generate the operation quality of the riveting operation based on the qualified probability.

[0124] In an embodiment of the present application, before step 102, the following steps may also be included:

[0125] The target event topic corresponding to the riveting operation is determined, and the pressure data of the riveting operation is recorded in the target event topic.

[0126] Among them, the target event topic can be the event topic corresponding to the riveting operation among all event topics. The event topic can be related to the message type of the published message in the subscription and publication mode. Different message types can correspond to different event topics. The event topic can be used to receive messages related to the event topic published by the publisher, and distribute the received messages to subscribers who subscribe to the event topic.

[0127] After obtaining the pressure data of the riveting operation, the target event topic corresponding to the riveting operation can be determined.

[0128] In actual applications, since each operating station can correspond to a neural network model, in order to input the pressure data of the riveting operation performed by each operating station into the corresponding neural network model, a corresponding event topic can be created in advance for each operating station, and corresponding message queues can be created for different event topics, and different operating stations correspond to different event topics. After obtaining the pressure data of the riveting operation, the pressure data can be published to the target event topic corresponding to the target operating station according to the target operating station corresponding to the riveting operation, and the pressure data of the riveting operation can be recorded in the target event topic, that is, the pressure data of the riveting operation is recorded in the message queue corresponding to the target event topic, so as to distribute the pressure data of the riveting operation to subscribers who subscribe to the target event topic through the target event topic.

[0129] In a specific implementation, the riveting device may be a publisher in a subscription-publishing mode, and after obtaining the pressure data of the riveting operation, the pressure data may be transmitted to a corresponding event topic.

[0130] In an embodiment of the present application, step 102 can be implemented in the following manner:

[0131] A target thread subscribing to a target event topic is determined, and pressure data of the riveting operation is transmitted to a neural network model corresponding to a target operation station through the target thread to obtain a qualified probability that the riveting operation is a qualified operation.

[0132] Among them, the target thread can be a thread that transmits data to the neural network model corresponding to the target operation station, and different threads can be used to transmit different data to the corresponding neural network model.

[0133] After determining the target event topic corresponding to the riveting operation and determining the neural network model corresponding to the target operation station, the target thread subscribing to the target event topic can be determined.

[0134] In actual applications, a thread for transmitting data to the neural network model can be created in advance for each neural network model, and the thread can be used as a subscriber in the subscription and publishing mode. Different threads can correspond to different neural network models, and different threads can subscribe to different event topics. After determining the target event topic, the target thread that subscribes to the target event topic can be determined.

[0135] After determining the target thread, the data in the target event topic can be obtained through the target thread, and the data can be transmitted to the neural network model corresponding to the target operation station to obtain the probability that the riveting operation is a qualified operation.

[0136] In an embodiment of the present application, pressure data is obtained, and the pressure data is used to reflect the pressure between the sheet material and the rivet when the riveting operation is performed on the sheet material. According to the pressure data and the neural network model, the probability that the riveting operation is a qualified operation is determined, so as to determine the operation quality of the riveting operation based on the qualified probability, and then the qualified probability of the riveting operation can be determined through the neural network model, thereby reducing the situation of manual sampling, thereby reducing the missed detection of connection defects, etc., and improving the quality of self-piercing riveting.

[0137] See also Figure 3 , Figure 3 A flowchart of another method for detecting the quality of a connection process provided by an embodiment of the present application is shown, which may specifically include the following steps:

[0138] Step 301, obtaining pressure data.

[0139] For the relevant description of step 301, please refer to step 101 and will not be repeated here.

[0140] Step 302: Determine the pressure change trend of the riveting operation according to the pressure data.

[0141] The pressure variation trend can be used to reflect the pressure variation between the sheet and the rivet when the riveting operation is performed on the sheet within a period of time.

[0142] In practical applications, after obtaining the pressure data, the displacement corresponding to each pressure data can be determined, and then based on the value of each pressure data and the displacement corresponding to each pressure data, the relationship between the pressure data and the displacement in the riveting operation can be determined, that is, the pressure change trend of the pressure data as the displacement increases.

[0143] In one embodiment of the present application, the pressure change trend can also be the pressure change trend of the pressure data changing with the increase of execution time. When the pressure change trend is the pressure change trend of the pressure data changing with the increase of execution time, the acquisition time of each pressure data can be determined after obtaining the pressure data, and then based on the acquisition time of each pressure data and the value of each pressure data, the pressure change trend of the pressure data changing with the increase of execution time in the riveting operation can be determined.

[0144] Step 303: input the pressure data into the neural network model to obtain the pass probability of the riveting operation.

[0145] After obtaining the pressure data, the pressure data can be input into the neural network model, and then the qualified probability of the riveting operation can be obtained through the processing of the neural network model.

[0146] Step 304: Generate the operation quality of the riveting operation according to the pressure variation trend and the qualified probability.

[0147] After obtaining the pressure variation trend of the riveting operation and the qualified probability of the riveting operation, the operation quality of the riveting operation can be evaluated according to the pressure variation trend and the qualified probability.

[0148] In practical applications, the qualified operations and unqualified operations corresponding to each operating station can be determined in advance, and the qualified pressure change trend corresponding to each qualified operation and the error pressure change trend corresponding to each unqualified operation can be determined. Then, after obtaining the pressure data of the riveting operation, the target operating station corresponding to the riveting operation can be determined, and the qualified pressure change trend and the error pressure change trend corresponding to the target operating station can be determined.

[0149] In a specific implementation, the qualified probability output by the neural network model can be compared with a preset qualified probability threshold, and the pressure change trend of the riveting operation can be compared with the qualified pressure change trend. Then, when the qualified probability output by the neural network model is greater than or equal to the qualified probability threshold, and the similarity between the pressure change trend of the riveting operation and the qualified pressure change trend is greater than or equal to the similarity threshold, it can be determined that the riveting operation is a qualified operation, and the operation quality of the riveting operation is determined to not produce connection defects that affect the connection strength, that is, there is no riveting quality problem in the riveting operation.

[0150] When the qualified probability output by the neural network model is less than the qualified probability threshold, and / or the similarity between the pressure change trend of the riveting operation and the qualified pressure change trend is less than the similarity threshold, it can be determined that the riveting operation is an unqualified operation, and the operation quality of the riveting operation is determined to be a connection defect that will affect the connection strength, that is, the riveting operation has a riveting quality problem.

[0151] In an embodiment of the present application, the following steps may also be included:

[0152] According to the execution position of the riveting operation on the sheet material, it is determined whether the riveting operation is a target riveting operation. When the riveting operation is a target riveting operation, a neural network model is used to identify the target features in the pressure change trend.

[0153] Among them, the execution part can be the part where the riveting operation is performed on the sheet material, the target riveting operation can be a riveting operation with a preset change characteristic in the pressure change trend, the preset change characteristic can indicate that there is an obvious change in the pressure change trend and the change is related to the process of the riveting operation, the preset change characteristic can be determined based on the process of the riveting operation, and the target characteristic can be a characteristic that the pressure change within a preset time period in the pressure change trend exceeds a preset change range.

[0154] In practical applications, the execution location of the riveting operation can be determined, and then whether the riveting operation is a target riveting operation can be determined based on the execution location of the riveting operation.

[0155] In a specific implementation, for each operating station, it is possible to determine in advance whether there is a target riveting operation from the riveting operations of the operating station in the past period of time, that is, to determine whether there is a preset change feature from the pressure change trend of each riveting operation in the past period of time. If there is a preset change feature in the pressure change trend of one of the riveting operations, then the riveting operation can be determined to be the target riveting operation, and the operating station can be determined to be the operating station where the target riveting operation exists. Furthermore, if the target operating station is an operating station where the target riveting operation exists, the riveting operation in the target operating station can be determined to be the target riveting operation, and if the target operating station is an operating station where the target riveting operation does not exist, then the riveting operation in the target operating station can be determined to be a non-target riveting operation.

[0156] In the case where the riveting operation is the target riveting operation, a neural network model may be used to identify target features in the pressure variation trend of the riveting operation.

[0157] It should be understood that since the processes for performing riveting operations at different parts may be different, the pressure change trend of each riveting operation will also be different, and the pressure change trend of the target riveting operation may have a change feature that the pressure change exceeds the preset change range within a preset time period, that is, the pressure change trend of the target riveting operation may have a preset change feature, and for operating stations at different parts, the corresponding neural network model can be trained with sample data to further identify whether there is a change feature in the pressure change trend of the riveting operation of the operating station that the pressure change exceeds the preset change range within a preset time period.

[0158] In practical applications, when a riveting operation is determined to be a target riveting operation, a neural network model corresponding to the target operation station may be called to identify target features in the pressure variation trend of the riveting operation.

[0159] In an embodiment of the present application, step 304 may also be performed in the following manner:

[0160] The operation quality of the riveting operation is generated based on the target characteristics, pressure change trend and qualified probability.

[0161] After obtaining the target characteristics, pressure change trend and qualified probability, the operation quality of the riveting operation can be evaluated based on the target characteristics, pressure change trend and qualified probability.

[0162] In practical applications, the target feature can be compared with the preset change feature, the qualified probability output by the neural network model can be compared with the preset qualified probability threshold, and the pressure change trend of the riveting operation can be compared with the qualified pressure change trend. Then, when the target feature matches the preset change feature, the qualified probability output by the neural network model is greater than or equal to the qualified probability threshold, and the similarity between the pressure change trend of the riveting operation and the qualified pressure change trend is greater than or equal to the similarity threshold, the riveting operation can be determined to be a qualified operation, and the operation quality of the riveting operation can be determined to be that no connection defects that affect the connection strength will be generated, that is, there is no riveting quality problem in the riveting operation.

[0163] However, when the target feature does not match the preset change feature, or the qualified probability output by the neural network model is less than the qualified probability threshold, or the similarity between the pressure change trend of the riveting operation and the qualified pressure change trend is less than the similarity threshold, it can be determined that the riveting operation is an unqualified operation, and the operation quality of the riveting operation is determined to be a connection defect that will affect the connection strength, that is, the riveting operation has a riveting quality problem.

[0164] In an embodiment of the present application, the following steps may also be included:

[0165] Show the pressure change trend to the user.

[0166] In practical applications, after the pressure change trend corresponding to the riveting operation is obtained, the pressure change trend can be displayed by a display device. Specifically, the display device can be a device for displaying images, such as a display, a mobile terminal with a display screen, and the like.

[0167] See also Figure 4 , Figure 4 A schematic diagram of a pressure change trend provided by an embodiment of the present application is shown. Figure 4As shown, the horizontal axis can represent the displacement of the rivet inserted into the sheet, the vertical axis can represent the pressure between the rivet and the sheet, and curve a can represent the pressure change trend corresponding to the riveting operation, that is, the relationship between the displacement of the rivet inserted into the sheet and the pressure between the rivet and the sheet in the riveting operation.

[0168] It needs to be understood that since at least one riveting operation can be performed on the same part of the sheet in the same operating station, that is, at least one rivet can be inserted into the part for connection, the pressure change trend corresponding to all riveting operations in the operating station can be displayed, that is, the pressure change trend of the pressure between each rivet and the sheet can be displayed.

[0169] In an embodiment of the present application, step 304 may also be performed in the following manner:

[0170] In response to the user's evaluation operation on the pressure change trend, the user's quality evaluation information on the riveting operation is generated, and the operation quality of the riveting operation is generated based on the quality evaluation information and the qualified probability.

[0171] Among them, the evaluation operation can be an operation in which the user evaluates whether the riveting operation is a qualified operation based on the pressure change trend, and the quality evaluation information can be information that the user evaluates whether the riveting operation is a qualified operation. The quality evaluation information may include information that the user evaluates the riveting operation as a qualified operation and information that the user evaluates the riveting operation as an unqualified operation.

[0172] After the pressure change trend is displayed to the user, the user can evaluate whether the riveting operation is a qualified operation based on the displayed pressure change trend, and then the user can output the result of the evaluation and perform a corresponding evaluation operation. After detecting the evaluation operation performed by the user, the system can respond to the user's evaluation operation on the pressure change trend and generate quality evaluation information corresponding to the evaluation operation. The operation quality of the riveting operation can be generated based on the quality evaluation information and the qualified probability.

[0173] In practical applications, the qualified probability output by the neural network model can be compared with a preset qualified probability threshold, and the pressure change trend of the riveting operation can be compared with the qualified pressure change trend. Then, when the quality assessment information is information that the user assesses the riveting operation as a qualified operation, the qualified probability output by the neural network model is greater than or equal to the qualified probability threshold, and the similarity between the pressure change trend of the riveting operation and the qualified pressure change trend is greater than or equal to the similarity threshold, it can be determined that the riveting operation is a qualified operation, and the operation quality of the riveting operation is determined to not produce connection defects that affect the connection strength, that is, there is no riveting quality problem in the riveting operation.

[0174] When the quality assessment information is information that the user assesses the riveting operation as an unqualified operation, or the qualified probability output by the neural network model is less than the qualified probability threshold, or the similarity between the pressure change trend of the riveting operation and the qualified pressure change trend is less than the similarity threshold, then it can be determined that the riveting operation is an unqualified operation, and the operation quality of the riveting operation is determined to produce connection defects that affect the connection strength, that is, the riveting operation has a riveting quality problem.

[0175] In one embodiment of the present application, after obtaining the operation quality of the riveting operation, the operation quality of the riveting operation and related information of the riveting operation can be recorded in a data warehouse, and the related information of the riveting operation may include the station identification of the operation station, the sheet identification of the sheet performing the riveting operation, the position in the sheet performing the riveting operation, the execution time of the riveting operation, the pressure change trend corresponding to the riveting operation and other information. Furthermore, the data warehouse can record the operation quality of all riveting operations in the past period of time and the related information of each riveting operation, so as to provide users with data query, data analysis and other services.

[0176] Specifically, data query can be to query the pressure change trend of unqualified operations in the past period of time, to query the operation of a specific unqualified operation in the past period of time, and data analysis can be to analyze the pressure change trend of riveting operations, to perform statistical analysis on abnormal operations of unqualified operations in the past period of time, and other operations.

[0177] In an embodiment of the present application, by acquiring pressure data of the riveting operation, the pressure data is the pressure between the rivet and the sheet material, and based on the pressure data, the pressure change trend of the riveting operation is determined, and the pressure data is input into a neural network model. The neural network model is used to determine the probability that the riveting operation is a qualified operation based on the pressure data, and then the qualified probability of the riveting operation can be obtained. According to the pressure change trend and the qualified probability, the operation quality of the riveting operation is generated, the situation of manual random inspection is reduced, thereby reducing the missed detection of connection defects and the like, and improving the quality of self-piercing riveting.

[0178] See also Figure 5 , Figure 5 FIG. 1 shows a schematic diagram of a connection process quality detection system provided by an embodiment of the present application. Figure 5 As shown, the detection system may include:

[0179] The data collection module 51 is used to collect the pressure data between the rivet and the sheet material during the riveting operation performed by each operating station. The riveting operation may be an operation of riveting the sheet material.

[0180] Specifically, the data acquisition module 51 can be deployed in the riveting equipment of each operating station, and when the riveting equipment performs the riveting operation, the pressure data between the rivet and the sheet material during the riveting operation can be collected by the data acquisition module 51 of each operating station.

[0181] The quality detection module 52 is used to distribute the pressure data to the neural network model corresponding to the operation station, obtain the output result of the neural network model, and generate the operation quality of the riveting operation performed at the operation station based on the output result.

[0182] The neural network model is used to determine the probability that the riveting operation is a qualified operation according to the pressure data, so as to generate the operation quality of the riveting operation based on the qualified probability, which is the output result of the neural network model.

[0183] In an embodiment of the present application, the detection system may further include:

[0184] The data storage module 53 is used to transfer the pressure data collected by the data collection module 51 to a pre-created data warehouse. The data warehouse can be a data set integrating at least one data source and at least one time.

[0185] The change data capture module 54 is used to identify changed data from the data warehouse using the change data capture technology, and transmit the changed data to the quality detection module 52 .

[0186] In actual applications, when the pressure data collected by the data acquisition module 51 is stored in the data warehouse, the data in the data warehouse will change, so that the change data capture module 54 can identify the pressure data and transmit the pressure data to the quality detection module 52.

[0187] In one embodiment of the present application, a corresponding neural network model can be deployed in the quality detection module 52 for each operating station, and then after receiving the pressure data transmitted by the change data capture module 54, the received pressure data can be distributed to the corresponding neural network model.

[0188] Exemplarily, the quality detection module 52 may adopt a subscription-publishing mode to distribute the received pressure data to the corresponding neural network model.

[0189] In actual applications, at least one event topic can be created in advance, and a message queue can be assigned to each event topic. Each event topic can be related to the type of message, and the type of message can be defined as the type of operation station, that is, different operation stations can correspond to different event topics, and the message queue corresponding to each event topic can be deployed in the quality detection module 52.

[0190] In a specific implementation, after receiving the pressure data transmitted by the change data capture module 54, the quality detection module 52 can determine the operating station corresponding to the pressure data, that is, determine the riveting operation that generates the pressure data and the operating station that performs the riveting operation, and then determine the event subject corresponding to the operating station and determine the message queue corresponding to the event subject, and transmit the pressure data to the message queue.

[0191] For each neural network model in the quality detection module 52, a thread for transmitting the pressure data in the message queue to the neural network model can be created in advance for each neural network model, and different neural network models can correspond to one of the threads. Since each neural network model can correspond to a different operating station, and different operating stations can correspond to different event topics, different neural network models can correspond to different event topics. For each neural network model, the thread corresponding to the neural network model can subscribe to the event topic corresponding to the neural network model, and the thread corresponding to the neural network model can transmit the pressure data in the message queue corresponding to the event topic to the neural network model, obtain the output result of the neural network model, and generate the operation quality of the riveting operation performed at the operating station based on the output result.

[0192] In one embodiment of the present application, the detection system further includes:

[0193] The visualization module 55 is used to display the pressure data to the user when the operation quality indicates that the riveting operation is a qualified operation.

[0194] Exemplarily, the visualization module 55 can also be used to display the pressure change trend of the riveting operation to the user when the riveting operation is a target riveting operation, and to receive and respond to the user's evaluation operation on the pressure change trend of the riveting operation, to generate quality assessment information, and to transmit the quality assessment information to the quality detection module 52 for quality assessment.

[0195] In one embodiment of the present application, the visualization module 55 can also be used to display to the user the pressure change trend of the riveting operation at each operating station over a period of time.

[0196] In one embodiment of the present application, the quality detection module 52 is also used to generate an alarm message and push the alarm message to the user to warn the user when the operation quality indicates that the riveting operation is an unqualified operation. When the operation quality indicates that the riveting operation is a qualified operation, the pressure data is transmitted to the visualization module 55 to display the pressure data of the riveting operation to the user.

[0197] Exemplarily, the quality detection module 52 is also used to determine whether the riveting operation is a target riveting operation, and when the riveting operation is a target riveting operation, transmit the pressure change trend of the riveting operation to the visualization module 55 for display through the visualization module 55, and after receiving the quality assessment information transmitted by the visualization module 55, distribute the pressure data to the neural network model corresponding to the operating station, obtain the pass probability of the riveting operation being a qualified operation, and generate the operation quality of the riveting operation based on the quality assessment information and the pass probability.

[0198] In one embodiment of the present application, the detection system further includes:

[0199] The training module 56 is used to obtain sample data of each operating station and train the neural network model based on the sample data to obtain a trained neural network model, so as to deploy the trained neural network model in the quality detection module 52.

[0200] In practical applications, the training module can also be used to perform preprocessing operations on the sample data and data enhancement operations on the negative sample data in the sample data after obtaining the sample data of each operating station, so as to train the neural network model based on the processed sample data.

[0201] In an embodiment of the present application, pressure data is obtained, and the pressure data is used to reflect the pressure between the sheet material and the rivet when the riveting operation is performed on the sheet material. According to the pressure data and the neural network model, the probability that the riveting operation is a qualified operation is determined, so as to determine the operation quality of the riveting operation based on the qualified probability, and then the qualified probability of the riveting operation can be determined through the neural network model, thereby reducing the situation of manual sampling, thereby reducing the missed detection of connection defects, etc., and improving the quality of self-piercing riveting.

[0202] See also Figure 6 , Figure 6 A flow chart of a method for detecting connection process quality provided by an embodiment of the present application is shown, which may specifically include the following steps:

[0203] Step 601, the riveting equipment collects pressure data of the riveting operation;

[0204] Step 602, the riveting equipment stores the pressure data in a data warehouse;

[0205] Step 603, obtaining the pressure data stored in the data warehouse through the change data capture module, and publishing the pressure data to the event topic corresponding to the target operation station of the riveting operation;

[0206] Specifically, the change data capture module can use the change data capture technology to obtain the data that has changed in the data warehouse;

[0207] Step 604, determining a target thread that subscribes to the event topic corresponding to the riveting operation, and distributing the pressure data published in the event topic corresponding to the riveting operation to the neural network model corresponding to the target operation station through the target thread, to obtain the qualified probability output by the neural network model;

[0208] Step 605, determining whether the riveting operation is a qualified operation according to the qualified probability, if so, the process ends, if not, executing step 606;

[0209] Step 606: Generate alarm information for the riveting operation, and push the alarm information to the user to warn the user.

[0210] In an embodiment of the present application, pressure data of the riveting operation is collected by a riveting device and stored in a data warehouse, and then the pressure data stored in the data warehouse is obtained by a change data capture module, and the pressure data is published to an event topic corresponding to a target operation station of the riveting operation, a target thread subscribing to the event topic corresponding to the riveting operation is determined, and the pressure data published in the event topic corresponding to the riveting operation is distributed to a neural network model corresponding to the target operation station through the target thread, so as to obtain a qualified probability output by the neural network model, reduce the situation of manual sampling, thereby reducing the situation of missed detection of connection defects, etc., and improve the quality of self-piercing riveting, so that it is possible to determine whether the riveting operation is a qualified operation based on the qualified probability, and if the riveting operation is an unqualified operation, generate an alarm message for the riveting operation, and push the alarm message to the user to warn the user.

[0211] See also Figure 7 , Figure 7 The following is a schematic diagram showing the structure of a connection process quality detection device provided by an embodiment of the present application, which may specifically include the following modules:

[0212] The acquisition module 701 is used to acquire pressure data; wherein the pressure data is used to reflect the pressure between the sheet material and the rivet when the riveting operation is performed on the sheet material;

[0213] The determination module 702 is used to determine the probability that the riveting operation is a qualified operation based on the pressure data and the neural network model, so as to determine the operation quality of the riveting operation based on the qualified probability; the neural network model is obtained by training a plurality of sample data, and the sample data includes positive sample data and negative sample data, the positive sample data includes the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is a qualified operation, and the negative sample data includes the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is an unqualified operation.

[0214] In one implementation, the determination module 702 may also be used to:

[0215] Determine the target operation station where the sheet material is located when performing riveting operation on the sheet material;

[0216] The pressure data is input into the neural network model corresponding to the target operating station to obtain the probability that the riveting operation is a qualified operation; wherein different operating stations correspond to different neural network models, and the neural network model corresponding to any operating station is used to determine the probability that the riveting operation performed on the sheet material at the operating station is a qualified operation.

[0217] In one implementation, the device further includes the following modules:

[0218] A recording module, used for determining a target event subject corresponding to the riveting operation before inputting the pressure data into a neural network model corresponding to the target operation station to obtain a qualified probability that the riveting operation is a qualified operation, and recording the pressure data of the riveting operation in the target event subject;

[0219] In one implementation, the determination module 702 may also be used to:

[0220] Determine a target thread that subscribes to a target event topic; wherein the target thread is a thread that transmits data to a neural network model corresponding to a target operation station, and different threads are used to transmit different data to corresponding neural network models;

[0221] The pressure data in the target event topic is transmitted to the neural network model corresponding to the target operation station through the target thread to obtain the qualified probability that the riveting operation is a qualified operation.

[0222] In one implementation, the determination module 702 may also be used to:

[0223] Before determining the probability that the riveting operation is a qualified operation according to the pressure data and the neural network model, generating a data sequence corresponding to the unqualified operation according to the pressure data of the unqualified operation;

[0224] The first data segment in the data sequence is processed based on the first random number to obtain a processed first data segment; wherein the first random number belongs to the first random number range,

[0225] The second data segment in the data sequence is divided into at least two data segments; wherein the average value of the second data segment is greater than the average value of the first data segment,

[0226] Processing the corresponding data segment based on the second random number corresponding to each data segment in the at least two data segments to obtain a processed second data segment; wherein the second random number belongs to a second random number range, and the second random number range is greater than the first random number range;

[0227] According to the processed first data segment and the processed second data segment, processed negative sample data is generated to train the neural network model based on the processed negative sample data.

[0228] In one implementation, the determination module 702 may also be used to:

[0229] Before processing the first data segment in the data sequence based on the first random number to obtain the processed first data segment, determining a target abnormal time period corresponding to the unqualified operation; wherein the target abnormal time period is a concentrated time period when abnormalities occur in the unqualified operation;

[0230] Based on the target abnormal time period, the data sequence is divided into a first data segment and a second data segment; wherein the second data segment is a concentrated time period when abnormalities occur.

[0231] In one implementation, the determination module 702 may also be used to:

[0232] Determine the pressure change trend of the riveting operation according to the pressure data; wherein the pressure change trend is used to reflect the pressure change between the sheet and the rivet when the riveting operation is performed on the sheet within a period of time;

[0233] The quality of the riveting operation is determined based on the pressure change trend and the probability of passing.

[0234] In one implementation, the determination module 702 may also be used to:

[0235] Determining whether the riveting operation is a target riveting operation according to the execution position of the riveting operation on the sheet material; wherein the target riveting operation is a riveting operation having a preset change feature in the pressure change trend;

[0236] In the case where the riveting operation is the target riveting operation, the neural network model is used to identify the target features in the pressure change trend;

[0237] The operational quality of the riveting operation is determined based on the target characteristics, pressure change trend and qualified probability.

[0238] In one implementation, the device further includes the following modules:

[0239] A display module is used to display the pressure change trend to the user;

[0240] In one implementation, the determination module 702 may also be used to:

[0241] Detection of evaluation operations for pressure change trends;

[0242] In response to the evaluation operation, determining quality evaluation information for the riveting operation;

[0243] Based on the quality assessment information and the probability of passing, the operational quality of the riveting operation is determined.

[0244] In an embodiment of the present application, pressure data is obtained, and the pressure data is used to reflect the pressure between the sheet material and the rivet when the riveting operation is performed on the sheet material. According to the pressure data and the neural network model, the probability that the riveting operation is a qualified operation is determined, so as to determine the operation quality of the riveting operation based on the qualified probability, and then the qualified probability of the riveting operation can be determined through the neural network model, thereby reducing the situation of manual sampling, thereby reducing the missed detection of connection defects, etc., and improving the quality of self-piercing riveting.

[0245] It should be noted that the information interaction, execution process and other contents between the above-mentioned devices are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0246] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0247] See also Figure 8 , Figure 8 A structural block diagram of a terminal device provided by an embodiment of the present application is shown as follows: Figure 8 As shown, this embodiment provides a terminal device 81, which includes: at least one processor 811, a memory 812, and a computer program 8121 stored in the memory 812 and executable on at least one processor 811, and when the processor 811 executes the computer program 8121, the steps in any of the above-mentioned method embodiments are implemented.

[0248] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments can be implemented.

[0249] An embodiment of the present application provides a computer program product. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0250] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device capable of carrying the computer program code to a terminal device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electrical carrier signal, a telecommunication signal, and a software distribution medium.

[0251] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for detecting connection process quality, characterized in that: The method comprises: Acquiring pressure data; wherein the pressure data is used to reflect the pressure between the sheet and the rivet when the sheet is riveted; According to the pressure data and the neural network model, the probability that the riveting operation is a qualified operation is determined, so as to determine the operation quality of the riveting operation based on the qualified probability; the neural network model is obtained by training a plurality of sample data, the sample data comprising positive sample data and negative sample data, the positive sample data comprising the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is a qualified operation, and the negative sample data comprising the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is an unqualified operation.

2. The method for detecting the connection process quality according to claim 1, characterized in that: The method further comprises: Determining a target operating station where the sheet material is located when performing the riveting operation on the sheet material; Wherein, determining the probability that the riveting operation is a qualified operation according to the pressure data and the neural network model includes: The pressure data is input into the neural network model corresponding to the target operating station to obtain the probability that the riveting operation is a qualified operation; wherein different operating stations correspond to different neural network models, and the neural network model corresponding to any operating station is used to determine the probability that the riveting operation performed on the sheet material at the operating station is a qualified operation.

3. The method for detecting the connection process quality according to claim 2, characterized in that: Before inputting the pressure data into the neural network model corresponding to the target operation station to obtain the probability that the riveting operation is a qualified operation, the method further includes: Determine a target event topic corresponding to the riveting operation, and record the pressure data of the riveting operation in the target event topic; The step of inputting the pressure data into the neural network model corresponding to the target operation station to obtain the probability that the riveting operation is a qualified operation includes: Determine a target thread that subscribes to the target event topic; wherein the target thread is a thread that transmits data to a neural network model corresponding to the target operation station, and different threads are used to transmit different data to corresponding neural network models; The pressure data in the target event topic is transmitted to the neural network model corresponding to the target operation station through the target thread to obtain the qualified probability that the riveting operation is a qualified operation.

4. The method for detecting the connection process quality according to any one of claims 1 to 3, characterized in that: Before determining the probability that the riveting operation is a qualified operation according to the pressure data and the neural network model, the method further includes: generating a data sequence corresponding to the unqualified operation according to the pressure data of the unqualified operation; Processing a first data segment in the data sequence based on a first random number to obtain a processed first data segment; wherein the first random number belongs to a first random number range; Dividing a second data segment in the data sequence into at least two data segments; wherein the average value of the second data segment is greater than the average value of the first data segment; Processing the corresponding data segment based on the second random number corresponding to each of the at least two data segments to obtain a processed second data segment; wherein the second random number belongs to a second random number range, and the second random number range is larger than the first random number range; Generate processed negative sample data according to the processed first data segment and the processed second data segment, so as to train the neural network model based on the processed negative sample data.

5. The method for detecting the quality of the connection process according to claim 4, characterized in that: Before processing the first data segment in the data sequence based on the first random number to obtain the processed first data segment, the method further includes: Determine a target abnormal time period corresponding to the unqualified operation; wherein the target abnormal time period is a concentrated time period during which abnormalities occur in the unqualified operation; Based on the target abnormal time period, the data sequence is divided into a first data segment and a second data segment; wherein the second data segment is a concentrated time period when abnormalities occur.

6. The method for detecting the quality of the connection process according to any one of claims 1 to 3 or 4 or 5, characterized in that: The method further comprises: Determine the pressure change trend of the riveting operation according to the pressure data; wherein the pressure change trend is used to reflect the pressure change between the sheet material and the rivet when the riveting operation is performed on the sheet material within a time period; Wherein, determining the operation quality of the riveting operation based on the qualified probability includes: The operation quality of the riveting operation is determined according to the pressure variation trend and the qualified probability.

7. The method for detecting the quality of the connection process according to claim 6, characterized in that: The method further comprises: Determining whether the riveting operation is a target riveting operation according to the execution position of the riveting operation on the sheet material; wherein the target riveting operation is a riveting operation having a preset change feature in the pressure change trend; In a case where the riveting operation is a target riveting operation, using the neural network model to identify a target feature in the pressure change trend; Determining the operation quality of the riveting operation according to the pressure change trend and the probability includes: The operation quality of the riveting operation is determined according to the target feature, the pressure change trend and the qualified probability.

8. The method for detecting the quality of the connection process according to claim 6, characterized in that: The method further comprises: showing the pressure variation trend; Wherein, determining the operation quality of the riveting operation based on the qualified probability includes: detecting an evaluation operation for the pressure change trend; In response to the evaluating operation, determining quality evaluation information for the riveting operation; The operation quality of the riveting operation is determined according to the quality assessment information and the pass probability.

9. A device for detecting the quality of a connection process, characterized in that: The device comprises: An acquisition module, used for acquiring pressure data; wherein the pressure data is used for reflecting the pressure between the sheet material and the rivet when the riveting operation is performed on the sheet material; A determination module is used to determine the probability that the riveting operation is a qualified operation based on the pressure data and the neural network model, so as to determine the operation quality of the riveting operation based on the qualified probability; the neural network model is trained by multiple sample data, and the sample data includes positive sample data and negative sample data, the positive sample data includes the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is a qualified operation, and the negative sample data includes the pressure data between the sheet material and the rivet when the riveting operation performed on the sheet material is an unqualified operation.

10. A connection process quality detection system, characterized in that: The system comprises: A data acquisition module, used to collect pressure data between the rivet and the sheet material during the riveting operation performed at each operating station; wherein the riveting operation is an operation of riveting the sheet material; The quality detection module is used to distribute the pressure data to the neural network model corresponding to the operation station, obtain the output result of the neural network model, and generate the operation quality of the riveting operation performed at the operation station based on the output result.

Images