Welding quality state prediction method and device based on statistical model, equipment and medium

Abstract

The invention provides a welding quality state prediction method and device based on a statistical model, equipment and a medium, relates to the technical field of laser welding manufacturing, and is used for solving the problem that a traditional method cannot feed back welding quality in real time. The method comprises the steps that the statistical model is constructed, and target values of parameters of the statistical model are solved; the statistical model is used for indicating the corresponding relation between the key welding parameters and the welding quality state; according to the key welding parameters of the weldment in the laser welding process, a welding parameter sequence is constructed; and the welding parameter sequence is input into the statistical model, solving is carried out according to the target value of the parameter, and the welding quality state of the weldment after laser welding is completed is predicted. According to the method, the welding quality state of the weldment after laser welding can be predicted in advance according to the key welding parameters of the weldment in the laser welding process, so that the welding quality is fed back in real time.

Description

Technical Field

[0001] The present application relates to the field of laser welding manufacturing technology, and provides a welding quality status prediction method, device, equipment and medium based on a statistical model. Background Art

[0002] Laser welding manufacturing technology is a material connection process that uses the high energy density of the laser beam to locally heat the weldment, so that two or more metal materials or non-metallic materials can be combined at the atomic level. Laser welding of titanium alloy plates and other weldments will produce a variety of quality conditions such as normal welding, undercut, non-penetration, cracks, deformation, etc. Welding quality is a key factor in ensuring product performance and service life.

[0003] The traditional welding quality status detection method mainly relies on the quality status of the weldment after laser welding. This method is time-consuming and labor-intensive and cannot provide real-time feedback on welding quality problems, making it difficult to effectively prevent the occurrence of welding defects. Summary of the invention

[0004] The present application provides a welding quality status prediction method, device, equipment and medium based on a statistical model, which are used to solve the problem that traditional methods cannot provide real-time feedback on welding quality.

[0005] In a first aspect, a welding quality state prediction method based on a statistical model is provided, comprising: Constructing a statistical model and solving the target values ​​of the parameters of the statistical model; the statistical model is used to indicate the corresponding relationship between the key welding parameters and the welding quality status; According to the key welding parameters of the weldment during the laser welding process, a welding parameter sequence is constructed; The welding parameter sequence is input into the statistical model, and the solution is performed according to the target value of the parameter to predict the welding quality state of the weldment after the laser welding is completed.

[0006] Optionally, the key welding parameters include welding power, welding speed, galvanometer focal length, welding part thickness, spot size and amplitude.

[0007] Optionally, solving the target value of the parameter of the statistical model includes: Initializing the parameters of the statistical model; the parameters include an initial state probability matrix, a state transfer matrix and an observation probability matrix; The value of the parameter is continuously updated through a forward-backward algorithm until a maximum number of iterations is reached to obtain a target value of the parameter.

[0008] Optionally, the step of inputting the welding parameter sequence into the statistical model, solving the model according to the target value of the parameter, and predicting the welding quality state of the weldment after laser welding is completed includes: Inputting the welding parameter sequence into the statistical model, solving it according to the target value of the parameter, and obtaining the maximum probability in the probability distribution of the welding quality state sequence; The welding quality state corresponding to the maximum probability is determined as the welding quality state of the weldment after laser welding is completed.

[0009] Optionally, the welding quality state sequence includes a normal state, an undercut state, a non-penetration state, a crack state and a deformation state.

[0010] Optionally, after inputting the welding parameter sequence into the statistical model, solving according to the target value of the parameter, and predicting the welding quality state of the weldment after laser welding, the method further includes: If the welding quality state is an abnormal state, adjusting key welding parameters of the weldment during the laser welding process according to the welding quality state; The laser welding of the weldment is continued according to the adjusted key welding parameters.

[0011] Optionally, after continuing to laser weld the weldment according to the adjusted key welding parameters, the method further includes: constructing an adjusted welding parameter sequence according to the adjusted key welding parameters; The adjusted welding parameter sequence is input into the statistical model, and the solution is performed according to the target value of the parameter to predict the welding quality state of the weldment after the parameters are adjusted and the laser welding is completed.

[0012] In a second aspect, a welding quality state prediction device based on a statistical model is provided, comprising: A model building module, used to build a statistical model and solve the target value of the parameters of the statistical model; the statistical model is used to indicate the corresponding relationship between the key welding parameters and the welding quality status; A sequence building module is used to build a welding parameter sequence according to the key welding parameters of the weldment during the laser welding process; The model prediction module is used to input the welding parameter sequence into the statistical model, solve it according to the target value of the parameter, and predict the welding quality state of the weldment after completing laser welding.

[0013] In a third aspect, the present application provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the welding quality status prediction method based on the statistical model described in the first aspect.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and a processor executes the computer program to implement the welding quality status prediction method based on a statistical model described in the first aspect.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present application provides a method for predicting welding quality status based on a statistical model, the method comprising: constructing a statistical model and solving the target values ​​of the parameters of the statistical model; the statistical model is used to indicate the correspondence between key welding parameters and welding quality status; according to the key welding parameters of the weldment during the laser welding process, a welding parameter sequence is constructed; the welding parameter sequence is input into the statistical model, and the target values ​​of the parameters are solved to predict the welding quality status of the weldment after laser welding is completed.

[0016] This application is based on a statistical model and makes predictions based on the key welding parameters of the weldment during the welding process. It can predict the welding quality status of the weldment after laser welding before completing laser welding, thereby providing real-time feedback on welding quality issues. Subsequently, the key welding parameters can be adjusted based on the welding quality status, thereby preventing the occurrence of welding defects and improving welding quality and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0018] Figure 1 A schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application; Figure 2 Schematic diagram of the welding quality state prediction method based on the statistical model provided in the embodiment of the present application Figure 1 ; Figure 3 Schematic diagram of the welding quality state prediction method based on the statistical model provided in the embodiment of the present application Figure 2 ; Figure 4 Schematic diagram of the welding quality state prediction method based on the statistical model provided in the embodiment of the present application Figure 3 ; Figure 5 A schematic diagram of the structure of a welding quality status prediction device based on a statistical model provided in an embodiment of the present application.

[0019] Markings in the figure: 101 - processor, 102 - communication bus, 103 - network interface, 104 - user interface, 105 - memory. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the drawings in the embodiment of the present application. Obviously, the described embodiment is only a part of the embodiment of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other arbitrarily. In addition, although the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in an order different from that here.

[0021] The traditional welding quality status detection method is to detect the finished product after laser welding, and judge whether the welding is qualified based on the final quality status performance. The disadvantages of this method are as follows: 1. Time-consuming and laborious: It is necessary to wait until the laser welding is completed before testing, which not only increases the production cycle, but also may cause welding defects to be difficult to repair when they are exposed in the final product.

[0022] Second, it is impossible to provide real-time feedback on problems: This method usually identifies problems only after production is completed. It lacks real-time quality monitoring and is unable to promptly detect and correct potential problems in welding.

[0023] 3. It is difficult to prevent welding defects: Because the inspection is carried out after completion, it is mainly a post-inspection and cannot make preventive corrections. Adjustments can often only be made when problems appear.

[0024] In view of this, the embodiment of the present application provides a welding quality status prediction method based on a statistical model. The method may be performed by a computer device. Figure 1 , which is a schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0025] like Figure 1As shown, the computer device may include: a processor 101, such as a central processing unit (CPU), a communication bus 102, a user interface 104, a network interface 103, and a memory 105. Among them, the communication bus 102 is used to realize the connection and communication between these components. The user interface 104 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 104 may include a standard wired interface and a wireless interface. The network interface 103 may include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 105 may be a high-speed random access memory (Random Access Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 105 may optionally also be a storage device independent of the aforementioned processor 101.

[0026] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0027] like Figure 1 As shown, the memory 105 as a storage medium may include an operating system, a network communication module, a user interface module, and a welding quality status prediction device based on a statistical model.

[0028] exist Figure 1 In the computer device shown, the network interface 103 is mainly used for data communication with the network server; the user interface 104 is mainly used for data interaction with the user; the processor 101 and the memory 105 in the computer device of the present invention can be set in the computer device, and the computer device calls the welding quality status prediction device based on the statistical model stored in the memory 105 through the processor 101, and executes the welding quality status prediction method based on the statistical model provided in the embodiment of the present application.

[0029] Please refer to Figure 2 , which is a schematic diagram of the process of predicting welding quality status based on the statistical model provided in the embodiment of the present application Figure 1 , the following is based on Figure 1 The computer equipment shown is Figure 2 A welding quality status prediction method based on statistical model is introduced as shown in the figure.

[0030] S201, construct a statistical model, and solve the target values ​​of the parameters of the statistical model.

[0031] In the specific implementation process, first, historical welding data is collected, including key welding parameters of various types of weldments during the welding process, and the welding quality status of each type of weldment after laser welding. These data can come from the measurement records in the actual production process. For example, when laser welding titanium alloy plates, weldments will produce various quality states such as normal welding, undercut, non-penetration, cracks, deformation, etc. Secondly, a suitable statistical model is selected to establish the relationship between key welding parameters and welding quality status. The statistical model can be a regression model, support vector machine, neural network, random forest, etc. Finally, the statistical model is trained based on the historical welding data to obtain the target values ​​of the parameters of the statistical model.

[0032] S202. Construct a welding parameter sequence according to key welding parameters of the weldment during the laser welding process.

[0033] In the specific implementation process, weldments refer to two or more parts connected together by welding, such as titanium alloy plates. The key welding parameters of weldments during laser welding can be collected first. The key welding parameters include welding power, welding speed, galvanometer focal length, welding part thickness, spot size and amplitude. Then, according to the key welding parameters, a welding parameter sequence is defined. O ={ o 1 ,o 2 ,...,o M},in, o M is the Mth key welding parameter, M is the total number of key welding parameters, M=6.

[0034] The key welding parameters are introduced below.

[0035] 1. Welding power: the output power of the laser beam.

[0036] 2. Welding speed: the speed at which the laser beam moves along the welding joint.

[0037] 3. Galvanometer focal length: The focal length of the laser beam focused to the welding area after passing through the galvanometer system.

[0038] 4. Welding parts thickness: the thickness of the material being welded.

[0039] 5. Spot size: The diameter of the spot formed after the laser beam is focused on the working surface.

[0040] 6. Amplitude: The maximum offset used by the laser scanning galvanometer during the scanning process.

[0041] S203, inputting the welding parameter sequence into the statistical model, solving it according to the target value of the parameter, and predicting the welding quality state of the weldment after completing the laser welding.

[0042] In the specific implementation process, the statistical model is used to indicate the corresponding relationship between the key welding parameters and the welding quality state. As long as the welding parameter sequence including the key welding parameters is input into the statistical model, the welding quality state of the weldment after laser welding can be predicted, which may be a normal state or an abnormal state. The abnormal state includes the undercut state, the non-penetration state, the crack state and the deformation state. The following introduces each state separately.

[0043] (1) Normal state: The weld joint is uniform and strong, the weld width and height meet the design requirements, the weld surface is flat, and there are no obvious defects in the welding.

[0044] (2) Undercut condition: Local deep cutting or depression occurs on the edge of the weld. This is usually caused by excessive heat, which causes the edge of the base material to be overheated, resulting in overflow of the molten pool or excessive melting.

[0045] (3) Non-penetration state: During the welding process, the molten pool fails to fully penetrate the bottom of the joint, resulting in the weld not completely penetrating the thickness of the workpiece, causing insufficient strength of the weld joint.

[0046] (4) Crack state: Cracks appear at the weld joint, usually caused by rapid temperature changes or concentrated thermal stress during the welding process.

[0047] (5) Deformation state: During the welding process, due to uneven thermal expansion and cooling contraction, the welding part bends or warps, affecting the size and shape of the workpiece.

[0048] In a possible embodiment, the parameters of the statistical model include the initial state probability matrix π, the state transition matrix A and the observation probability matrix B, the steps of solving the target values ​​of the parameters of the statistical model include: Initialize the parameters of the statistical model; continuously update the values ​​of the parameters of the statistical model through the forward-backward algorithm until the maximum number of iterations is reached to obtain the target values ​​of the parameters of the statistical model.

[0049] Please refer to Figure 3 , which is a schematic diagram of the process of the welding quality state prediction method based on the statistical model provided in the embodiment of the present application Figure 2 .

[0050] S301, define welding quality status sequence S ={ s 1, s 2, ... , s N}.

[0051] in, s N Indicates NWelding quality status, N is the total number of welding quality states. Since the welding quality states include normal state, undercut state, non-penetration state, crack state and deformation state, N=5.

[0052] S302, define welding parameter sequence O= { o 1, o 2, ... , o M}.

[0053] in, o M Indicates M The key welding parameters are: M is the total number of key welding parameters, which include welding power, welding speed, galvanometer focal length, welding part thickness, spot size and amplitude. M =6.

[0054] S303, initializing statistical model f= ( A,B,π ) A , B and π The value of .

[0055] Initialize the state transfer matrix A [ a ij ], observation probability matrix B[ b j ( o k )] and the initial state probability matrix π( π 1, π 2, ... , π N ),in, π i Indicates i Welding quality status s i The initial state probability, i∈ [1 ,N ]. a ij Indicates that from i Welding quality status s i Transfer to j Welding quality status s j The probability of b j ( o k ) indicates thej Welding quality status s j The next observation k Key welding parameters o k The probability of k∈ [1 ,M ].

[0056] S304, using minimum and maximum normalization to process the initial state probability matrix π and the state transfer matrix A , scale its data to the interval [0,1].

[0057] In order to reduce the data discreteness of the random matrix and reduce the error caused by data discreteness in the parameter solution of the statistical model, the initial state probability matrix is ​​normalized to obtain the normalized initial state probability matrix, and the state transfer matrix is ​​normalized to obtain the normalized state transfer matrix. The normalization formula is as follows:

[0058] in, π norm represents the normalized initial state probability matrix, Indicates i Welding quality status s i The initial state probability, Min ( π ) represents the minimum value in the initial state probability matrix π, Max ( π ) represents the maximum value in the initial state probability matrix π. A norm represents the normalized state transfer matrix, a ij Indicates that from i Welding quality status s i Transfer to j Welding quality status s j The probability of Min ( A ) represents the minimum value in the state transfer matrix A, Max ( A ) represents the maximum value in the state transfer matrix A, N is the total number of welding quality conditions.

[0059] S305 , solving the values ​​of the parameters A, B and π of the statistical model through a forward algorithm.

[0060]

[0061] in, Indicates that the first k key welding parameters in the welding parameter sequence are observed i Welding quality status s i The probability of Indicates that the first k-1 key welding parameters are observed j Welding quality status s j The probability of Indicates that from j Welding quality status s j Transfer to i Welding quality status s i The probability of b i ( o k ) indicates the i Welding quality status s i The next observation k Key welding parameters o k The probability of, N is the total number of welding quality states, N=5.

[0062] S306: Solving the parameters of the statistical model through the backward algorithm A , B and π Value 。

[0063]

[0064] in, Indicates that the first k key welding parameters in the welding parameter sequence are observed i The welding quality status is Under the condition of i Welding quality status s i probability. Indicates that the first k+1 key welding parameters in the welding parameter sequence are observed j The welding quality status is s j Under the condition of j Welding quality status s j probability. aij Indicates that from i Welding quality status s i Transfer to j Welding quality status s j The probability of Indicates j Welding quality status s j The next observation k+ 1 Key welding parameters o k+1 probability.

[0065] S307: Determine whether the number of iterations is less than or equal to the maximum number of iterations.

[0066] If the number of iterations is greater than the maximum number of iterations (for example, 200 times), S308 is executed. If the number of iterations is less than or equal to the maximum number of iterations (for example, 200 times), the process returns to S305 and continues to execute.

[0067] S308, parameters A, B and π The value of is determined as the target value of the parameter of the statistical model.

[0068] When the number of iterations is greater than the maximum number of iterations (for example, 200), the parameters A, B and π The value of is the target value of the parameter of the statistical model.

[0069] In the embodiment of the present application, the hidden state in the statistical model cannot be directly observed, and the forward-backward algorithm effectively estimates the probability distribution of the hidden state through the forward and backward processes, making the learning process of the model more reliable. By iteratively updating the parameters, the model will gradually approach the optimal solution, thereby improving the accuracy of the prediction.

[0070] In a possible embodiment, the specific steps of S202 include: The welding parameter sequence is input into the statistical model, and the solution is performed according to the target value of the parameter to obtain the maximum probability in the probability distribution of the welding quality state sequence; the welding quality state corresponding to the maximum probability is determined as the welding quality state of the weldment after laser welding is completed.

[0071] Please refer to Figure 4 , which is a schematic diagram of the process of predicting welding quality status based on the statistical model provided in the embodiment of the present application Figure 3 .

[0072] S401, initialization observes the first key welding parameter in the welding parameter sequence i The probability of a certain welding quality state.

[0073]

[0074] in, Indicates the welding parameter sequence O= { o 1, o 2, ... , o M The first key welding parameter o 1 Observe the i Welding quality status s i The probability of Indicates i Welding quality status s i The initial state probability, Indicates i Welding quality status s i The first critical welding parameter was observed under The probability of, N is the total number of welding quality states, N=5.

[0075] S402. Recursively deduce the probability sequence of each welding quality state observed under the first k key welding parameters in the welding parameter sequence according to the conditional probability.

[0076]

[0077] in, Indicates the welding parameter sequence O= { o 1, o 2, ... , o M}The first k key welding parameters are observed i Welding quality status s i The probability sequence of Indicates the welding parameter sequence O= { o 1, o 2, ... , o M}The first k-1 key welding parameters were observed j Welding quality status s j The probability sequence of Indicates that from j Welding quality status s j Transfer to i Welding quality status si The probability of b i ( o k ) indicates the i Welding quality status s i The next observation k Key welding parameters o k The probability of, M is the total number of key welding parameters, M=6.

[0078] S403: Obtain the maximum probability in the probability distribution.

[0079]

[0080] in, represents the maximum probability, Indicates the welding parameter sequence O= { o 1, o 2, ... , o M}The first k key welding parameters are observed i Welding quality status s i The probability sequence of .

[0081] S404. The welding quality state corresponding to the maximum probability is the welding quality state corresponding to the solved welding parameter sequence.

[0082] The maximum probability The corresponding i Welding quality status s i Determined as welding parameter sequence O= { o 1, o 2, ... , o M}The corresponding welding quality status, that is, the welding quality status of the weldment after laser welding is completed.

[0083] In an embodiment of the present application, the most likely welding quality state can be determined by calculating the maximum probability in the probability distribution. This method selects the most likely state from all possible quality states by maximizing the probability, thereby providing a reasonable basis for evaluating the welding quality.

[0084] In a possible embodiment, after inputting the welding parameter sequence into the statistical model, solving according to the optimal parameters of the statistical model, and predicting the welding quality state of the weldment after laser welding, the method further includes: If the welding quality status is abnormal, the key welding parameters of the weldment during the laser welding process are adjusted according to the welding quality status; and laser welding of the weldment is continued according to the adjusted key welding parameters.

[0085] In the specific implementation process, after the welding quality state of the weldment after laser welding is completed is predicted by the statistical model, if the predicted welding quality state is normal, the weldment is continued to be laser welded according to the previous key welding parameters. If the predicted welding quality state is abnormal, the abnormal state includes the undercut state, the non-penetration state, the crack state and the deformation state, the previous key welding parameters can be adjusted according to the welding quality state, and the weldment is continued to be laser welded according to the adjusted key welding parameters.

[0086] Since the thickness of the welded parts cannot be adjusted, the key welding parameters to be adjusted include welding power, welding speed, galvanometer focal length, spot size and amplitude. Due to the welding quality status of the weldment, the methods of adjusting the key welding parameters are different, which are introduced below.

[0087] The first situation is the biting edge state.

[0088] If the welding quality is undercut, the welding power can be appropriately reduced to avoid excessive heat concentration on the edge. Increase the welding speed to reduce the heat input per unit time to prevent excessive heating of the edge of the base material. Adjust the focal length of the galvanometer to ensure that the laser beam is more concentrated, reduce excessive heat dispersion, and avoid excessive heating of the edge area. Increase the spot size to reduce excessive heat concentration in the joint area during welding and avoid excessive burn-through.

[0089] The second situation is the non-penetration state.

[0090] If the welding quality status is not penetrated, you can increase the welding power, increase the current or power, and ensure that the heat of the molten pool is deep enough into the base material. Reduce the welding speed and extend the heating time of each point so that the molten pool has enough time to penetrate to the bottom. Properly adjust the focal length of the galvanometer to ensure that the laser beam is focused at the best position and increase the depth of the welding molten pool. Reduce the spot size to make the energy more concentrated and increase the penetration depth of the welding. Increase the amplitude to help the heat distribution more evenly and prevent the local temperature from being too low to cause non-penetration.

[0091] The third situation is the crack state.

[0092] If the welding quality state is a crack state, you can reduce the welding power, reduce the local temperature gradient, and avoid cracks caused by excessive temperature differences. Appropriately reduce the welding speed, extend the welding time, reduce the cooling rate, and reduce thermal stress. Adjust the focal length of the galvanometer to make the focus of the laser beam in the best position to avoid cracks caused by overheating. Appropriately increase the spot size to ensure uniform heat distribution and reduce local overheating. Increase the amplitude to help homogenize the molten pool, reduce the temperature difference in the welding area, and reduce the risk of cracks.

[0093] The fourth situation is the deformed state.

[0094] If the welding quality state is a crack state, you can reduce the welding power, reduce local heating, avoid local overheating and excessive thermal expansion, and cause deformation. Increase the welding speed, reduce the heat input per unit time, and avoid deformation caused by excessive heating. Adjust the focal length of the galvanometer to ensure that the energy distribution of the laser beam is uniform and reduce local deformation caused by heat concentration. Increase the spot size, reduce energy concentration, help heat to be evenly distributed, and thus reduce deformation. Appropriately increase the amplitude to help heat be evenly distributed and reduce deformation caused by local overheating.

[0095] In the embodiment of the present application, by adjusting the key welding parameters according to the welding quality status and continuing the welding process, the problems of undercut, non-penetration, cracks and deformation of the weldment are avoided, and the welding process is ensured to return to normal state. By timely adjustment, the occurrence of defective products is reduced, thereby improving the welding quality. And it can reduce downtime and rework time, thereby improving welding efficiency. By reasonably adjusting the key welding parameters during the welding process, an efficient, accurate and stable welding process can be achieved, thereby meeting various complex and precise welding requirements.

[0096] In a possible embodiment, after continuing to perform laser welding on the weldment, the method further includes: According to the adjusted key welding parameters, an adjusted welding parameter sequence is constructed; the adjusted welding parameter sequence is input into the statistical model, and the solution is performed according to the target value of the parameter to predict the welding quality state of the weldment after the parameters are adjusted and the laser welding is completed.

[0097] In an embodiment of the present application, during the welding process, the welding quality status is closely monitored. After adjusting the key welding parameters, the welding quality status is predicted based on the adjusted key welding parameters to confirm whether the adjusted key welding parameters are effective, thereby ensuring that the final welding quality meets the standards.

[0098] To sum up, considering that the quality state is closely related to the thickness of the weldment and the welding parameters of the laser welding process, the present application provides a welding quality state prediction method based on a statistical model. By real-time monitoring of the key welding parameters of the welding process, the laser welding quality state of the input key welding parameters is predicted based on the statistical model. According to the predicted welding quality state, it is determined whether the key welding parameters need to be adjusted, thereby improving the welding quality and welding efficiency.

[0099] Based on the same invention concept, please refer to Figure 5 The present application also provides a welding quality state prediction device based on a statistical model, the device comprising: A model building module is used to build a statistical model and solve the target value of the parameters of the statistical model; the statistical model is used to indicate the corresponding relationship between the key welding parameters and the welding quality status; A sequence building module is used to build a welding parameter sequence according to the key welding parameters of the weldment during the laser welding process; The model prediction module is used to input the welding parameter sequence into the statistical model, solve it according to the target value of the parameter, and predict the welding quality state of the weldment after laser welding is completed.

[0100] It should be noted that each module in the welding quality status prediction device based on the statistical model in this embodiment corresponds one-to-one to each step in the welding quality status prediction method based on the statistical model in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned welding quality status prediction method based on the statistical model, which will not be repeated here.

[0101] Based on the same inventive concept, the present application also provides a computer device, which includes a processor, a memory, and a computer program stored in the memory. When the computer program is executed by the processor, the aforementioned statistical model-based welding quality status prediction method is implemented.

[0102] Based on the same inventive concept, the present application also provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the aforementioned welding quality state prediction method based on a statistical model is implemented.

[0103] In some embodiments, the computer readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or various devices including one or any combination of the above memories. The computer may be various computing devices including intelligent terminals and servers.

[0104] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0105] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file storing other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0106] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.

[0107] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0108] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0109] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for a multimedia terminal device (which can be a mobile phone, a computer, a television receiver, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0110] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the specification and accompanying drawings of the present application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A welding quality status prediction method based on a statistical model, characterized in that: include: Constructing a statistical model and solving the target values ​​of the parameters of the statistical model; the statistical model is used to indicate the corresponding relationship between the key welding parameters and the welding quality status; According to the key welding parameters of the weldment during the laser welding process, a welding parameter sequence is constructed; The welding parameter sequence is input into the statistical model, and the solution is performed according to the target value of the parameter to predict the welding quality state of the weldment after the laser welding is completed.

2. The welding quality state prediction method based on the statistical model according to claim 1, characterized in that: The key welding parameters include welding power, welding speed, galvanometer focal length, welding part thickness, spot size and amplitude.

3. The welding quality state prediction method based on the statistical model according to claim 1, characterized in that: The step of solving the target value of the parameter of the statistical model comprises: Initializing the parameters of the statistical model; the parameters include an initial state probability matrix, a state transfer matrix and an observation probability matrix; The value of the parameter is continuously updated through a forward-backward algorithm until a maximum number of iterations is reached to obtain a target value of the parameter.

4. The welding quality state prediction method based on the statistical model according to claim 1, characterized in that: The step of inputting the welding parameter sequence into the statistical model, solving the problem according to the target value of the parameter, and predicting the welding quality state of the weldment after laser welding is completed includes: Inputting the welding parameter sequence into the statistical model, solving it according to the target value of the parameter, and obtaining the maximum probability in the probability distribution of the welding quality state sequence; The welding quality state corresponding to the maximum probability is determined as the welding quality state of the weldment after laser welding is completed.

5. The welding quality state prediction method based on the statistical model according to claim 4, characterized in that: The welding quality state sequence includes a normal state, an undercut state, a non-penetration state, a crack state and a deformation state.

6. The welding quality state prediction method based on statistical model according to claim 1, characterized in that: After inputting the welding parameter sequence into the statistical model, solving according to the target value of the parameter, and predicting the welding quality state of the weldment after laser welding, the method further includes: If the welding quality state is an abnormal state, adjusting key welding parameters of the weldment during the laser welding process according to the welding quality state; The laser welding of the weldment is continued according to the adjusted key welding parameters.

7. The welding quality status prediction method based on the statistical model according to claim 6, characterized in that: After continuing to laser weld the weldment according to the adjusted key welding parameters, the method further includes: constructing an adjusted welding parameter sequence according to the adjusted key welding parameters; The adjusted welding parameter sequence is input into the statistical model, and the solution is performed according to the target value of the parameter to predict the welding quality state of the weldment after the parameters are adjusted and the laser welding is completed.

8. A welding quality status prediction device based on a statistical model, characterized in that: include: A model building module, used to build a statistical model and solve the target value of the parameters of the statistical model; the statistical model is used to indicate the corresponding relationship between the key welding parameters and the welding quality status; A sequence building module is used to build a welding parameter sequence according to the key welding parameters of the weldment during the laser welding process; The model prediction module is used to input the welding parameter sequence into the statistical model, solve it according to the target value of the parameter, and predict the welding quality state of the weldment after completing laser welding.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the welding quality status prediction method based on a statistical model as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the processor executes the computer program to implement the welding quality status prediction method based on the statistical model as described in any one of claims 1 to 7.

Images