Method, device and equipment for detecting welding signal of welding spot of vehicle body and storage medium

Abstract

The invention relates to the technical field of welding quality detection, and discloses a detection method, device and equipment for a welding signal of a welding spot of a vehicle body and a storage medium. According to the embodiment of the invention, the accuracy of the parameter value is improved by means of preprocessing and cross validation of the parameter value, so that the accuracy of the target welding signal obtained based on the parameter value is ensured, and finally the accuracy of the difference value output based on the target welding signal and the final state is ensured. All the target welding signals are processed through the preset prediction model, the output difference values are matched with the preset threshold value, and therefore it is guaranteed that the state finally determined according to the matching result is consistent. And by adopting the method, the welding signal generated in the welding process of the automobile body welding spot is automatically detected, manual visual inspection is not needed, and the efficiency of detecting the welding signal of the welding spot is greatly improved.

Description

Technical Field

[0001] This application relates to the technical field of welding quality detection, and particularly to a method, device, equipment and storage medium for detecting welding signals of vehicle body weld points. Background Art

[0002] During the automobile manufacturing process, the quality of weld points directly affects the overall strength and safety of the vehicle body. Traditional methods for detecting the quality of weld points often rely on manual visual inspection or destructive testing. These methods are not only inefficient but also difficult to ensure the accuracy of the detection results. Summary of the Invention

[0003] In view of the above problems, this application provides a method, device, equipment and storage medium for detecting welding signals of vehicle body weld points, which is used to solve the technical problems in the prior art that are not only inefficient but also difficult to ensure the accuracy of the detection results, and realizes the state detection of the target welding signal. By preprocessing the parameter values and cross-verifying them, the accuracy of the parameter values is improved, thereby ensuring the accuracy of the target welding signal obtained based on the parameter values, and finally ensuring the accuracy of the difference value and the final state output based on the target welding signal. It realizes that all target welding signals are processed using a preset prediction model, and the output difference values are all matched with a preset threshold, so as to ensure the consistency of the state finally determined according to the matching result. And this method is used to automatically detect the welding signals generated during the welding process of vehicle body weld points without manual visual inspection, greatly improving the efficiency of detecting welding signals of weld points.

[0004] According to one aspect of the embodiments of this application, a method for detecting welding signals of vehicle body weld points is provided. The method includes: collecting an initial welding signal generated during the welding process of a vehicle body weld point; wherein, the initial welding signal includes the collection time and the initial welding parameter value corresponding to the collection time; preprocessing the initial welding parameter value to obtain a preprocessed transitional welding parameter value; based on the collection time, cross-verifying the specific parameter values in the transitional welding parameter value to obtain a qualified transitional welding parameter value, and using the transitional welding parameter value as the target welding parameter value; inputting the target welding signal into a preset prediction model, and detecting whether the difference value output by the preset prediction model matches a preset threshold; wherein, the target welding signal includes the collection time and the target welding parameter value corresponding to the collection time; according to the matching result, determining the state of the target welding signal corresponding to the difference value.

[0005] In an alternative manner, the initial welding parameter values include an initial current value, an initial voltage value, and an initial dynamic resistance value; the step of preprocessing the initial welding parameter values to obtain the preprocessed transition welding parameter values further includes: performing a filtering process on the initial current value to obtain a transition current value; performing a voltage calibration process on the initial voltage value to obtain a transition voltage value; and analyzing the initial dynamic resistance value based on a preset analysis method to perform a denoising process on the initial dynamic resistance value to obtain a transition dynamic resistance value.

[0006] In an alternative manner, the specific parameter values in the transition welding parameter values include a transition current value, a transition voltage value, and a transition dynamic resistance value; the step of cross-verifying the specific parameter values in the transition welding parameter values based on the acquisition time to obtain the qualified transition welding parameter values further includes: performing pairwise correlation verification on the transition current value, the transition voltage value, and the transition dynamic resistance value at the same acquisition time; if the pairwise correlation verifications are all qualified, it is determined that the transition current value, the transition voltage value, and the transition dynamic resistance value corresponding to the acquisition time are all verified to be qualified.

[0007] In an alternative manner, the step of performing pairwise correlation verification on the transition current value, the transition voltage value, and the transition dynamic resistance value at the same acquisition time further includes: combining the transition current value, the transition voltage value, and the transition dynamic resistance value at the same acquisition time pairwise to obtain three combinations each including two values; detecting whether the two values in each combination conform to the preset relationship of the corresponding combination, and determining that the verification is qualified when it is detected that they conform to the preset relationship; wherein the preset relationships include a preset voltage-current relationship, a preset current-resistance relationship, and a preset voltage-resistance relationship; if the three combinations corresponding to the same acquisition time are all verified to be qualified, it is determined that the pairwise correlation verifications are all qualified.

[0008] In an alternative manner, the step of inputting the target welding signal into a preset prediction model and detecting whether the difference value output by the preset prediction model matches a preset threshold further includes: using a preset neural network to identify and extract the timing features in the target welding signal; establishing a reconstructed sample signal according to the timing features; and calculating the difference value between the reconstructed sample signal and the target welding signal.

[0009] In an alternative manner, the state includes a normal welding state and an abnormal welding state; the step of determining the state of the target welding signal corresponding to the difference value according to the matching result further includes: if the matching result matches a preset threshold, determining that the target welding signal corresponding to the difference value is in the normal welding state; if the matching result does not match the preset threshold, determining that the target welding signal corresponding to the difference value is in the abnormal welding state.

[0010] In an alternative manner, after the step of, if the matching result does not match the preset threshold, determining that the target welding signal corresponding to the difference value is in the abnormal welding state, the method further includes: obtaining the position information corresponding to the target welding signal; wherein the position information includes the acquisition station information of the target welding signal and the line information of the station where the station is located; establishing abnormal alarm information according to the position information and the abnormal welding state; wherein the abnormal alarm information includes a line stop instruction; sending the abnormal alarm information to the Internet of Things, so that the Internet of Things issues the line stop instruction according to the abnormal alarm information, and sends the abnormal alarm information to the staff user terminal for reminder.

[0011] According to another aspect of the embodiments of the present application, there is provided a detection device for a welding signal of a vehicle body solder joint, including: a signal acquisition module, configured to acquire an initial welding signal generated during the welding of the vehicle body solder joint; wherein the initial welding signal includes an acquisition time and an initial welding parameter value corresponding to the acquisition time; a preprocessing module, configured to preprocess the initial welding parameter value to obtain a preprocessed transition welding parameter value; a cross-validation module, configured to perform cross-validation on specific parameter values in the transition welding parameter value based on the acquisition time to obtain a qualified transition welding parameter value through verification, and use the transition welding parameter value as a target welding parameter value; a model prediction module, configured to input a target welding signal into a preset prediction model and detect whether a difference value output by the preset prediction model matches a preset threshold; wherein the target welding signal includes the acquisition time and a target welding parameter value corresponding to the acquisition time; a state determination module, configured to determine the state of the target welding signal corresponding to the difference value according to the matching result.

[0012] According to another aspect of the embodiments of the present application, there is provided a device, including: a controller; a memory, configured to store one or more programs, and when the one or more programs are executed by the controller, enable the controller to implement the method for detecting a welding signal of a vehicle body solder joint described in any one of the above claims.

[0013] According to another aspect of the embodiments of the present application, there is provided a computer-readable storage medium storing at least one executable instruction, which, when running on a device / equipment, causes the device / equipment to perform the operations of the method for detecting the welding signal of the vehicle body solder joint as described in any one of the above claims.

[0014] In the embodiments of the present application, the initial welding parameter values included in the collected initial welding signal are preprocessed to obtain transition welding parameter values; based on the acquisition time, the specific parameter values in the transition welding parameter values are cross-validated, and the transition welding parameter values that pass the validation are used as target welding parameter values; the target welding signal composed of the target welding parameter values and the corresponding acquisition time is input into a preset prediction model, and it is detected whether the difference value output by the preset prediction model matches a preset threshold; finally, according to the matching result, the state of the target welding signal is determined. Thus, by preprocessing and cross-validating the parameter values, the accuracy of the parameter values is improved, and further the accuracy of the target welding signal obtained based on the parameter values is ensured, and finally the accuracy of the difference value output based on the target welding signal and the final state is ensured. By processing all target welding signals with a preset prediction model and matching the output difference values with a preset threshold, the consistency of the state finally determined according to the matching result is ensured. And by using this method, the welding signals generated during the welding process of the vehicle body solder joints are automatically detected without manual visual inspection, greatly improving the efficiency of detecting the welding signals of the solder joints.

[0015] The above description is only an overview of the technical solutions of the embodiments of the present application. In order to be able to understand the technical means of the embodiments of the present application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features and advantages of the embodiments of the present application more obvious and understandable, the specific embodiments of the present application are given below. Brief Description of the Drawings

[0016] The drawings are only used to illustrate the embodiments and are not considered as a limitation to the present application. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0017] Figure 1 A flowchart showing an embodiment of the method for detecting the welding signal of the vehicle body solder joint provided by the present application is shown;

[0018] Figure 2 A graph showing the change curves of current, voltage, and dynamic resistance in the initial welding signal in an embodiment is shown;

[0019] Figure 3 A flowchart showing another embodiment of the method for detecting the welding signal of the vehicle body solder joint provided by the present application is shown;

[0020] Figure 4 Shows a schematic flowchart of another embodiment of the method for detecting the welding signal of vehicle body welding points provided by the present application;

[0021] Figure 5 Shows a schematic flowchart of still another embodiment of the method for detecting the welding signal of vehicle body welding points provided by the present application;

[0022] Figure 6 Shows a schematic flowchart of another embodiment of the method for detecting the welding signal of vehicle body welding points provided by the present application;

[0023] Figure 7 Shows a schematic flowchart of another embodiment of the method for detecting the welding signal of vehicle body welding points provided by the present application;

[0024] Figure 8 Shows a schematic flowchart of still another embodiment of the method for detecting the welding signal of vehicle body welding points provided by the present application;

[0025] Figure 9 Shows a schematic flowchart of another embodiment of the method for detecting the welding signal of vehicle body welding points provided by the present application;

[0026] Figure 10 Shows a schematic structural diagram of an embodiment of the device for detecting the welding signal of vehicle body welding points provided by the present application;

[0027] Figure 11 Shows a schematic structural diagram of an embodiment of the device provided by the present application. Detailed Description of the Invention

[0028] Here, exemplary embodiments will be described in detail, and examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0029] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0030] The flowcharts shown in the accompanying drawings are merely illustrative and not necessarily inclusive of all content and operations / steps, nor are they necessarily to be executed in the order described. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may change according to the actual situation.

[0031] As used in this application, "a plurality of" means two or more. "And / or" describes the association relationship of associated objects and indicates that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0032] During the automobile manufacturing process, the quality of welding spots directly affects the overall strength and safety of the vehicle body. Traditional methods for detecting the quality of welding spots often rely on manual visual inspection or destructive testing. These methods are not only inefficient but also difficult to ensure the accuracy and consistency of the detection results. Therefore, it is of great significance to develop an efficient and accurate method for detecting the quality of vehicle body welding spots.

[0033] The automotive body-in-white is mainly formed by connecting stamping parts through resistance spot welding technology, and resistance spot welding accounts for more than 90% of the total welding volume of the body-in-white. During the welding process, welding parameters such as voltage, current, dynamic resistance, and welding time directly affect the strength of the welding spots. Therefore, precise control and detection of these parameters are the key to ensuring welding quality.

[0034] Patent CN 114813838 A discloses a method and system for detecting the quality of welding spots based on dynamic resistance signals. The invention collects dynamic resistance signals at each workstation in the welding shop, constructs a large database of vehicle body welding spot quality information, and uses the low-rank sparse decomposition method to construct a reference curve, and establishes an evaluation index for the stability of the welding process based on this curve. Further, a one-dimensional convolutional neural network model based on channel attention mechanism and residual connection is developed for welding spot quality detection and classification. This method combines deep learning algorithms and information on the physical process of spot welding, improving the accuracy and reliability of welding spot quality detection.

[0035] Welding parameters that affect welding quality during the welding process include welding current, welding voltage, dynamic resistance, welding duration, and other related parameters. Patent CN 114813838 A only provides a detection method based on dynamic resistance and cannot comprehensively describe the quality of welding.

[0036] Patent CN 114813838 A constructs a large database of body weld quality and establishes evaluation indicators for welding process stability based on the database. The accuracy of weld quality detection in this solution is too dependent on the integrity of the data. If errors, omissions or delays occur during data collection, it will affect the construction of subsequent reference curves and the evaluation of welding process stability.

[0037] One of the purposes of this technical solution is to comprehensively monitor welding process parameters, including welding current, welding voltage, and dynamic resistance, and provide a more complete welding signal detection solution.

[0038] The second purpose of this technical solution is to propose an anomaly detection algorithm combining VAE and LSTM to improve the accuracy of welding quality detection.

[0039] The third purpose of the technical solution is to propose a method for intercepting the quality control line of a solder joint. When the online monitoring system for solder joint quality detects that there is a quality problem with the solder joint, the line body can be linked to realize the function of intercepting the abnormal solder joint control line.

[0040] In order to explain the technical solution in detail, the following embodiments are used for detailed description:

[0041] Figure 1 The flowchart of an embodiment of the method for detecting welding signals of a body welding spot of the present application is shown, and the method is executed by a computer device. Figure 1 As shown, the method comprises the following steps:

[0042] Step S110: collecting initial welding signals generated by the welding points of the vehicle body during the welding process.

[0043] The initial welding signal includes the acquisition time and the initial welding parameter value corresponding to the acquisition time. The initial welding parameter value includes the initial current value, the initial voltage value, and the initial dynamic resistance value. The initial and subsequent transitions and targets here are all pre-defined to distinguish the welding parameter values ​​after different processing, and they have no meaning in themselves.

[0044] Specifically, a welding machine IOT has built-in high-precision sensors that can accurately collect key parameters such as current, voltage, dynamic resistance, etc. during the welding process in real time. These sensors can ensure the accuracy and reliability of the data. The collected initial welding signal can be established as follows Figure 2 The curve change diagram of welding current, voltage and dynamic resistance is shown. The initial welding signal is displayed more intuitively, that is, the initial welding parameter values ​​are arranged in the time sequence of the corresponding acquisition time to obtain a change curve diagram.

[0045] Step S120: preprocessing the initial welding parameter values ​​to obtain preprocessed transition welding parameter values.

[0046] Among them, the preprocessing includes filtering, smoothing the curve, and denoising, etc. The transitional welding parameter values include transitional current value, transitional voltage value, and transitional dynamic resistance value.

[0047] Specifically, the data cleaning of the welding current curve, welding voltage curve, and dynamic resistance curve is a key link to ensure the accuracy of welding quality assessment and process control. The accuracy of these data curves directly affects the reliability and stability of the welding results. Therefore, it is necessary to preprocess the initial welding parameter values to obtain more accurate, stable, and reliable transitional welding parameter values.

[0048] Step S130: Based on the acquisition time, cross-verify the specific parameter values in the transitional welding parameter values to obtain the qualified transitional welding parameter values, and use the transitional welding parameter values as the target welding parameter values.

[0049] Among them, the target welding parameter values include target current value, target voltage value, and target dynamic resistance value;

[0050] Specifically, after the cleaning of the above three data curves, cross-validation and correlation analysis are also required to ensure that the logical relationship between the current, voltage, and dynamic resistance data is reasonable and consistent. Because there is a natural physical relationship between the current, voltage, and dynamic resistance, the current value, voltage value, and dynamic resistance value at the same acquisition time are pairwise correlated. This process not only improves the accuracy of the data but also provides a solid data foundation for subsequent welding quality control.

[0051] Step S140: Input the target welding signal into the preset prediction model, and detect whether the difference value output by the preset prediction model matches the preset threshold.

[0052] Among them, the target welding signal includes the acquisition time and the target welding parameter values corresponding to the acquisition time; the preset threshold is to generate a threshold interval based on historical data for judging whether a sample is abnormal.

[0053] Specifically, the preset prediction model is used to predict the difference value corresponding to the input target welding signal. The difference value is the error between the reconstructed sample established based on the target welding signal and the target welding signal. Calculate the error between the original signal sample and the reconstructed sample. This error reflects the reconstruction ability of the model for the original image and indirectly reflects the abnormality degree of the signal sample.

[0054] Step S150: According to the matching result, determine the state of the target welding signal corresponding to the difference value.

[0055] Specifically, if the difference value matches the preset threshold, it indicates that the state of the target welding signal corresponding to the difference value is the welding abnormal state.

[0056] Beneficial effects: By preprocessing the initial welding parameter values included in the collected initial welding signals, transitional welding parameter values are obtained; based on the acquisition time, cross-validation is performed on the specific parameter values in the transitional welding parameter values, and the verified transitional welding parameter values are used as target welding parameter values; the target welding signals composed of the target welding parameter values and the corresponding acquisition times are input into a preset prediction model, and it is detected whether the difference values output by the preset prediction model match a preset threshold; finally, according to the matching result, the state of the target welding signal is determined.

[0057] Thus, by means of preprocessing and cross-validation of the parameter values, the accuracy of the parameter values is improved, thereby ensuring the accuracy of the target welding signals obtained based on the parameter values, and finally ensuring the accuracy of the difference values and the final state output based on the target welding signals.

[0058] By processing all target welding signals with a preset prediction model and matching the output difference values with the preset threshold, the consistency of the state finally determined according to the matching result is ensured. And by using this method, the welding signals generated during the welding process of the body solder joints are automatically detected, without manual visual inspection, greatly improving the efficiency of solder joint welding signal detection.

[0059] In some embodiments, the initial welding parameter values include an initial current value, an initial voltage value, and an initial dynamic resistance value; as Figure 3 shown, step S120 further includes:

[0060] Step S121: Perform filtering processing on the initial current value to obtain a transitional current value.

[0061] Specifically, for the initial current value, that is, Figure 2 the welding current curve in, abnormal data points caused by equipment transient response and line interference need to be carefully removed. By comparing the theoretical current range with the measured data, a filtering algorithm (such as median filtering, low-pass filtering) is used to smooth the curve and remove noise to ensure that the current curve can truly reflect the current change trend during the welding process.

[0062] Step S122: Perform voltage calibration processing on the initial voltage value to obtain a transitional voltage value.

[0063] Specifically, for the initial voltage value, that is, Figure 2 the welding voltage curve in, the abnormal fluctuations of the welding voltage curve are due to unstable power supply, poor cable contact, or the thermoelectric effect in the welding area. During the cleaning process, the voltage data needs to be calibrated to eliminate measurement errors, and data points that significantly deviate from the normal voltage range are identified and removed through algorithms.

[0064] Step S123: Analyze the initial dynamic resistance value based on a preset analysis method to denoise the initial dynamic resistance value and obtain a transitional dynamic resistance value.

[0065] Specifically, during the welding process, the dynamic resistance is affected by various factors, including the contact resistance between the workpieces, the contact resistance between the electrode and the workpieces, and the change in the resistance of the workpieces themselves. During the cleaning process, time-domain and frequency-domain analyses are performed on the resistance data (i.e., the initial dynamic resistance value) to identify and remove the non-linear changes and system noise caused by temperature changes, thereby obtaining the transitional dynamic resistance value.

[0066] Beneficial effects: The initial current value, the initial voltage value, and the initial dynamic resistance value are preprocessed respectively. By filtering and calibrating the initial welding parameter values, the accuracy of the parameter values is improved, thereby ensuring the accuracy of the target welding signal obtained based on the parameter values, and finally ensuring the accuracy of the difference value and the final state output based on the target welding signal.

[0067] In some embodiments, the specific parameter values in the transitional welding parameter values include a transitional current value, a transitional voltage value, and a transitional dynamic resistance value; as Figure 4 shown, step S130 further includes:

[0068] Step S131: Perform pairwise correlation verification on the transitional current value, the transitional voltage value, and the transitional dynamic resistance value at the same acquisition time.

[0069] Specifically, because there is a natural physical relationship between current, voltage, and dynamic resistance, the current value, voltage value, and dynamic resistance value at the same acquisition time are pairwise correlated.

[0070] Step S132: If all pairwise correlation verifications are qualified, determine that the transitional current value, the transitional voltage value, and the transitional dynamic resistance value corresponding to the acquisition time are all verified to be qualified.

[0071] Specifically, the current value, voltage value, and dynamic resistance value at the same acquisition time should be pairwise verified to be qualified. If any two of them are not qualified in the verification, the data at the corresponding acquisition moment is problematic. Therefore, only the transitional welding parameter values that are qualified in the pairwise verification of any two of them are qualified.

[0072] Beneficial effects: Further refine the steps of cross-verifying the specific parameter values in the transitional welding parameter values based on the acquisition time. By performing pairwise correlation verification on the transitional current value, the transitional voltage value, and the transitional dynamic resistance value at the same acquisition time, cross-verification of the specific parameter values in the transitional welding parameter values is achieved. By verifying the transitional welding parameter values, the accuracy of the transitional welding parameter values is further improved.

[0073] In some embodiments, such asFigure 5 As shown, step S131 further includes:

[0074] Step S210: Combine the transition current value, transition voltage value, and transition dynamic resistance value at the same acquisition time in pairs to obtain three combinations each including two values.

[0075] Specifically, the three combinations corresponding to the same acquisition time are: the combination of the transition current value and the transition voltage value, the combination of the transition voltage value and the transition dynamic resistance value, and the combination of the transition current value and the transition dynamic resistance value.

[0076] Step S220: Detect whether the two values in each combination conform to the preset relationship of the corresponding combination, and determine that the verification is qualified when it is detected that they conform to the preset relationship.

[0077] Among them, the preset relationships include a preset voltage-current relationship, a preset current-resistance relationship, and a preset voltage-resistance relationship.

[0078] Specifically, the preset relationships are determined according to the circuit connection set for the body solder joints. Detect respectively whether the combination of the transition current value and the transition voltage value conforms to the preset voltage-current relationship, whether the combination of the transition voltage value and the transition dynamic resistance value conforms to the preset voltage-resistance relationship, and whether the combination of the transition current value and the transition dynamic resistance value conforms to the preset current-resistance relationship.

[0079] Step S230: If the three combinations corresponding to the same acquisition time are all verified to be qualified, determine that the pairwise correlation verification is all qualified.

[0080] Specifically, as long as each combination conforms to the corresponding preset relationship, the combination is verified to be qualified. If the three combinations corresponding to the same acquisition time are all verified to be qualified, that is, the three combinations corresponding to the same acquisition time all conform to their respective corresponding preset relationships, then it can be determined that the pairwise correlation verification is all qualified.

[0081] Beneficial effects: Further refine the steps of pairwise correlation verification of the transition current value, transition voltage value, and transition dynamic resistance value at the same acquisition time. By combining the transition current value, transition voltage value, and transition dynamic resistance value at the same acquisition time in pairs, and respectively detecting whether the three obtained combinations conform to the preset relationships of the corresponding combinations, finally determine whether the corresponding combination is verified to be qualified according to the detection results, and further determine whether the pairwise correlation verification is qualified. Thereby ensuring that the logical relationship among the current, voltage, and dynamic resistance data at the same acquisition time is reasonable and consistent, realizing cross-verification of the specific parameter values in the transition welding parameter values, and improving the accuracy of the transition welding parameter values through the verification of the transition welding parameter values.

[0082] In some embodiments, as Figure 6 shown, step S140 further includes:

[0083] Step S141: Identify and extract the temporal features in the target welding signal using a preset neural network.

[0084] Specifically, the target welding signal includes the acquisition time and the corresponding target welding parameter values, and the temporal features are the target welding parameter values sorted according to the acquisition time. The preset neural network is a CNN or an LSTM. LSTM, Long Short-Term Memory, is a type of recurrent neural network for processing sequential data. CNN, Convolutional Neural Networks, is a class of feedforward neural networks with convolutional computations and a deep structure.

[0085] Step S142: Establish a reconstructed sample signal based on the temporal features.

[0086] Specifically, use the trained VAE model to reconstruct the target welding signal according to the temporal features to obtain the reconstructed sample signal. VAE is a generative model that generates similar samples by learning the distribution of normal data. The VAE model is trained using a training dataset containing normal data, and the training dataset includes the normal welding current, welding voltage, and dynamic resistance curve data of 300 vehicles.

[0087] Step S143: Calculate the difference value between the reconstructed sample signal and the target welding signal.

[0088] Specifically, calculate the error between the target welding signal and the reconstructed sample signal, that is, the difference value. This error reflects the reconstruction ability of the VAE model for the target welding signal and indirectly reflects the abnormality degree of the target welding signal. The difference value is the output of the preset prediction model.

[0089] Beneficial effects: Further refine the step of inputting the target welding signal into the preset prediction model and detecting whether the difference value output by the preset prediction model matches the preset threshold. Combining the advantages of VAE in the aspect of generative models and the ability of LSTM in feature extraction for deep learning and anomaly detection can capture the internal features and laws of the target welding signal, effectively distinguish normal and abnormal signals, and further explore the temporal features in the signal, making the anomaly detection more accurate. This not only reduces the false alarm rate but also improves the detection efficiency, ensuring the timely discovery and handling of welding quality problems.

[0090] In some embodiments, as Figure 7 shown, step S141 further includes:

[0091] Step S310: Arrange the target welding parameter values in the order of acquisition time to obtain a target parameter sequence.

[0092] Specifically, the target parameter sequence is the values of each target welding parameter in chronological order, which is convenient for subsequent data feature acquisition.

[0093] Step S320: Move and collect on the target parameter sequence using a window of a preset size to obtain multiple one-dimensional feature matrices.

[0094] Specifically, the window of the preset size is the preset convolution kernel, and the convolutional layer includes the convolution kernel. Each one-dimensional feature matrix is the value of the target welding parameter collected each time the window of the preset size moves.

[0095] Step S330: Perform convolution on the multiple one-dimensional feature matrices based on the preset convolutional layer to obtain temporal features.

[0096] Specifically, multiply the data in each one-dimensional feature matrix element by element with the corresponding convolution kernel, and add the obtained product results to obtain the temporal features.

[0097] Beneficial effects: The steps of how to obtain the temporal features are further refined. By arranging the target welding parameter values in the order of acquisition time to obtain the target parameter sequence, then using a window of a preset size for sliding window acquisition to obtain multiple one-dimensional feature matrices, and finally performing convolution on the multiple one-dimensional feature matrices based on the preset convolutional layer to obtain the temporal features, the extraction of the temporal features of the target welding signal is realized, enriching the technical solution of the present invention.

[0098] In some embodiments, the state includes a normal welding state and an abnormal welding state; as Figure 8 shown, step S150 further includes:

[0099] Step S151: If the matching result matches the preset threshold, it is determined that the target welding signal corresponding to the difference value is in the normal welding state.

[0100] Specifically, the matching result matches the preset threshold, that is, the difference value matches the preset threshold, that is, the corresponding target welding signal is in the normal welding state, that is, when the corresponding initial welding signal is collected, the body solder joints are welded normally.

[0101] Step S152: If the matching result does not match the preset threshold, it is determined that the target welding signal corresponding to the difference value is in the abnormal welding state.

[0102] Specifically, the matching result does not match the preset threshold, that is, the difference value does not match the preset threshold and exceeds the preset threshold range, that is, the corresponding target welding signal is in the abnormal welding state, that is, when the corresponding initial welding signal is collected, the body solder joints are welded abnormally.

[0103] Beneficial effect: The step of determining the state of the target welding signal corresponding to the difference value according to the matching result is further refined, enriching the technical solution. The state of the corresponding target welding signal is determined according to the matching result; and it is refined that when matching, the corresponding target welding signal is in a normal welding state, and when not matching, the corresponding target welding signal is in an abnormal welding state.

[0104] In some embodiments, Figure 9 As shown, after step S152, the following steps are also included:

[0105] Step S410: Acquire position information corresponding to the target welding signal.

[0106] The location information includes the collection station information of the target welding signal and the line information where the station is located.

[0107] Specifically, the target welding signal corresponds to the initial welding signal, and the position information corresponding to the target welding signal is the acquisition position of the corresponding initial welding signal, that is, the workstation where the corresponding body welding point is located and the workline where the workstation is located.

[0108] Step S420: Establish abnormal alarm information according to the position information and abnormal welding status.

[0109] Among them, the abnormal alarm information includes the line stop instruction.

[0110] Specifically, when the target welding signal is detected to be in an abnormal welding state, it is necessary to stop the line and control the line of the workstation where the body welding point corresponding to the initial welding signal is collected. Therefore, the abnormal alarm information includes the abnormal welding state and the workstation where the corresponding body welding point is located and the workline where the workstation is located.

[0111] Step S430: Send the abnormal alarm information to the Internet of Things, so that the Internet of Things issues a line stop instruction according to the abnormal alarm information, and sends the abnormal alarm information to the staff user terminal as a reminder.

[0112] Specifically, the abnormal alarm information and the workstations that need to be stopped are fed back to the Internet of Things. The Internet of Things notifies the production engineers of the alarm information and the stop information through the instant messaging system. At the same time, the Internet of Things system sends the alarm information to the PLC of the connected workstation corresponding to the abnormal welding point and issues a stop command. The PLC executes the stop command and links the stop workstation to sound and light alarm. After the relevant engineer handles and confirms, he presses the reset button and the line body resumes operation. In actual use, the abnormal alarm information also includes the workstation information that needs to be stopped, the body number BSN, the welding point number, the PLC information, the welding clamp information, etc. PLC, Programmable Logic Controller, programmable logic controller.

[0113] After detecting an abnormal welding state, the system feeds back the abnormal alarm information to the Internet of Things. The Internet of Things notifies the production-related engineers in the instant messaging system of the alarm information and stores the alarm information at the same time. Before the vehicle body arrives at the inspection station, the PLC sends a quality status confirmation request. The Internet of Things system sends the determination information of the BSN solder joint quality of the corresponding vehicle body to the PLC control line of the inspection station and prompts the information that needs to be repaired. After the repair is completed, press the repair confirmation button to release.

[0114] When an abnormal welding state is detected, it can immediately trigger a control line stop instruction and notify the production engineer through the Internet of Things platform in real time. At the same time, it can also send a quality status confirmation request in advance before the vehicle body arrives at the inspection station to ensure that problems are handled in a timely manner. After the repair is completed, it is released by pressing the repair confirmation button, realizing the closed-loop management of the production line.

[0115] Beneficial effects: After determining that it is in an abnormal welding state, an abnormal alarm information is generated and sent to the Internet of Things, so that the Internet of Things issues a stop instruction according to the abnormal alarm information and sends the abnormal alarm information to the staff user terminal for reminder. Thus, it is ensured that abnormal problems can be discovered and processed in the first time, effectively preventing the continuous production of unqualified products and improving product quality.

[0116] Figure 10 The structural schematic diagram of the embodiment of the detection device for the welding signal of the vehicle body solder joints of the present application is shown. Please refer to Figure 10 As shown, the device 500 includes: a signal acquisition module 510, a preprocessing module 520, a cross-validation module 530, a model prediction module 540, and a state determination module 550.

[0117] The signal acquisition module 510 is used to acquire the initial welding signal generated during the welding of the vehicle body solder joints; wherein, the initial welding signal includes the acquisition time and the initial welding parameter value corresponding to the acquisition time;

[0118] The preprocessing module 520 is used to preprocess the initial welding parameter value to obtain the preprocessed transitional welding parameter value;

[0119] The cross-validation module 530 is used to perform cross-validation on the specific parameter values in the transitional welding parameter value based on the acquisition time, obtain the qualified transitional welding parameter value after validation, and use the transitional welding parameter value as the target welding parameter value;

[0120] The model prediction module 540 is used to input the target welding signal into a preset prediction model and detect whether the difference value output by the preset prediction model matches a preset threshold; wherein, the target welding signal includes the acquisition time and the target welding parameter value corresponding to the acquisition time;

[0121] A status determination module 550, configured to determine the status of the target welding signal corresponding to the difference value according to the matching result.

[0122] Advantageous effects: In this embodiment, by preprocessing the initial welding parameter values included in the collected initial welding signal, transitional welding parameter values are obtained; based on the acquisition time, cross-verification is performed on the specific parameter values in the transitional welding parameter values, and the verified transitional welding parameter values are used as target welding parameter values; the target welding signal composed of the target welding parameter values and the corresponding acquisition time is input into a preset prediction model, and it is detected whether the difference value output by the preset prediction model matches a preset threshold; finally, according to the matching result, the status of the target welding signal is determined. Thus, by means of preprocessing and cross-verifying the parameter values, the accuracy of the parameter values is improved, and further, the accuracy of the target welding signal obtained based on the parameter values is ensured, and finally, the accuracy of the difference value and the final status output based on the target welding signal is ensured.

[0123] By processing all target welding signals with a preset prediction model and matching the output difference values with a preset threshold, the consistency of the status finally determined according to the matching result is ensured.

[0124] Moreover, by implementing this embodiment, the welding signals generated during the welding process of the body weld points can be automatically detected without manual visual inspection, greatly improving the efficiency of detecting the weld signals of the weld points.

[0125] It should be noted that the detection device for the body weld point welding signal provided in the above embodiment and the detection method for the body weld point welding signal provided in the foregoing embodiment belong to the same concept. The specific manners in which each module and unit perform operations have been described in detail in the method embodiment and will not be elaborated here.

[0126] Figure 11 The structural schematic diagram of the embodiment of the device of the present application is shown, and it shows the structural schematic diagram of the computer system of the device suitable for implementing the embodiment of the present application. The specific implementation of the device in the specific embodiment of the present application is not limited.

[0127] Please refer to Figure 11 As shown, the device includes: a controller; a memory, configured to store one or more programs, and when the one or more programs are executed by the controller, to execute the above-mentioned detection method for the body weld point welding signal.

[0128] Please continue to refer to Figure 11As shown, the computer system 600 of the device includes a Central Processing Unit (CPU) 601, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 602 or the program loaded from the storage section 608 into the Random Access Memory (RAM) 603, such as executing the methods in the above embodiments. In the RAM 603, various programs and data required for system operation are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An Input / Output (I / O) interface 605 is also connected to the bus 604.

[0129] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including, for example, a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), etc. and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read from it can be installed into the storage section 608 as needed.

[0130] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 609, and / or installed from the removable medium 611. When the computer program is executed by the Central Processing Unit (CPU) 601, various functions defined in the system of the present application are executed.

[0131] Another aspect of the present application also provides a computer-readable storage medium, in which at least one executable instruction is stored, and when the executable instruction runs on a device, it causes the device to execute the operations of the method for detecting the body solder joint welding signal in any one of the above embodiments.

[0132] Beneficial effects: In this embodiment, the initial welding parameter values included in the collected initial welding signal are preprocessed to obtain transition welding parameter values; based on the acquisition time, cross-validation is performed on the specific parameter values in the transition welding parameter values, and the verified transition welding parameter values are used as target welding parameter values; the target welding signal composed of the target welding parameter values and the corresponding acquisition time is input into a preset prediction model, and it is detected whether the difference value output by the preset prediction model matches a preset threshold; finally, according to the matching result, the state of the target welding signal is determined. Thus, by preprocessing and cross-validating the parameter values, the accuracy of the parameter values is improved, thereby ensuring the accuracy of the target welding signal obtained based on the parameter values, and finally ensuring the accuracy of the difference value and the final state output based on the target welding signal.

[0133] By processing all target welding signals with a preset prediction model and matching the output difference values with a preset threshold, the consistency of the finally determined state based on the matching result is ensured.

[0134] Moreover, by implementing this embodiment, the welding signals generated during the welding process of the body solder joints can be automatically detected without manual visual inspection, greatly improving the efficiency of solder joint welding signal detection.

[0135] It should be noted that the computer-readable medium shown in the embodiments of the present application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable computer program is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0136] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. Among them, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in an order different from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, as well as the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0137] The units involved in the embodiments described in this application can be implemented in software or in hardware, and the described units can also be provided in a processor. Among them, the names of these units do not constitute a limitation to the units themselves in some cases.

[0138] According to one aspect of the embodiments of the present application, a computer system is further provided, including a central processing unit (CPU), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage section into a random access memory (RAM), such as performing the methods in the above embodiments. In the RAM, various programs and data required for system operations are also stored. The CPU, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0139] The following components are connected to the I / O interface: an input section including a keyboard, a mouse, etc.; an output section including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section including a hard disk, etc.; and a communication section including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as required. A removable medium, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive as required so that a computer program read from it can be installed into the storage section as required.

[0140] The above content is only a preferred exemplary embodiment of the present application and is not used to limit the implementation of the present application. Those of ordinary skill in the art can easily make corresponding adaptations or modifications according to the main concept and spirit of the present application. Therefore, the protection scope of the present application should be subject to the protection scope required by the claims.

Claims

1. A method for detecting welding signals of a vehicle body welding point, characterized in that: The method comprises: Collecting an initial welding signal generated by a welding point of a vehicle body during a welding process; wherein the initial welding signal includes a collection time and an initial welding parameter value corresponding to the collection time; Preprocessing the initial welding parameter value to obtain a preprocessed transition welding parameter value; Based on the acquisition time, cross-validate the specific parameter values ​​in the transition welding parameter values ​​to obtain qualified transition welding parameter values, and use the transition welding parameter values ​​as target welding parameter values; Inputting a target welding signal into a preset prediction model, and detecting whether a difference value output by the preset prediction model matches a preset threshold; wherein the target welding signal includes the acquisition time and a target welding parameter value corresponding to the acquisition time; According to the matching result, the state of the target welding signal corresponding to the difference value is determined.

2. The method according to claim 1, characterized in that The initial welding parameter values ​​include an initial current value, an initial voltage value and an initial dynamic resistance value; The step of preprocessing the initial welding parameter value to obtain the preprocessed transition welding parameter value further includes: Performing filtering on the initial current value to obtain a transition current value; Performing voltage calibration processing on the initial voltage value to obtain a transition voltage value; The initial dynamic resistance value is analyzed based on a preset analysis method to perform denoising on the initial dynamic resistance value to obtain a transition dynamic resistance value.

3. The method according to claim 1, characterized in that The specific parameter values ​​in the transition welding parameter values ​​include transition current value, transition voltage value and transition dynamic resistance value; The step of cross-validating specific parameter values ​​in the transition welding parameter values ​​based on the acquisition time to obtain qualified transition welding parameter values ​​further includes: Performing pairwise correlation verification on the transition current value, the transition voltage value and the transition dynamic resistance value at the same acquisition time; If both the pairwise correlation checks are qualified, it is determined that the transition current value, the transition voltage value, and the transition dynamic resistance value corresponding to the acquisition time are all qualified.

4. The method according to claim 3, characterized in that The step of performing pairwise correlation verification on the transition current value, the transition voltage value and the transition dynamic resistance value at the same acquisition time further comprises: Combining the transition current value, the transition voltage value, and the transition dynamic resistance value at the same acquisition time in pairs to obtain three combinations including two values; Detecting whether two values ​​in each combination meet the preset relationship of the corresponding combination, and determining that the verification is qualified when it is detected that the preset relationship is met; wherein the preset relationship includes a preset voltage-current relationship, a preset current-resistance relationship, and a preset voltage-resistance relationship; If the three combinations corresponding to the same collection time are all verified to be qualified, it is determined that the pairwise correlation verifications are all qualified.

5. The method according to claim 1, characterized in that The step of inputting the target welding signal into a preset prediction model and detecting whether the difference value output by the preset prediction model matches a preset threshold value further includes: Using a preset neural network to identify and extract the timing characteristics in the target welding signal; Establishing a reconstructed sample signal according to the timing characteristics; The difference value between the reconstructed sample signal and the target welding signal is calculated.

6. The method according to claim 1, characterized in that The states include a normal welding state and an abnormal welding state; The step of determining the state of the target welding signal corresponding to the difference value according to the matching result further includes: If the matching result matches the preset threshold, it is determined that the target welding signal corresponding to the difference value is in a normal welding state; If the matching result does not match the preset threshold, it is determined that the target welding signal corresponding to the difference value is in an abnormal welding state.

7. The method according to claim 6, characterized in that After the step of determining that the target welding signal corresponding to the difference value is in an abnormal welding state if the matching result does not match the preset threshold, the method further includes: Acquire the position information corresponding to the target welding signal; wherein the position information includes the acquisition station information of the target welding signal and the line information where the station is located; Establishing abnormal alarm information according to the position information and the abnormal welding state; wherein the abnormal alarm information includes a line stop instruction; The abnormal alarm information is sent to the Internet of Things, so that the Internet of Things issues the line stop instruction according to the abnormal alarm information, and sends the abnormal alarm information to the staff user terminal as a reminder.

8. A device for detecting welding signals of a vehicle body welding spot, characterized in that: The device comprises: A signal acquisition module, used to acquire an initial welding signal generated by a welding point of a vehicle body during a welding process; wherein the initial welding signal includes an acquisition time and an initial welding parameter value corresponding to the acquisition time; A preprocessing module, used for preprocessing the initial welding parameter value to obtain a preprocessed transition welding parameter value; A cross-validation module, used to cross-validate specific parameter values ​​in the transition welding parameter values ​​based on the acquisition time, obtain qualified transition welding parameter values, and use the transition welding parameter values ​​as target welding parameter values; A model prediction module, used for inputting a target welding signal into a preset prediction model, and detecting whether a difference value output by the preset prediction model matches a preset threshold; wherein the target welding signal includes the acquisition time and a target welding parameter value corresponding to the acquisition time; The state determination module is used to determine the state of the target welding signal corresponding to the difference value according to the matching result.

9. A device, characterized in that: include: Controller; The memory is used to store one or more programs. When the one or more programs are executed by the controller, the controller implements the method for detecting welding signals of vehicle body welding points as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one executable instruction. When the executable instruction is executed on the device / equipment, the device / equipment executes the operation of the method for detecting welding signals of vehicle body welding points as described in any one of claims 1 to 7.

Images