Automobile battery welding monitoring method and device, computer equipment and storage medium

Abstract

The embodiment of the application belongs to the field of welding monitoring, and relates to a kind of automobile battery welding monitoring methods: when welding automobile battery, the optical monitoring data, temperature monitoring data and plasma monitoring data of welding area are acquired, and time sequence feature extraction and spatial feature extraction are carried out respectively, and first optical data, first temperature data and first plasma data are generated according to the extracted features;First optical data, first temperature data and first plasma data are interactively fused to obtain second optical data, second temperature data and second plasma data;Obtain defect type identification, and add weight to second optical data, second temperature data and second plasma data according to defect type identification;Second optical data, second temperature data and second plasma data with weight are input into the defect detection network corresponding to the defect type identification to obtain the welding defect detection result.The application improves the accuracy of automobile battery welding monitoring.

Description

Technical Field

[0001] This application relates to the field of welding monitoring technology, and in particular to a method, device, computer equipment and storage medium for monitoring welding of automotive batteries. Background Technology

[0002] With the rapid development of new energy vehicles, the production of automotive power batteries has become a critical link. The quality of battery welding directly affects the safety, performance, and lifespan of the battery, making the monitoring of the welding process essential. Currently, automotive power battery welding monitoring technology processes data from various sensors independently, ignoring the correlation between different sensors and failing to effectively integrate the data. This affects the accurate identification of complex welding defects (such as porosity, cracks, and incomplete welds), resulting in a high false detection rate. Summary of the Invention

[0003] The purpose of this application is to provide a method, apparatus, computer equipment, and storage medium for monitoring automotive battery welding, in order to solve the problem of low accuracy in monitoring automotive battery welding.

[0004] To address the aforementioned technical problems, this application provides a method for monitoring automotive battery welding, employing the following technical solution:

[0005] When welding automotive batteries, acquire optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area;

[0006] Temporal and spatial features are extracted from the optical monitoring data, temperature monitoring data, and plasma monitoring data, respectively, and first optical data, first temperature data, and first plasma data are generated based on the extracted features.

[0007] The first optical data, the first temperature data, and the first plasma data are interactively fused to obtain second optical data, second temperature data, and second plasma data.

[0008] Obtain the defect type identifier, and add weights to the second optical data, the second temperature data, and the second plasma data according to the defect type identifier;

[0009] The weighted second optical data, second temperature data, and second plasma data are input into a defect detection network corresponding to the defect type identifier to obtain welding defect detection results.

[0010] Furthermore, the steps of acquiring optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area include:

[0011] Optical monitoring data is obtained by acquiring the morphology of the molten pool and weld seam in the welding area using optical sensors.

[0012] Temperature monitoring data is obtained by collecting the temperature field distribution of the molten pool in the welding area using temperature sensors.

[0013] Plasma monitoring data is obtained by collecting the radiation characteristics of the plasma in the welding area using a plasma sensor.

[0014] Furthermore, the step of performing temporal feature extraction and spatial feature extraction on the optical monitoring data, the temperature monitoring data, and the plasma monitoring data, respectively, and generating first optical data, first temperature data, and first plasma data based on the extracted features includes:

[0015] The change information of the image sequence in the optical monitoring data is extracted as optical temporal features, the spatial image features of the image sequence in the optical monitoring data are extracted, and the first optical data is obtained based on the optical temporal features and the spatial image features.

[0016] The temperature change information and temperature gradient of the temperature monitoring data are extracted as temperature time series features. The temperature field distribution and heat-affected zone are extracted from the temperature monitoring data as temperature spatial features. The first temperature data is obtained based on the temperature time series features and the temperature spatial features.

[0017] The radiation intensity change information and plasma fluctuation information of the plasma monitoring data are extracted as plasma temporal features, and the plasma radiation distribution and plasma spectral features of the plasma monitoring data are extracted as plasma spatial features. The first plasma data is obtained based on the plasma temporal features and the plasma spatial features.

[0018] Furthermore, the step of interactively fusing the first optical data, the first temperature data, and the first plasma data to obtain the second optical data, the second temperature data, and the second plasma data includes:

[0019] The first temperature data and the first plasma data are fused into the first optical data through a cross-modal attention mechanism to obtain the second optical data;

[0020] The first optical data and the first plasma data are fused into the first temperature data through the cross-modal attention mechanism to obtain the second temperature data;

[0021] The first optical data and the first temperature data are fused into the first plasma data through the cross-modal attention mechanism to obtain the second plasma data.

[0022] Furthermore, the step of fusing the first temperature data and the first plasma data into the first optical data through a cross-modal attention mechanism to obtain the second optical data includes:

[0023] The first optical data and the first temperature data are fused using a cross-modal attention mechanism to obtain the first optical fused data;

[0024] The first optical data and the first plasma data are fused using the cross-modal attention mechanism to obtain the second optical fused data;

[0025] The first optical fusion data and the second optical fusion data are concatenated to obtain the second optical data.

[0026] Furthermore, the step of fusing the first optical data and the first temperature data through a cross-modal attention mechanism to obtain the first optical fused data includes:

[0027] The first optical data is linearly transformed to obtain the optical query vector;

[0028] Calculate the temperature feature key and temperature feature value of the first temperature data;

[0029] Calculate the attention score between the optical query vector and the temperature feature key;

[0030] The temperature feature values ​​are weighted according to the attention scores to obtain weighted temperature data;

[0031] The weighted temperature data and the first optical data are added together to obtain the first optical fusion data.

[0032] Furthermore, the step of adding weights to the second optical data, the second temperature data, and the second plasma data based on the defect type identifier includes:

[0033] Obtain a weight combination corresponding to the defect type identifier, the weight combination being learned in advance;

[0034] Weights are added to the second optical data, the second temperature data, and the second plasma data according to the weight combination; or...

[0035] Based on the defect type identifier, the weights of the second optical data, the second temperature data, and the second plasma data are calculated using a self-attention mechanism.

[0036] To address the aforementioned technical problems, this application also provides an automotive battery welding monitoring device, which employs the following technical solution:

[0037] The monitoring and acquisition module is used to acquire optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area when welding automotive batteries.

[0038] The feature extraction module is used to extract temporal features and spatial features from the optical monitoring data, the temperature monitoring data and the plasma monitoring data respectively, and generate first optical data, first temperature data and first plasma data based on the extracted features;

[0039] An interactive fusion module is used to interactively fuse the first optical data, the first temperature data, and the first plasma data to obtain second optical data, second temperature data, and second plasma data.

[0040] The weighting module is used to obtain the defect type identifier and add weights to the second optical data, the second temperature data and the second plasma data according to the defect type identifier;

[0041] The defect detection module is used to input the weighted second optical data, second temperature data, and second plasma data into a defect detection network corresponding to the defect type identifier to obtain welding defect detection results.

[0042] To address the aforementioned technical problems, this application also provides a computer device, which includes a memory and a processor. The memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the automotive battery welding monitoring method described above.

[0043] To address the aforementioned technical problems, this application also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by a processor, implement the steps of the automotive battery welding monitoring method described above.

[0044] Compared with existing technologies, the embodiments of this application have the following main advantages: The extraction of temporal and spatial features from optical monitoring data, temperature monitoring data, and plasma monitoring data effectively captures key changes during the welding process, avoiding the problem that single sensor data cannot fully reflect welding defects; the use of a cross-modal attention mechanism to interactively fuse the first optical data, first temperature data, and first plasma data enhances the correlation between different sensor data and improves the accuracy of defect identification; finally, the dynamic adjustment of the weights of the second optical data, second temperature data, and second plasma data based on the defect type identifier makes the contribution of each sensor data to the defect detection result more precise, further reducing the false detection rate; and the selection of the corresponding defect detection network based on the defect type identifier for welding defect detection enables the accurate identification of complex defects during the welding process, improving the accuracy of automotive battery welding monitoring. Attached Figure Description

[0045] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0047] Figure 2 This is a flowchart of one embodiment of the automotive battery welding monitoring method according to this application;

[0048] Figure 3 This is a schematic diagram of a structure of an embodiment of the automotive battery welding monitoring device according to this application;

[0049] Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0051] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0052] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0053] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0054] During the welding of automotive power batteries, terminal devices 101, 102, and 103 can collect various data from the welding area, such as optical monitoring data, temperature monitoring data, and plasma monitoring data, and interact with server 105 via network 104 to receive or send messages. Terminal devices 101, 102, and 103 can be, but are not limited to, various industrial computers, personal computers, laptops, and sensors. Server 105 can be equipped with an automotive battery welding monitoring system.

[0055] It should be noted that the automotive battery welding monitoring method provided in this application embodiment is generally executed by a server, and correspondingly, the automotive battery welding monitoring device is generally installed in the server.

[0056] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0057] Continue to refer to Figure 2 A flowchart of one embodiment of the automotive battery welding monitoring method according to this application is shown. The automotive battery welding monitoring method includes the following steps:

[0058] Step S201: When welding the car battery, acquire optical monitoring data, temperature monitoring data and plasma monitoring data of the welding area.

[0059] In this embodiment, the automotive battery welding monitoring method operates on electronic devices (e.g., Figure 1The server shown can communicate with the terminal device via wired or wireless connection. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other currently known or future known wireless connection methods.

[0060] Specifically, during the welding of automotive batteries (which can be automotive power batteries), the automotive battery welding monitoring system collects monitoring data of the welding area, obtaining optical monitoring data, temperature monitoring data, and plasma monitoring data. The optical monitoring data includes visual information about the molten pool and weld seam; the temperature monitoring data includes temperature distribution information; and the plasma monitoring data includes the radiation characteristics of the plasma.

[0061] Furthermore, the steps of acquiring optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area may include: acquiring the molten pool morphology and weld morphology of the welding area through an optical sensor to obtain optical monitoring data; acquiring the molten pool temperature field distribution of the welding area through a temperature sensor to obtain temperature monitoring data; and acquiring the radiation characteristics of the plasma in the welding area through a plasma sensor to obtain plasma monitoring data.

[0062] Specifically, the automotive battery welding monitoring system can collect monitoring data in three dimensions: optics, temperature, and plasma. The system can monitor the molten pool and weld morphology in the welding area in real time using optical sensors, obtaining optical monitoring data to provide morphological information for subsequent defect detection. The size, shape, and changes of the molten pool during welding are important evaluation criteria for welding quality. For example, an excessively large or irregular molten pool may lead to welding defects such as overheating or incomplete welding. The width, depth, and morphology of the weld directly affect the welding quality.

[0063] Temperature sensors can monitor the temperature field distribution of the molten pool in real time during welding, thus obtaining temperature monitoring data. The temperature field distribution reflects the heat transfer and distribution during welding and is a crucial factor in judging weld quality. During welding, the temperature of the molten pool region is a key quality indicator; excessively high temperatures may lead to overheating or material loss, while excessively low temperatures may result in insufficient penetration. Changes in the temperature field determine the size of the heat-affected zone (HAZ). If the temperature in this zone exceeds the annealing temperature of the base material, it may cause changes in the material's physical properties, affecting weld quality.

[0064] Plasma sensors (which can be spectrometers) can monitor the radiation characteristics of plasma during the welding process. By measuring the radiation intensity and wavelength, they provide information about the welding process, thus obtaining plasma monitoring data. The radiation characteristics of the plasma can reveal the energy distribution, metal evaporation, and possible formation of porosity or cracks during the welding process. Changes in plasma intensity and spectrum are often closely related to welding defects such as porosity and cracks.

[0065] In this embodiment, optical monitoring data reflects the welding morphology, temperature monitoring data reflects the thermal state, and plasma monitoring data provides clues about potential defects such as metal evaporation and porosity. Acquiring multi-dimensional monitoring data is beneficial for more comprehensive and accurate monitoring of welding quality.

[0066] Step S202: Perform temporal feature extraction and spatial feature extraction on the optical monitoring data, temperature monitoring data and plasma monitoring data respectively, and generate first optical data, first temperature data and first plasma data based on the extracted features.

[0067] Specifically, optical monitoring data, temperature monitoring data, and plasma monitoring data are input into a spatiotemporal feature fusion network to extract temporal and spatial features from the monitoring data. The monitoring data contains temporal information and can form a time series. The spatiotemporal feature fusion network can use a bidirectional LSTM (Long Short-Term Memory) network to process the time series data and extract the dynamic changes during the welding process. The spatiotemporal feature fusion network can also use a 3D convolutional neural network (3D-CNN) to extract the spatial features of the monitoring data and process the spatial distribution information of the welding area, such as changes in the molten pool morphology.

[0068] Ultimately, the first optical data, first temperature data, and first plasma data are obtained. Among them, the first optical data includes temporal and spatial features extracted from optical monitoring data, the first temperature data includes temporal and spatial features extracted from temperature monitoring data, and the first plasma data includes temporal and spatial features extracted from plasma monitoring data.

[0069] Furthermore, step S202 may include: extracting change information of image sequences from optical monitoring data as optical temporal features, extracting spatial image features of image sequences from optical monitoring data, and obtaining first optical data based on optical temporal features and spatial image features; extracting temperature change information and temperature gradient from temperature monitoring data as temperature temporal features, extracting temperature field distribution and heat-affected zone from temperature monitoring data as temperature spatial features, and obtaining first temperature data based on temperature temporal features and temperature spatial features; extracting radiation intensity change information and plasma fluctuation information from plasma monitoring data as plasma temporal features, extracting plasma radiation distribution and plasma spectral features from plasma monitoring data as plasma spatial features, and obtaining first plasma data based on plasma temporal features and plasma spatial features.

[0070] Specifically, optical monitoring data consists of image sequences with temporal information. The changes in these image sequences are extracted as optical temporal features. These features reflect the temporal information of image changes within the sequence, specifically the evolution of the molten pool and weld morphology. Changes in the image sequence reflect the expansion of the molten pool or the offset of the weld, and these changes can be used to identify potential problems in the welding process (such as porosity or cracks). Optical temporal features can include dynamic changes, color variations, or brightness changes in the images.

[0071] Spatial image features are extracted from image sequences in optical monitoring data as optical spatial features. Optical spatial features focus on spatial information of the image, such as the size and shape of the molten pool, the width of the weld, and the morphology, texture, and brightness distribution of the welded area. Through the extraction of optical spatial features, the geometric and morphological information of the welded area can be used to analyze the stability and quality of the weld. The optical temporal features are combined with spatial image features to generate the first optical data.

[0072] Temperature change information and temperature gradient are extracted from temperature monitoring data as temperature time-series features. Temperature change information can reflect the stability of the heat source and the temperature fluctuation of the molten pool during welding. For example, a sudden increase or decrease in temperature may correspond to anomalies in the welding process (such as overheating or insufficient penetration). The temperature gradient refers to the temperature difference at different time steps, reflecting the rate of temperature change. It can detect the non-uniformity of heat distribution in the welding area, thereby revealing potential defects in the welding process (such as porosity, cold welds, etc.).

[0073] Temperature field distribution and heat-affected zone (HAZ) are extracted from temperature monitoring data as spatial temperature features. These spatial features focus on the temperature distribution within the welding area, including the temperature field distribution (temperature distribution map of the welding area) and the size and shape of the HAZ. The high temperature in the HAZ can affect the microstructure and properties of the material, thus impacting weld quality. Combining temporal and spatial temperature features generates initial temperature data, which is used to analyze the thermal state during the welding process.

[0074] Radiation intensity variation and plasma fluctuation information from plasma monitoring data are extracted as plasma temporal features. Radiation intensity variation information reflects changes in plasma radiation intensity, indicating temperature fluctuations and metal flow during the welding process, and can be used to identify defects such as porosity formation and crack propagation. Plasma fluctuation information reflects unstable aspects of the welding process, especially when defects such as porosity and cracks occur, where plasma intensity and wavelength fluctuate.

[0075] Plasma radiation distribution and plasma spectral characteristics are extracted from plasma monitoring data as plasma spatial features. Plasma radiation distribution, the spatial distribution of plasma, reflects the thermal state and metal flow in different regions during the welding process, helping to detect irregular welding behavior and thus determine the presence of welding defects. Plasma spectral characteristics, measured at different wavelengths, can determine whether there is undesirable metal evaporation or gas dissolution during the welding process, thereby inferring possible defect types. Combining plasma temporal and spatial features generates the first plasma data.

[0076] In this embodiment, temporal and spatial features are extracted from optical monitoring data, temperature monitoring data, and plasma monitoring data to comprehensively capture key changes in the welding process. Temporal features help capture the dynamic changes in the welding process, while spatial features provide geometric information about the welding process. This multi-dimensional feature extraction method provides more accurate input data for defect detection and improves the accuracy of automotive battery welding monitoring.

[0077] Step S203: The first optical data, the first temperature data, and the first plasma data are interactively fused to obtain the second optical data, the second temperature data, and the second plasma data.

[0078] Specifically, the system performs cross-modal fusion of the first optical data, the first temperature data, and the first plasma data, that is, it interactively fuses the temporal and spatial features extracted by different sensors to enhance the correlation between the data.

[0079] Furthermore, step S203 may include: fusing the first temperature data and the first plasma data into the first optical data through a cross-modal attention mechanism to obtain the second optical data; fusing the first optical data and the first plasma data into the first temperature data through a cross-modal attention mechanism to obtain the second temperature data; and fusing the first optical data and the first temperature data into the first plasma data through a cross-modal attention mechanism to obtain the second plasma data.

[0080] Specifically, the cross-modal attention mechanism is a weighted mechanism that calculates the similarity between different data sources. This mechanism enables a particular modality (such as optical data) to actively focus on important information from other modalities (such as temperature data and plasma data), thereby achieving the fusion of intermodal information. In this application, the cross-modal attention mechanism weights data from different sensors, thereby effectively fusing the features of optical, temperature, and plasma data.

[0081] First, the first temperature data and the first plasma data are fused into the first optical data through a cross-modal attention mechanism: the first temperature data and the first plasma data are used as query information, and the influence of these data on the first optical data is calculated through the cross-modal attention mechanism. After dynamic weighting, the fused optical data (i.e., the second optical data) is obtained.

[0082] Similarly, the first optical data and the first plasma data are fused into the first temperature data through a cross-modal attention mechanism. The first optical data and the first plasma data affect the weighting of the first temperature data, resulting in the fused temperature data (i.e., the second temperature data).

[0083] Finally, the first optical data and the first temperature data are fused into the first plasma data through a cross-modal attention mechanism, incorporating the influence of the first optical data and the first temperature data into the first plasma data to generate enhanced plasma data (i.e., second plasma data).

[0084] In this embodiment, by using a cross-modal attention mechanism to fuse data from different sensors, the intrinsic relationships between the data are automatically captured. The second optical data, second temperature data, and second plasma data generated after fusion can provide more comprehensive and higher quality feature information, better describe the changes and defects in the welding process, and improve the accuracy of welding defect detection.

[0085] Furthermore, the step of fusing the first temperature data and the first plasma data into the first optical data through the cross-modal attention mechanism to obtain the second optical data may include: fusing the first optical data and the first temperature data through the cross-modal attention mechanism to obtain the first optical fused data; fusing the first optical data and the first plasma data through the cross-modal attention mechanism to obtain the second optical fused data; and concatenating the first optical fused data and the second optical fused data to obtain the second optical data.

[0086] Specifically, by using a cross-modal attention mechanism, the first optical data and the first temperature data are fused to generate the first optical fused data. Combining the optical data (involving molten pool morphology and weld morphology) with the temperature data (involving molten pool temperature changes and heat-affected zone) can enhance the expression of the influence of temperature on the weld morphology and help detect defects caused by temperature fluctuations (such as insufficient penetration or overheating).

[0087] Similarly, the first optical data and the first plasma data are fused to generate second optical fused data. By calculating the similarity between the optical data and the plasma data, the plasma data (involving plasma radiation intensity and fluctuation information) enhances the defect identification capability of the optical data, providing more important information, especially when detecting defects related to plasma instability (such as porosity).

[0088] The first optical fusion data is concatenated with the second optical fusion data (concat()) to form the second optical data. The concatenation operation integrates the features obtained from the two cross-modal fusions, forming a rich feature vector containing optical, temperature, and plasma information, which can provide a more comprehensive input for subsequent defect detection networks.

[0089] In this embodiment, the first and second optical fusion data generated by weighted fusion through cross-modal attention mechanism not only express optical data but also incorporate key information from temperature and plasma data. Multi-dimensional feature fusion can provide richer and more comprehensive welding process information, which helps to accurately identify welding defects.

[0090] Furthermore, the steps described above for fusing the first optical data and the first temperature data through a cross-modal attention mechanism to obtain the first optical fused data may include: performing a linear transformation on the first optical data to obtain an optical query vector; calculating the temperature feature key and temperature feature value of the first temperature data; calculating the attention score between the optical query vector and the temperature feature key; weighting the temperature feature value according to the attention score to obtain weighted temperature data; and adding the weighted temperature data and the first optical data to obtain the first optical fused data.

[0091] Specifically, the first optical data undergoes a linear transformation to extract the most relevant features for fusion with other modalities (such as temperature data), resulting in an optical query vector. The linear transformation can be obtained by multiplying the first optical data with a trainable weight matrix. This linear transformation maps the first optical data from a high-dimensional space to a space more suitable for interaction with other modal data.

[0092] The first temperature data is projected by multiplying it with the key projection matrix to obtain the temperature feature key, and then multiplying it with the value projection matrix to obtain the temperature feature value. The temperature feature key contains key information about the first temperature data and can be matched with the optical query vector; while the temperature feature value is the specific numerical value of the temperature data, which is used for final fusion after weighting.

[0093] The similarity between the optical query vector and the temperature feature key is calculated to obtain the attention score, which represents the correlation between the first temperature data and the first optical data. The first temperature data with a high attention score will occupy a more important position in the fusion.

[0094] Based on the calculated attention scores, the temperature feature values ​​are weighted to obtain weighted temperature data. The weighted temperature data can be adjusted according to the query vector of the optical data, so that the most relevant parts of the temperature data contribute the most to the final optical data fusion.

[0095] The weighted temperature data is added to the first optical data to obtain the first optical fusion data. This data contains comprehensive information from both optical and temperature data, enabling a more accurate reflection of the multimodal characteristics of the welding process and providing rich data input for subsequent defect detection.

[0096] In this embodiment, the first optical data and the first temperature data are organically fused through a cross-modal attention mechanism, ensuring that the data from each sensor plays a better role in defect detection. The attention calculation between the optical query vector and the temperature feature key enables the system to dynamically adjust the weights according to the correlation of different data, enhance key information, and ensure that the most relevant part of the first temperature data has the greatest impact on the fusion result of the first optical data. The dynamic weighting mechanism improves the accuracy of the fused data and effectively enhances the accuracy of welding defect detection.

[0097] It is understood that, referring to the fusion process of the first optical data and the first temperature data, the first optical data and the first plasma data are fused in the same way to obtain the second optical fused data. The first optical fused data and the second optical fused data are then concatenated to obtain the second optical data. This application will not elaborate further.

[0098] Referring to the process of fusing first temperature data and first plasma data into first optical data to obtain second optical data, the first optical data and first plasma data are fused into first temperature data in the same manner to obtain second temperature data; and the first optical data and first temperature data are fused into first plasma data in the same manner to obtain second plasma data. This application will not elaborate further.

[0099] Step S204: Obtain the defect type identifier and add weights to the second optical data, second temperature data and second plasma data according to the defect type identifier.

[0100] Specifically, each defect type has a corresponding defect type identifier, and the system can detect multiple defect types. Based on the defect type identifier (e.g., porosity, cracks, poor welds), the second optical data, temperature data, and plasma data are weighted and dynamically adjusted according to the importance of the sensors to better adapt to the characteristics of different defect types. For example, for porosity defects, temperature data may be more important, so it is given a larger weight; for crack defects, optical data may be more critical, so its weight is increased.

[0101] Furthermore, the steps of adding weights to the second optical data, the second temperature data, and the second plasma data based on the defect type identifier may include: obtaining a weight combination corresponding to the defect type identifier, wherein the weight combination is obtained in advance through learning; adding weights to the second optical data, the second temperature data, and the second plasma data based on the weight combination; or, calculating the weights of the second optical data, the second temperature data, and the second plasma data based on the defect type identifier using a self-attention mechanism.

[0102] Specifically, weight combinations corresponding to different defect types can be obtained in advance through training. These weight combinations are learned during the training phase using machine learning algorithms (such as deep neural networks), based on the different dependencies of each defect type (such as porosity, cracks, etc.) on optical data, temperature data, and plasma data in the training data. Each defect type will have a specific set of weights, representing the importance of the three data sources under that defect type.

[0103] Weight combinations are obtained based on the defect type identifier, and the second optical data, second temperature data, and second plasma data are weighted accordingly. Different defect types correspond to different weight combinations, giving higher weights to relevant data (such as temperature data, which is more important in insufficient penetration defects), thus making the system pay more attention to the contribution of relevant data when detecting that defect type.

[0104] Alternatively, a self-attention mechanism can calculate the weights of the second optical data, second temperature data, and second plasma data based on the defect type identifier. By calculating the similarity and correlation between the data sources, the weight of each data source is automatically adjusted, improving the detection accuracy for specific defect types. The self-attention mechanism can flexibly handle the feature differences of different types of defects, adjust the weights in real time, and improve the adaptability and accuracy of detection.

[0105] In this embodiment, the weight combination corresponding to the defect type identifier is obtained, and an appropriate weight is automatically assigned to each defect type, so that when the system detects a specific defect, it focuses on the most relevant data source to improve detection accuracy; or, through a self-attention mechanism, the weights of different data sources are dynamically adjusted according to the characteristics of each defect type. The adaptive weighting method enables the system to flexibly cope with various welding defects and improves detection accuracy.

[0106] Step S205: Input the weighted second optical data, second temperature data, and second plasma data into the defect detection network corresponding to the defect type identifier to obtain the welding defect detection result.

[0107] Specifically, different neural networks are used for different types of welding defects. Different types of defects (such as porosity, cracks, and incomplete welds) have different physical characteristics and behavioral patterns during their formation, and may exhibit different time-series characteristics and spatial distribution patterns. For example, porosity is usually closely related to the non-uniformity of the temperature field, and may require higher weighting of temperature sensor data; cracks usually show linear or non-linear changes in the image, and rely more on the spatial features provided by optical sensors; incomplete welds may show that the energy or temperature in certain areas is too low, requiring comprehensive analysis of data from multiple sensors. The signal characteristics of each defect type are different, and their detection methods should be customized according to the differences in characteristics to improve diagnostic accuracy.

[0108] For each defect type, different network structures are employed to optimize feature extraction and recognition. For porosity detection, convolutional neural networks (CNNs) can be designed to analyze spatial variations in the temperature field, as temperature changes typically exhibit a certain spatial distribution pattern. For cracks, image recognition networks (such as CNNs or ResNets) can be used to process welding images acquired by optical sensors, as cracks usually appear as linear or fissure-like features in images. For incomplete welds, regression analysis (such as fully connected layers) may be combined to analyze power fluctuations during the welding process and identify areas of substandard power. Using different neural network models or structures can better adapt to the characteristics of different defect types, thereby improving diagnostic accuracy.

[0109] Weighted second optical data, second temperature data, and second plasma data are input into a defect detection network corresponding to defect type identifiers to detect welding defects. A trained deep learning model accurately classifies defects in the welding process, yielding the welding defect detection results.

[0110] In this embodiment, the temporal and spatial features of optical monitoring data, temperature monitoring data, and plasma monitoring data are extracted to effectively capture key changes in the welding process, avoiding the problem that single sensor data cannot fully reflect welding defects. A cross-modal attention mechanism is used to interactively fuse the first optical data, first temperature data, and first plasma data, enhancing the correlation between different sensor data and improving the accuracy of defect identification. Finally, the weights of the second optical data, second temperature data, and second plasma data are dynamically adjusted according to the defect type identifier, making the contribution of each sensor data to the defect detection result more precise, further reducing the false detection rate. Furthermore, the corresponding defect detection network is selected based on the defect type identifier for welding defect detection, enabling accurate identification of complex defects in the welding process and improving the accuracy of automotive battery welding monitoring.

[0111] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0112] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0113] Further reference Figure 3 As a response to the above Figure 2 The implementation of the method shown in this application provides an embodiment of an automotive battery welding monitoring device, which is similar to... Figure 2Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0114] like Figure 3 As shown, the automotive battery welding monitoring device 300 described in this embodiment includes: a monitoring acquisition module 301, a feature extraction module 302, an interactive fusion module 303, a weight addition module 304, and a defect detection module 305, wherein:

[0115] The monitoring and acquisition module 301 is used to acquire optical monitoring data, temperature monitoring data and plasma monitoring data of the welding area when welding automotive batteries.

[0116] The feature extraction module 302 is used to extract temporal features and spatial features from optical monitoring data, temperature monitoring data and plasma monitoring data respectively, and generate first optical data, first temperature data and first plasma data based on the extracted features.

[0117] The interactive fusion module 303 is used to interactively fuse the first optical data, the first temperature data, and the first plasma data to obtain the second optical data, the second temperature data, and the second plasma data.

[0118] The weighting module 304 is used to obtain the defect type identifier and add weights to the second optical data, the second temperature data and the second plasma data according to the defect type identifier.

[0119] The defect detection module 305 is used to input weighted second optical data, second temperature data and second plasma data into a defect detection network corresponding to the defect type identifier to obtain welding defect detection results.

[0120] In this embodiment, the temporal and spatial features of optical monitoring data, temperature monitoring data, and plasma monitoring data are extracted to effectively capture key changes in the welding process, avoiding the problem that single sensor data cannot fully reflect welding defects. A cross-modal attention mechanism is used to interactively fuse the first optical data, first temperature data, and first plasma data, enhancing the correlation between different sensor data and improving the accuracy of defect identification. Finally, the weights of the second optical data, second temperature data, and second plasma data are dynamically adjusted according to the defect type identifier, making the contribution of each sensor data to the defect detection result more precise, further reducing the false detection rate. Furthermore, the corresponding defect detection network is selected based on the defect type identifier for welding defect detection, enabling accurate identification of complex defects in the welding process and improving the accuracy of automotive battery welding monitoring.

[0121] In some optional implementations of this embodiment, the monitoring and acquisition module 301 may include: an optical acquisition submodule, a temperature acquisition submodule, and a plasma acquisition submodule, wherein:

[0122] The optical acquisition submodule is used to acquire the morphology of the molten pool and weld seam in the welding area through optical sensors to obtain optical monitoring data.

[0123] The temperature acquisition submodule is used to collect the temperature field distribution of the molten pool in the welding area through temperature sensors to obtain temperature monitoring data.

[0124] The plasma acquisition submodule is used to collect the radiation characteristics of the plasma in the welding area through a plasma sensor to obtain plasma monitoring data.

[0125] In this embodiment, optical monitoring data reflects the welding morphology, temperature monitoring data reflects the thermal state, and plasma monitoring data provides clues about potential defects such as metal evaporation and porosity. Acquiring multi-dimensional monitoring data is beneficial for more comprehensive and accurate monitoring of welding quality.

[0126] In some optional implementations of this embodiment, the feature extraction module 302 may include: an optical extraction submodule, a temperature extraction submodule, and a plasma extraction submodule, wherein:

[0127] The optical extraction submodule is used to extract the change information of the image sequence in the optical monitoring data as optical temporal features, extract the spatial image features of the image sequence in the optical monitoring data, and obtain the first optical data based on the optical temporal features and spatial image features.

[0128] The temperature extraction submodule is used to extract temperature change information and temperature gradient from temperature monitoring data as temperature time series features, extract temperature field distribution and heat-affected zone from temperature monitoring data as temperature spatial features, and obtain the first temperature data based on the temperature time series features and temperature spatial features.

[0129] The plasma extraction submodule is used to extract radiation intensity change information and plasma fluctuation information from plasma monitoring data as plasma temporal features, extract plasma radiation distribution and plasma spectral features from plasma monitoring data as plasma spatial features, and obtain the first plasma data based on the plasma temporal features and plasma spatial features.

[0130] In this embodiment, temporal and spatial features are extracted from optical monitoring data, temperature monitoring data, and plasma monitoring data to comprehensively capture key changes in the welding process. Temporal features help capture the dynamic changes in the welding process, while spatial features provide geometric information about the welding process. This multi-dimensional feature extraction method provides more accurate input data for defect detection and improves the accuracy of automotive battery welding monitoring.

[0131] In some optional implementations of this embodiment, the interactive fusion module 303 may include: an optical fusion submodule, a temperature fusion submodule, and a plasma fusion submodule, wherein:

[0132] The optical fusion submodule is used to fuse the first temperature data and the first plasma data into the first optical data through a cross-modal attention mechanism to obtain the second optical data.

[0133] The temperature fusion submodule is used to fuse the first optical data and the first plasma data into the first temperature data through a cross-modal attention mechanism to obtain the second temperature data.

[0134] The plasma fusion submodule is used to fuse the first optical data and the first temperature data into the first plasma data through a cross-modal attention mechanism to obtain the second plasma data.

[0135] In this embodiment, by using a cross-modal attention mechanism to fuse data from different sensors, the intrinsic relationships between the data are automatically captured. The second optical data, second temperature data, and second plasma data generated after fusion can provide more comprehensive and higher quality feature information, better describe the changes and defects in the welding process, and improve the accuracy of welding defect detection.

[0136] In some optional implementations of this embodiment, the optical fusion submodule may include: a first fusion unit, a second fusion unit, and a data connection unit, wherein:

[0137] The first fusion unit is used to fuse the first optical data and the first temperature data through a cross-modal attention mechanism to obtain the first optical fusion data.

[0138] The second fusion unit is used to fuse the first optical data and the first plasma data through a cross-modal attention mechanism to obtain the second optical fused data.

[0139] The data connection unit is used to connect the first optical fusion data and the second optical fusion data to obtain the second optical data.

[0140] In this embodiment, the first and second optical fusion data generated by weighted fusion through cross-modal attention mechanism not only express optical data but also incorporate key information from temperature and plasma data. Multi-dimensional feature fusion can provide richer and more comprehensive welding process information, which helps to accurately identify welding defects.

[0141] In some optional implementations of this embodiment, the first fusion unit may include: a linear transformation subunit, a key value calculation subunit, a fraction calculation subunit, a value weighting subunit, and an optical fusion subunit, wherein:

[0142] The linear transformation subunit is used to perform a linear transformation on the first optical data to obtain the optical query vector.

[0143] The key value calculation subunit is used to calculate the temperature feature key and temperature feature value of the first temperature data.

[0144] The fractional calculation subunit is used to calculate the attention score between the optical query vector and the temperature feature key.

[0145] The value-weighted subunit is used to weight the temperature feature values ​​according to the attention score to obtain weighted temperature data.

[0146] The optical fusion subunit is used to add the weighted temperature data and the first optical data to obtain the first optical fusion data.

[0147] In this embodiment, the first optical data and the first temperature data are organically fused through a cross-modal attention mechanism, ensuring that the data from each sensor plays a better role in defect detection. The attention calculation between the optical query vector and the temperature feature key enables the system to dynamically adjust the weights according to the correlation of different data, enhance key information, and ensure that the most relevant part of the first temperature data has the greatest impact on the fusion result of the first optical data. The dynamic weighting mechanism improves the accuracy of the fused data and effectively enhances the accuracy of welding defect detection.

[0148] In some optional implementations of this embodiment, the weight addition module 304 may include: a combination acquisition submodule, a weight addition submodule, and a weight calculation submodule, wherein:

[0149] The combined acquisition submodule is used to obtain the weight combination corresponding to the defect type identifier. The weight combination is obtained in advance through learning.

[0150] The weighting submodule is used to add weights to the second optical data, the second temperature data, and the second plasma data based on the weight combination.

[0151] The weight calculation submodule is used to calculate the weights of the second optical data, the second temperature data, and the second plasma data based on the defect type identifier using a self-attention mechanism.

[0152] In this embodiment, the weight combination corresponding to the defect type identifier is obtained, and an appropriate weight is automatically assigned to each defect type, so that when the system detects a specific defect, it focuses on the most relevant data source to improve detection accuracy; or, through a self-attention mechanism, the weights of different data sources are dynamically adjusted according to the characteristics of each defect type. The adaptive weighting method enables the system to flexibly cope with various welding defects and improves detection accuracy.

[0153] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed] for details. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0154] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0155] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0156] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for automotive battery welding monitoring methods. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0157] In some embodiments, the processor 42 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions for the automotive battery welding monitoring method.

[0158] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0159] The computer device provided in this embodiment can execute the above-described automotive battery welding monitoring method. The automotive battery welding monitoring method here can be any of the automotive battery welding monitoring methods described in the various embodiments above.

[0160] In this embodiment, the temporal and spatial features of optical monitoring data, temperature monitoring data, and plasma monitoring data are extracted to effectively capture key changes in the welding process, avoiding the problem that single sensor data cannot fully reflect welding defects. A cross-modal attention mechanism is used to interactively fuse the first optical data, first temperature data, and first plasma data, enhancing the correlation between different sensor data and improving the accuracy of defect identification. Finally, the weights of the second optical data, second temperature data, and second plasma data are dynamically adjusted according to the defect type identifier, making the contribution of each sensor data to the defect detection result more precise, further reducing the false detection rate. Furthermore, the corresponding defect detection network is selected based on the defect type identifier for welding defect detection, enabling accurate identification of complex defects in the welding process and improving the accuracy of automotive battery welding monitoring.

[0161] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the automotive battery welding monitoring method described above.

[0162] In this embodiment, the temporal and spatial features of optical monitoring data, temperature monitoring data, and plasma monitoring data are extracted to effectively capture key changes in the welding process, avoiding the problem that single sensor data cannot fully reflect welding defects. A cross-modal attention mechanism is used to interactively fuse the first optical data, first temperature data, and first plasma data, enhancing the correlation between different sensor data and improving the accuracy of defect identification. Finally, the weights of the second optical data, second temperature data, and second plasma data are dynamically adjusted according to the defect type identifier, making the contribution of each sensor data to the defect detection result more precise, further reducing the false detection rate. Furthermore, the corresponding defect detection network is selected based on the defect type identifier for welding defect detection, enabling accurate identification of complex defects in the welding process and improving the accuracy of automotive battery welding monitoring.

[0163] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0164] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A method for monitoring automotive battery welding, characterized in that, Includes the following steps: When welding automotive batteries, acquire optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area; Temporal and spatial features are extracted from the optical monitoring data, temperature monitoring data, and plasma monitoring data, respectively, and first optical data, first temperature data, and first plasma data are generated based on the extracted features. By using a cross-modal attention mechanism, the first optical data, the first temperature data, and the first plasma data are interactively fused to obtain the second optical data, the second temperature data, and the second plasma data. Obtain the defect type identifier, and add weights to the second optical data, the second temperature data, and the second plasma data according to the defect type identifier; The weighted second optical data, second temperature data, and second plasma data are input into a defect detection network corresponding to the defect type identifier to obtain welding defect detection results. The steps for acquiring optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area include: Optical monitoring data is obtained by acquiring the morphology of the molten pool and weld seam in the welding area using optical sensors. Temperature monitoring data is obtained by collecting the temperature field distribution of the molten pool in the welding area using temperature sensors. The radiation characteristics of the plasma in the welding area are collected by a plasma sensor to obtain plasma monitoring data. The steps of performing temporal feature extraction and spatial feature extraction on the optical monitoring data, the temperature monitoring data, and the plasma monitoring data, respectively, and generating first optical data, first temperature data, and first plasma data based on the extracted features include: The change information of the image sequence in the optical monitoring data is extracted as optical temporal features, the spatial image features of the image sequence in the optical monitoring data are extracted, and the first optical data is obtained based on the optical temporal features and the spatial image features. The temperature change information and temperature gradient of the temperature monitoring data are extracted as temperature time series features. The temperature field distribution and heat-affected zone are extracted from the temperature monitoring data as temperature spatial features. The first temperature data is obtained based on the temperature time series features and the temperature spatial features. The radiation intensity change information and plasma fluctuation information of the plasma monitoring data are extracted as plasma temporal features, and the plasma radiation distribution and plasma spectral features of the plasma monitoring data are extracted as plasma spatial features. The first plasma data is obtained based on the plasma temporal features and the plasma spatial features.

2. The automotive battery welding monitoring method according to claim 1, characterized in that, The step of interactively fusing the first optical data, the first temperature data, and the first plasma data to obtain the second optical data, the second temperature data, and the second plasma data includes: The first temperature data and the first plasma data are fused into the first optical data through a cross-modal attention mechanism to obtain the second optical data; The first optical data and the first plasma data are fused into the first temperature data through the cross-modal attention mechanism to obtain the second temperature data; The first optical data and the first temperature data are fused into the first plasma data through the cross-modal attention mechanism to obtain the second plasma data.

3. The automotive battery welding monitoring method according to claim 2, characterized in that, The step of fusing the first temperature data and the first plasma data into the first optical data through a cross-modal attention mechanism to obtain the second optical data includes: The first optical data and the first temperature data are fused using a cross-modal attention mechanism to obtain the first optical fused data; The first optical data and the first plasma data are fused using the cross-modal attention mechanism to obtain the second optical fused data; The first optical fusion data and the second optical fusion data are concatenated to obtain the second optical data.

4. The automotive battery welding monitoring method according to claim 3, characterized in that, The step of fusing the first optical data and the first temperature data through a cross-modal attention mechanism to obtain the first optical fused data includes: The first optical data is linearly transformed to obtain the optical query vector; Calculate the temperature feature key and temperature feature value of the first temperature data; Calculate the attention score between the optical query vector and the temperature feature key; The temperature feature values ​​are weighted according to the attention scores to obtain weighted temperature data; The weighted temperature data and the first optical data are added together to obtain the first optical fusion data.

5. The automotive battery welding monitoring method according to claim 1, characterized in that, The step of adding weights to the second optical data, the second temperature data, and the second plasma data according to the defect type identifier includes: Obtain a weight combination corresponding to the defect type identifier, the weight combination being learned in advance; Weights are added to the second optical data, the second temperature data, and the second plasma data according to the weight combination; or... Based on the defect type identifier, the weights of the second optical data, the second temperature data, and the second plasma data are calculated using a self-attention mechanism.

6. A monitoring device for automotive battery welding, characterized in that, include: The monitoring and acquisition module is used to acquire optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area when welding automotive batteries. The feature extraction module is used to extract temporal features and spatial features from the optical monitoring data, the temperature monitoring data and the plasma monitoring data respectively, and generate first optical data, first temperature data and first plasma data based on the extracted features; The interactive fusion module is used to interactively fuse the first optical data, the first temperature data, and the first plasma data through a cross-modal attention mechanism to obtain second optical data, second temperature data, and second plasma data. The weighting module is used to obtain the defect type identifier and add weights to the second optical data, the second temperature data and the second plasma data according to the defect type identifier; The defect detection module is used to input the weighted second optical data, second temperature data and second plasma data into a defect detection network corresponding to the defect type identifier to obtain welding defect detection results; The acquisition of optical monitoring data, temperature monitoring data, and plasma monitoring data of the welding area includes: Optical monitoring data is obtained by acquiring the morphology of the molten pool and weld seam in the welding area using optical sensors. Temperature monitoring data is obtained by collecting the temperature field distribution of the molten pool in the welding area using temperature sensors. The radiation characteristics of the plasma in the welding area are collected by a plasma sensor to obtain plasma monitoring data. The step of performing temporal feature extraction and spatial feature extraction on the optical monitoring data, the temperature monitoring data, and the plasma monitoring data, respectively, and generating first optical data, first temperature data, and first plasma data based on the extracted features includes: The change information of the image sequence in the optical monitoring data is extracted as optical temporal features, the spatial image features of the image sequence in the optical monitoring data are extracted, and the first optical data is obtained based on the optical temporal features and the spatial image features. The temperature change information and temperature gradient of the temperature monitoring data are extracted as temperature time series features. The temperature field distribution and heat-affected zone are extracted from the temperature monitoring data as temperature spatial features. The first temperature data is obtained based on the temperature time series features and the temperature spatial features. The radiation intensity change information and plasma fluctuation information of the plasma monitoring data are extracted as plasma temporal features, and the plasma radiation distribution and plasma spectral features of the plasma monitoring data are extracted as plasma spatial features. The first plasma data is obtained based on the plasma temporal features and the plasma spatial features.

7. A computer device comprising a memory and a processor, the memory storing computer-readable instructions, wherein the processor, when executing the computer-readable instructions, implements the steps of the automotive battery welding monitoring method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the automotive battery welding monitoring method as described in any one of claims 1 to 5.

Images