Control system, method, apparatus, medium and program product for a welding process

Abstract

The present disclosure provides a control system, method, apparatus, medium and program product for a welding process. The control system comprises a sensor module configured to collect data in the welding process; the data analysis module is configured to pre-process data in the welding process, extract key features from the pre-processed data, perform weighted fusion processing on the key features to obtain comprehensive features, and perform defect identification through a machine learning model based on the comprehensive features to identify defect features; the knowledge graph module is configured to obtain a repair strategy corresponding to the defect characteristics according to the defect characteristics and by utilizing a knowledge graph; the control module is configured to transmit the data received from the sensor module in the welding process to the data analysis module, receive the repair strategy and generate a repair instruction based on the repair strategy; and the repairing module is configured to automatically adjust welding parameters or control an executing mechanism to conduct repairing operation based on the repairing instruction.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent manufacturing, and in particular to a control system, method, equipment, medium and program product for a welding process. Background Art

[0002] With the rapid development of intelligent manufacturing, welding, as one of the key processes in the manufacturing industry, has a direct impact on the reliability and service life of products. In related technologies, welding defect repair mainly relies on manual intervention. Especially in complex welding scenarios, manual repair often takes a lot of time and manpower, and is prone to secondary defects, affecting the overall welding quality and production efficiency. Summary of the invention

[0003] A technical problem solved by the present disclosure is that the methods in the related art affect welding quality and production efficiency.

[0004] According to one aspect of the present disclosure, a control system for a welding process is provided, comprising: a sensor module, configured to collect data during a welding process; a data analysis module, configured to preprocess the data during the welding process, extract key features from the preprocessed data, perform weighted fusion processing on the key features to obtain comprehensive features, and perform defect identification based on the comprehensive features and through a machine learning model to identify defect features; a knowledge graph module, configured to obtain a repair strategy corresponding to the defect feature based on the defect feature and using a knowledge graph, wherein the knowledge graph is a knowledge graph including welding parameters, defect features and repair strategies; a control module, configured to transmit the data during the welding process received from the sensor module to the data analysis module, receive the repair strategy, and generate a repair instruction based on the repair strategy; and a repair module, configured to automatically adjust the welding parameters or control the actuator to perform a repair operation based on the repair instruction.

[0005] In some embodiments, the knowledge graph is stored in the form of a graph structure, the nodes of the knowledge graph represent the welding parameters, the defect characteristics or the repair strategies, and the edges of the knowledge graph represent the association relationships between different nodes.

[0006] In some embodiments, the sensor module includes: a visual sensor configured to collect image data during the welding process; an acoustic sensor configured to collect sound wave signal data during the welding process; and an infrared sensor configured to collect temperature distribution characteristic data during the welding process; wherein the data during the welding process includes the image data, the sound wave signal data and the temperature distribution characteristic data.

[0007] In some embodiments, the data analysis module is configured to perform denoising, normalization and time alignment processing on the image data, the acoustic wave signal data and the temperature distribution characteristic data.

[0008] In some embodiments, the key features include: image key features in the image data , the acoustic spectrum features in the acoustic wave signal data and the temperature gradient characteristics in the temperature distribution characteristic data ; wherein the comprehensive characteristics for

[0009] ,

[0010] in, is the weight of the key features of the image, is the weight of the acoustic spectrum feature, is the weight of the temperature gradient feature.

[0011] In some embodiments, the data analysis module is configured to perform defect identification based on the comprehensive features, combined with the knowledge graph, and through the machine learning model.

[0012] In some embodiments, the control module is further configured to dynamically schedule the data collection frequency and data transmission rate of the sensor module.

[0013] In some embodiments, the sensor module is also configured to collect data during the repair process; the data analysis module is also configured to determine whether the repair result meets the expected result based on the data during the repair process; if the repair result meets the expected result, the data during the repair process is transmitted to the knowledge graph module; if the repair result does not meet the expected result, re-repair information is sent to the knowledge graph module; the knowledge graph module is also configured to record the data during the repair process and update the knowledge graph when the repair result meets the expected result; if the repair result does not meet the expected result, a re-repair strategy different from the current repair strategy is sent to the control module based on the re-repair information; the control module is also configured to send a re-repair instruction to the repair module based on the re-repair strategy, so that the repair module performs the repair operation again.

[0014] In some embodiments, the control module is further configured to detect a current working state of the sensor module, and adjust working parameters of the sensor module according to historical data of the working state of the sensor module and the current working state.

[0015] According to another aspect of the present disclosure, a control method for a welding process is provided, including: receiving data in a welding process; preprocessing the data in the welding process, extracting key features from the preprocessed data, performing weighted fusion processing on the key features to obtain comprehensive features, performing defect identification based on the comprehensive features and through a machine learning model to identify defect features; obtaining a repair strategy corresponding to the defect feature based on the defect feature and using a knowledge graph, wherein the knowledge graph is a knowledge graph including welding parameters, defect features and repair strategies; generating repair instructions based on the repair strategy; and automatically adjusting welding parameters or controlling an actuator to perform a repair operation based on the repair instructions.

[0016] In some embodiments, the knowledge graph is stored in the form of a graph structure, the nodes of the knowledge graph represent the welding parameters, the defect characteristics or the repair strategies, and the edges of the knowledge graph represent the association relationships between different nodes.

[0017] In some embodiments, the data during the welding process includes image data, acoustic wave signal data, and temperature distribution characteristic data.

[0018] In some embodiments, the data in the welding process is preprocessed, including: denoising, normalizing and time aligning the image data, the acoustic wave signal data and the temperature distribution characteristic data.

[0019] In some embodiments, the key features include: image key features in the image data , the acoustic spectrum features in the acoustic wave signal data and the temperature gradient characteristics in the temperature distribution characteristic data ; wherein the comprehensive characteristics for

[0020] ,

[0021] in, is the weight of the key features of the image, is the weight of the acoustic spectrum feature, is the weight of the temperature gradient feature.

[0022] In some embodiments, defect identification is performed based on the comprehensive features and through a machine learning model, including: based on the comprehensive features, combined with the knowledge graph, and defect identification is performed through the machine learning model.

[0023] In some embodiments, the control method further includes: dynamically scheduling the data acquisition frequency and data transmission rate of the sensor module.

[0024] In some embodiments, the control method also includes: receiving data from a repair process; judging whether a repair result meets an expected result based on the data from the repair process; if the repair result meets the expected result, recording the data from the repair process and updating the knowledge graph; if the repair result does not meet the expected result, obtaining a re-repair strategy different from the current repair strategy; generating a re-repair instruction based on the re-repair strategy; and re-performing the repair operation based on the re-repair instruction.

[0025] In some embodiments, the control method further includes: detecting a current working state of the sensor module; and adjusting working parameters of the sensor module according to historical data of the working state of the sensor module and the current working state.

[0026] According to another aspect of the present disclosure, an electronic device is provided, including: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the control method as described above based on instructions stored in the memory.

[0027] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which computer instructions are stored. When the computer instructions are executed by a processor, the control method as described above is implemented.

[0028] According to another aspect of the present disclosure, a computer program product is provided. The computer program product includes a computer program or instructions. When the computer program or instructions are executed by a processor, the control method as described above is implemented.

[0029] In the above control system, data from the welding process is collected, and defects are identified after relevant processing of these data. The repair strategy corresponding to the defect is obtained through the knowledge graph, and the corresponding repair instructions are generated. Then, based on the repair instructions, the welding parameters are automatically adjusted or the actuator is controlled to perform the repair operation. In this way, the automation and intelligence of the repair defect of the welding process is realized, and the welding quality and production efficiency are improved.

[0030] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0032] The present disclosure may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0033] Figure 1is a block diagram schematically illustrating a control system for a welding process according to some embodiments of the present disclosure;

[0034] Figure 2 is a block diagram schematically illustrating a structure of a sensor module according to some embodiments of the present disclosure;

[0035] Figure 3 is a block diagram schematically illustrating a structure of a data analysis module according to some embodiments of the present disclosure;

[0036] Figure 4 is a block diagram schematically illustrating a structure of a knowledge graph module according to some embodiments of the present disclosure;

[0037] Figure 5 is a block diagram schematically illustrating a control module according to some embodiments of the present disclosure;

[0038] Figure 6 is a block diagram schematically illustrating a repair module according to some embodiments of the present disclosure;

[0039] Figure 7 is a flow chart illustrating a control method for a welding process according to some embodiments of the present disclosure;

[0040] Figure 8 is a flow chart showing a control method for a welding process according to other embodiments of the present disclosure;

[0041] Fig. 9 is a block diagram schematically showing a structure of an electronic device according to some embodiments of the present disclosure;

[0042] Fig.10 is a block diagram schematically showing a structure of an electronic device according to some other embodiments of the present disclosure. DETAILED DESCRIPTION

[0043] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0044] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0045] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0046] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0047] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0048] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0049] Figure 1 Schematically shows a block diagram of a control system for a welding process according to some embodiments of the present disclosure. Figure 1 As shown, the control system includes a sensor module 110 , a data analysis module 120 , a knowledge graph module 130 , a control module 140 and a repair module 150 .

[0050] The sensor module 110 is configured to collect data during the welding process.

[0051] The data analysis module 120 is configured to preprocess the data in the welding process, extract key features from the preprocessed data, perform weighted fusion processing on the key features to obtain comprehensive features, and perform defect identification based on the comprehensive features and through a machine learning model to identify defect features. Here, the machine learning model is a machine learning algorithm, and a known machine learning algorithm can be used.

[0052] The knowledge graph module 130 is configured to obtain a repair strategy corresponding to the defect feature based on the defect feature and using the knowledge graph. The knowledge graph is a knowledge graph including welding parameters, defect features and repair strategies.

[0053] For example, the knowledge graph is stored in the form of a graph structure. The nodes of the knowledge graph represent welding parameters, defect characteristics or repair strategies, and the edges of the knowledge graph represent the association relationships between different nodes.

[0054] The control module 140 is configured to transmit the data in the welding process received from the sensor module 110 to the data analysis module 120 , receive the repair strategy, generate a repair instruction based on the repair strategy, and send the repair instruction to the repair module 150 .

[0055] The repair module 150 is configured to automatically adjust welding parameters or control actuators to perform repair operations based on the repair instructions.

[0056] Here, the welding parameters are relevant parameters in the welding process, such as welding current, welding voltage, welding speed, etc. The repair module 150 can automatically adjust the welding parameters based on the repair instruction, so that the adjusted welding parameters meet the desired welding effect.

[0057] For example, the actuator may be a welding robot. That is, the repair module 150 may control the welding robot to perform a repair operation based on the repair instruction.

[0058] So far, a control system for a welding process according to some embodiments of the present disclosure is provided. The control system includes: a sensor module, configured to collect data during the welding process; a data analysis module, configured to preprocess the data during the welding process, extract key features from the preprocessed data, perform weighted fusion processing on the key features to obtain comprehensive features, perform defect identification based on the comprehensive features and through a machine learning model to identify defect features; a knowledge graph module, configured to obtain a repair strategy corresponding to the defect feature based on the defect feature and using the knowledge graph, wherein the knowledge graph is a knowledge graph containing welding parameters, defect features and repair strategies; a control module, configured to transmit the data received from the sensor module during the welding process to the data analysis module, and receive the repair strategy, and generate a repair instruction based on the repair strategy; and a repair module, configured to automatically adjust the welding parameters or control the actuator to perform a repair operation based on the repair instruction. In this control system, data from the welding process is collected and defects are identified after relevant processing. The repair strategy corresponding to the defect is obtained through the knowledge graph, and the corresponding repair instructions are generated. Then, based on the repair instructions, the welding parameters are automatically adjusted or the actuator is controlled to perform the repair operation. In this way, the automation and intelligence of the repair defect of the welding process is realized, and the welding quality and production efficiency are improved.

[0059] Moreover, in the above system, the association relationships in the knowledge graph can be used to quickly infer the best repair plan, realizing the combination of static knowledge and real-time data, which makes it easier for the system to respond and make decisions quickly after detecting defects in real time, and then repair welding defects in time during the welding process, thereby improving the product yield.

[0060] Combine the following Figures 2 to 6 Describe the structure or function of each module in detail.

[0061] Figure 2 is a block diagram schematically showing a structure of a sensor module according to some embodiments of the present disclosure.

[0062] like Figure 2 As shown, the sensor module 110 may include: a visual sensor 111 , an acoustic sensor 112 , and an infrared sensor 113 .

[0063] The vision sensor 111 is configured to collect image data during the welding process.

[0064] The acoustic sensor 112 is configured to collect acoustic wave signal data during the welding process.

[0065] The infrared sensor 113 is configured to collect temperature distribution characteristic data during the welding process.

[0066] In this way, the data during the welding process may include image data, acoustic wave signal data, and temperature distribution characteristic data.

[0067] In the related art, welding quality detection technology usually relies on a single type of sensor. The detection method based on a single sensor has limitations in the comprehensiveness and accuracy of information acquisition due to the single data source. In the above embodiment of the present disclosure, a visual sensor, an acoustic sensor and an infrared sensor are set as a sensor module (which can be called a multimodal sensor module) to facilitate the subsequent use of multiple sensor data such as vision, acoustics, infrared, etc., so as to achieve more comprehensive and accurate welding quality detection.

[0068] In other embodiments, in some scenarios, one or two sensors among visual sensors, acoustic sensors and infrared sensors can be used as sensor modules, and the detection of welding defects in these specific scenarios can also be achieved.

[0069] In other embodiments, the sensor module may include other sensors in addition to the visual sensor, the acoustic sensor and the infrared sensor, which may further improve the accuracy of defect detection.

[0070] In some embodiments, Figure 2 As shown, a synchronous clock system 114 can be provided in the sensor module, so that the visual sensor, acoustic sensor and infrared sensor can realize real-time data collection and synchronous transmission through the synchronous clock system, and the time between various sensor data can be kept consistent and spatially aligned.

[0071] In this embodiment, all sensors are connected to a unified synchronous clock system 114. The synchronous clock system provides an accurate time reference for each sensor to ensure synchronous data collection at the same time point. The synchronous clock system can be implemented in hardware (for example, using GPS (Global Positioning System) time synchronization or Network Time Protocol (NTP), etc.), or it can be based on software algorithms to calibrate and adjust the time differences between different sensors.

[0072] For example, during the data collection process, the data collected by each sensor will be accompanied by a precise timestamp, which is generated by a synchronized clock system to ensure that data from different sensors can be compared and aligned under the same time reference.

[0073] For example, in the data fusion stage, the data stream is corrected and aligned using the data alignment algorithm using the time-stamped data. This algorithm takes into account the possible time deviation of the sensor and achieves accurate data alignment through methods such as interpolation and delay correction.

[0074] In other embodiments, various types of data in the welding process are spatially aligned, that is, data from different sensors correspond to the same position or area in space, so that the system can fuse and analyze multimodal data in a unified spatial coordinate system, thereby improving the accuracy and reliability of detection.

[0075] In the above embodiment, the sensor module includes a visual sensor, an acoustic sensor, and an infrared sensor. These sensors are used to collect image data, acoustic wave signals, and temperature distribution characteristics during the welding process. These sensors realize real-time data collection and synchronous transmission through a synchronous clock system, and ensure the time consistency and spatial alignment between various sensor data as much as possible.

[0076] Figure 3 Schematically illustrates a structural block diagram of a data analysis module according to some embodiments of the present disclosure. The data analysis module is mainly responsible for fusing and analyzing the data collected by the multi-modal sensor.

[0077] like Figure 3 As shown, the data analysis module 120 can pre-process the data in the welding process. For example, the data analysis module 120 is configured to perform denoising, normalization and time alignment on the image data, the acoustic wave signal data and the temperature distribution characteristic data.

[0078] Here, a denoising method known to those skilled in the art may be used to perform denoising processing.

[0079] For example, in the normalization process, the normalized data for

[0080] , (1)

[0081] in, is the original data, is the mean of the data, is the standard deviation of the data.

[0082] By normalizing the image data, acoustic wave signal data and temperature distribution characteristic data respectively, it is convenient to perform weighted fusion of these different types of data.

[0083] Time alignment refers to aligning data at the same time. For example, if 10 images and 20 sound wave signals are collected in the same time period, the images and sound wave signals at the same time are aligned, and other sound wave signals can be removed. Time alignment can facilitate weighted fusion of data.

[0084] In other embodiments, the data analysis module 120 may also be configured to perform spatial alignment processing on the image data, the acoustic wave signal data, and the temperature distribution characteristic data. That is, the image data, the acoustic wave signal data, and the temperature distribution characteristic data collected at the same part are aligned. Through the spatial alignment processing, weighted fusion of data can be facilitated.

[0085] In the above embodiments, by performing pre-processing such as denoising, normalization, time alignment and space alignment on the collected visual data, acoustic data and infrared data, a unified standard format is provided for subsequent data processing, thereby facilitating the processing of information redundancy.

[0086] like Figure 3 As shown, the data analysis module 120 can also extract key features from the preprocessed data. For example, the key features include: image key features in the image data , Acoustic spectrum features in acoustic wave signal data and temperature gradient characteristics in temperature distribution characteristic data For example, the key image features may include weld width and / or weld pool shape parameters, etc.

[0087] like Figure 3 As shown, the data analysis module 120 can also perform weighted fusion processing on the key features to obtain comprehensive features. That is, these features are fused through a weighted fusion algorithm to generate a comprehensive feature vector.

[0088] For example, the comprehensive characteristics for

[0089] , (2)

[0090] in, is the weight of the key features of the image (i.e., visual data features), is the weight of the acoustic spectrum feature, is the weight of the temperature gradient feature. These weight values ​​can be dynamically adjusted according to the reliability of various types of data in the actual production environment.

[0091] In the weighted fusion process, the weights of the key image features, acoustic spectrum features, and temperature gradient features (i.e., infrared temperature features) can be dynamically adjusted according to the reliability in the specific application scenario and production environment. That is, the specific values ​​of these weights can be adjusted and optimized according to the actual usage.

[0092] In some embodiments, the weights of the key image features (ie, visual data features) can range from 0.3 to 0.5. For example, in an environment with good lighting conditions, the visual sensor can provide high-quality image data, so a higher weight can be given (for example, For example, in an environment with a lot of smoke or unfavorable lighting, the accuracy of the visual data may be affected and the weight may be reduced (for example, can be 0.35).

[0093] In some embodiments, the weights of the acoustic spectrum features can range from 0.2 to 0.4. For example, if the acoustic signal generated by the welding process is very obvious and has a significant feature, the weight of the acoustic feature can be increased (for example, For example, if there is background noise or the sound feature is not obvious, the weight may be reduced (for example, can be 0.25).

[0094] In some embodiments, the weight of the temperature gradient feature The range can be 0.3 to 0.5. For example, in a scenario where welding temperature is a key indicator for defect detection, the importance of the temperature gradient feature is high, and the weight may be set to 0.45. For another example, if the ambient temperature changes greatly, making the infrared data unstable, the weight may be reduced to 0.35.

[0095] The above weight , and The sum of these weights is 1, and the adjustment of these weights can be optimized through experimental data analysis to find the best combination suitable for various production environments. Dynamic adjustment of weights can flexibly respond to changes in the quality of various data sources to improve the accuracy and reliability of defect detection.

[0096] In the above embodiment, key features such as weld width, molten pool shape parameters, spectrum characteristics, temperature gradient, etc. are extracted from various types of data, and then these features are fused through a weighted fusion algorithm to generate a comprehensive feature vector. The weight value of the weighted fusion is dynamically adjusted according to the reliability of the data, which improves the accuracy of feature extraction and information fusion analysis.

[0097] like Figure 3As shown, the data analysis module 120 can also perform defect identification based on comprehensive features and through a machine learning model to identify defect features. For example, the data analysis module is configured to perform defect identification based on comprehensive features, combined with a knowledge graph, and through a machine learning model. In this way, the accuracy of defect identification can be improved.

[0098] The fused comprehensive feature vector is combined with the defect model in the knowledge graph and a machine learning algorithm (for example, a support vector machine (SVM) or a convolutional neural network (CNN)) is used to perform defect identification.

[0099] For example, the process of defect identification includes the following steps.

[0100] In step 1, feature fusion is performed. That is, the visual, acoustic, and infrared sensor data are integrated into a comprehensive feature vector.

[0101] In step 2, the knowledge graph is used to provide context. That is, the knowledge graph is used to provide semantic and prior information related to the defect.

[0102] In step 3, the model is trained. That is, the labeled data is used to train a machine learning model (e.g., SVM or CNN, etc.) to identify the relationship between features and defect types.

[0103] In step 4, online recognition is performed. In practical applications, the feature vector of new data is input into the model and defect recognition is performed in combination with the knowledge graph.

[0104] By combining the data of multiple sensors such as vision, acoustics, and infrared, a more comprehensive and accurate welding quality inspection can be achieved. Multimodal data fusion technology can effectively make up for the shortcomings of a single data source and enhance the robustness and detection accuracy of the system through collaborative analysis of different types of sensor data. The above process enhances the accuracy and efficiency of defect detection by integrating different technologies. For example, the accuracy of the above detection algorithm of the embodiment of the present disclosure can reach more than 95% in an experimental environment, which is significantly better than the 85% of the single sensor data detection method in the related art.

[0105] In addition, by structuring and associating information such as parameters, processes, and defect types in the welding process through knowledge graphs, data analysis and decision support can be effectively performed.

[0106] Figure 4 It is a block diagram schematically showing the structure of a knowledge graph module according to some embodiments of the present disclosure.

[0107] The knowledge graph module 130 is responsible for managing relevant knowledge such as welding process, defect type and repair strategy, and analyzes and makes decisions on real-time data through reasoning mechanism. A knowledge base including welding parameters, defect characteristics and repair strategies can be established based on welding process standards, historical data and expert knowledge. The knowledge base is stored in the form of a graph structure (i.e., knowledge graph).

[0108] In the knowledge base, nodes represent parameters (e.g., welding parameters or defect characteristics) or knowledge points (e.g., repair strategies). For example, the knowledge base may include parameter nodes (representing welding parameters), defect nodes (representing defect characteristics), and repair nodes (representing repair strategies). For another example, the knowledge base may also include: standard files and / or experience files, etc. Standard files may include process selection, defect judgment, and repair methods, etc.

[0109] In a knowledge base, an edge represents the association between one node and another node. For example, an edge can represent the correspondence between a defect and a repair strategy.

[0110] By using the associations in the knowledge graph, the system can use the reasoning mechanism to quickly infer the best repair solution after detecting a defect. For example, when it is detected that the weld temperature is too high, the system can infer that the repair solution needs to reduce the welding current or increase the cooling time. The reasoning mechanism can use a logistic regression algorithm to improve the efficiency and accuracy of reasoning.

[0111] In the above embodiment, the knowledge graph module is responsible for managing relevant knowledge such as welding process, defect type, and repair strategy, and analyzes and makes decisions on real-time data through reasoning mechanism. By establishing a knowledge base containing welding parameters, defect characteristics and repair strategies, the optimal repair plan is quickly inferred using the association relationship in the knowledge graph, realizing the combination of static knowledge and real-time data.

[0112] Furthermore, the subsequent control module converts the repair plan into specific adjusted welding parameters or operating instructions, which enables the system to respond and make decisions quickly after detecting defects in real time.

[0113] Figure 5 is a block diagram schematically illustrating a structure of a control module according to some embodiments of the present disclosure.

[0114] The control module 140 is the core coordination unit of the system, responsible for managing the communication and coordination between the sensor module, data analysis module, knowledge graph module and repair module.

[0115] In some embodiments, the control module 140 is further configured to dynamically schedule the data acquisition frequency and data transmission rate of the sensor module. Here, the control module can dynamically schedule the data acquisition frequency and transmission rate of each sensor of the sensor module according to system requirements to improve the real-time performance of the system.

[0116] After detecting a welding defect, the control module sends a repair instruction to the repair module based on the reasoning results of the knowledge graph module. In some cases, the control module can also monitor the execution of the repair process.

[0117] In some embodiments, the control module 140 is further configured to detect the current working state of the sensor module, and adjust the working parameters of the sensor module according to the historical data of the working state of the sensor module and the current working state. For example, the historical data may include the working state of the sensor module and the working parameters of the sensor module corresponding to the working state. Therefore, during the operation of the system, the control module can self-check the working state of each module, and automatically adjust or optimize the parameters in certain specific cases to improve the stability and efficiency of the system.

[0118] Figure 6 is a block diagram schematically illustrating a structure of a repair module according to some embodiments of the present disclosure.

[0119] The repair module 150 is used to execute the repair solution inferred by the system. The repair module 150 can automatically adjust the parameters of the welding equipment (i.e., welding parameters), such as welding current, welding speed, etc., to deal with the detected welding defects. In the case of complex defects (e.g., cracks, pores), the repair module 150 can perform repair operations by controlling the welding robot, such as repair welding, grinding, etc. For example, the accuracy of the welding robot control can be within ±0.1 mm (millimeter) to improve the accuracy of the repair.

[0120] After the repair is completed, the system can perform defect detection again to test the effect of the repair. If the repair effect does not meet the expected effect, the system will further adjust the parameters or perform a second repair.

[0121] So far, the structure or function of each of the above modules has been described in detail.

[0122] In some embodiments, the sensor module is also configured to collect data during the repair process. The data analysis module is also configured to determine whether the repair result meets the expected result based on the data during the repair process. If the repair result meets the expected result, the data in the repair process is transmitted to the knowledge graph module. If the repair result does not meet the expected result, the re-repair information is sent to the knowledge graph module. The knowledge graph module is also configured to record the data in the repair process and update the knowledge graph when the repair result meets the expected result. If the repair result does not meet the expected result, the re-repair strategy different from the current repair strategy is sent to the control module based on the re-repair information. The control module is also configured to send a re-repair instruction to the repair module based on the re-repair strategy so that the repair module performs the repair operation again. In this way, after the repair is completed, the system can detect the quality of the weld again to determine whether the repair result meets the expected result. If the repair result of the detection meets the expected result, the system records the repair process and updates the knowledge graph; if the repair result of the detection does not meet the expected result, the system will perform a secondary repair. In this way, by updating the knowledge graph, it is convenient to provide a more accurate repair plan in the future, and the effect of the repair can be improved through secondary repair.

[0123] Therefore, after each repair operation, the system will record the data of the detection and repair process in the knowledge graph module for analysis and updating of the knowledge base, so that the knowledge graph can be dynamically updated and self-learned, improving the level of integration of the knowledge graph and real-time data.

[0124] In the related art, welding defect detection and repair are usually carried out separately, that is, welding is completed first, and then detection and repair are carried out. This method has a significant time lag, resulting in defects not being repaired in time during the welding process, and may even cause product scrapping due to delays.

[0125] In the embodiment of the present disclosure, through the real-time data acquisition of the sensor module, the rapid defect identification of the data analysis module, and the real-time decision generation of the knowledge graph module, the system can continuously monitor the quality during the welding process, and once a defect is found, the repair process is immediately triggered; the control module, as the coordination unit of the system, can immediately issue a repair instruction according to the reasoning result of the knowledge graph after detecting the defect, and monitor the execution of the repair process in real time, thereby eliminating the time interval between detection and repair; the repair module can automatically adjust the parameters of the welding equipment or control the welding robot to perform repair operations, such as repair welding, grinding, etc., immediately after the system detects the welding defect. Through the real-time feedback mechanism, the system will detect the quality of the weld again after the repair is completed to ensure the repair result. Therefore, the control system of the embodiment of the present disclosure can realize real-time monitoring, defect detection, intelligent analysis and automatic repair of the welding process through the collaborative work of the above modules, which can solve the problem that welding defect detection and repair are usually carried out separately in the related art, so that the system can detect and repair defects in time during the welding process.

[0126] So far, the embodiments of the present disclosure provide a real-time detection and automatic repair system for welding defects based on multimodal data fusion and knowledge graph. The system integrates multiple sensor data, knowledge graph reasoning and automated control technology to achieve high-precision defect detection and automatic repair in the welding process, thereby improving the overall welding quality, reducing production costs and enhancing production efficiency.

[0127] Figure 7 is a flow chart showing a control method for a welding process according to some embodiments of the present disclosure. For example, the control method can be executed in an electronic device. Figure 7 As shown, the control method includes steps S702 to S710.

[0128] In step S702, data during the welding process is received.

[0129] For example, the data during the welding process includes image data, acoustic wave signal data, and temperature distribution characteristic data.

[0130] In step S704, the data of the welding process is preprocessed, key features are extracted from the preprocessed data, and the key features are weighted fused to obtain comprehensive features. Defect identification is performed based on the comprehensive features and through a machine learning model to identify defect features.

[0131] In some embodiments, the data in the welding process is preprocessed, including: denoising, normalizing and time aligning the image data, the acoustic wave signal data and the temperature distribution characteristic data.

[0132] In some embodiments, the key features include: image key features in the image data , Acoustic spectrum features in acoustic wave signal data and temperature gradient characteristics in temperature distribution characteristic data .

[0133] In some embodiments, the comprehensive features for

[0134] ,

[0135] in, is the weight of the key features of the image, is the weight of the acoustic spectrum feature, is the weight of the temperature gradient feature.

[0136] In some embodiments, defect identification is performed based on comprehensive features and through a machine learning model, including: defect identification is performed based on comprehensive features, combined with a knowledge graph, and through a machine learning model.

[0137] In step S706, a repair strategy corresponding to the defect feature is obtained based on the defect feature and using the knowledge graph. The knowledge graph is a knowledge graph including welding parameters, defect features and repair strategies.

[0138] The knowledge graph is stored in the form of a graph structure. The nodes of the knowledge graph represent welding parameters, defect characteristics or repair strategies, and the edges of the knowledge graph represent the association relationships between different nodes.

[0139] In step S708, a repair instruction is generated based on the repair strategy.

[0140] In step S710, based on the repair instruction, the welding parameters are automatically adjusted or the actuator is controlled to perform the repair operation.

[0141] So far, a control method for a welding process according to some embodiments of the present disclosure is provided. The control method includes: receiving data in a welding process; preprocessing the data in the welding process, extracting key features from the preprocessed data, performing weighted fusion processing on the key features to obtain comprehensive features, performing defect identification based on the comprehensive features and through a machine learning model to identify defect features; obtaining a repair strategy corresponding to the defect feature based on the defect feature and using a knowledge graph, wherein the knowledge graph is a knowledge graph containing welding parameters, defect features and repair strategies; generating repair instructions based on the repair strategy; and automatically adjusting welding parameters or controlling an actuator to perform a repair operation based on the repair instructions. In the control method, by collecting data in the welding process, and identifying defects after relevant processing of these data, a repair strategy corresponding to the defect is obtained through a knowledge graph, and a corresponding repair instruction is generated, and then based on the repair instruction, the welding parameters are automatically adjusted or the actuator is controlled to perform a repair operation, so that the automation and intelligence of repairing defects in the welding process are realized, and the welding quality and production efficiency are improved.

[0142] Moreover, in the above method, the association relationship in the knowledge graph can be used to quickly infer the best repair plan, realizing the combination of static knowledge and real-time data, so that the system can respond and make decisions quickly after detecting defects in real time, and then repair welding defects in time during the welding process, thereby improving the product yield.

[0143] In some embodiments, the control method further includes: dynamically scheduling the data acquisition frequency and data transmission rate of the sensor module.

[0144] In some embodiments, the above-mentioned control method also includes: receiving data during the repair process; judging whether the repair result meets the expected result based on the data during the repair process; if the repair result meets the expected result, recording the data during the repair process and updating the knowledge graph; if the repair result does not meet the expected result, obtaining a re-repair strategy different from the current repair strategy; generating a re-repair instruction based on the re-repair strategy; and re-performing the repair operation based on the re-repair instruction.

[0145] In some embodiments, the control method further includes: detecting a current working state of the sensor module; and adjusting working parameters of the sensor module according to historical data of the working state of the sensor module and the current working state.

[0146] Figure 8is a flow chart showing a control method for a welding process according to other embodiments of the present disclosure. This embodiment can be applied to real-time detection and automatic repair of robot welding quality. The method can be executed by a control system provided by an embodiment of the present disclosure. The control system can be implemented by software and / or hardware and can be configured in an electronic device. Figure 8 As shown, the control method includes steps S801 to S807.

[0147] In step S801, the device is initialized.

[0148] Start the system, load the welding process parameters and repair strategies in the knowledge graph through the control module, and set the acquisition frequency, accuracy requirements and data preprocessing parameters of each sensor in the sensor module. The operator can view the initialization status of the system through the user interface and make corresponding adjustments.

[0149] In step S802, data collection is performed.

[0150] The sensor module starts to collect visual data (i.e., image data), acoustic data, and infrared data in real time during the welding process. The time consistency of various data is ensured by synchronizing the clock system. The collected data is transmitted to the data analysis module through the control module.

[0151] In step S803, data fusion and analysis are performed.

[0152] The data analysis module performs preprocessing, feature extraction, and weighted fusion on the collected data to generate a comprehensive feature vector. Then, the system uses a machine learning algorithm (e.g., a defect recognition algorithm) to identify defects in the welding process and transmits the recognition results to the knowledge graph module. For example, after identifying a defect, the data analysis module transmits the comprehensive feature vector to the knowledge graph module.

[0153] In step S804, a repair strategy is obtained.

[0154] The knowledge graph module uses the reasoning mechanism to generate the optimal repair plan based on the detected defect type and location. The repair plan includes welding parameter adjustment or robot operation instructions. After the repair plan is generated, the control module issues an execution command.

[0155] In step S805, automatic repair is performed.

[0156] The repair module automatically adjusts the welding parameters or controls the welding robot to perform the repair operation according to the instructions of the control module. In addition, the repair results can be monitored in real time through sensors.

[0157] In step S806, feedback and adjustment are performed.

[0158] After the repair is completed, the system will test the weld quality again to determine whether the repair result (i.e., repair effect) meets the expected result (i.e., expected effect). If the repair result obtained by the test meets the expected result, the system will record the parameter data during the repair process and update the knowledge graph; if the repair result does not meet the expected result, the system will continue to adjust and perform a second repair.

[0159] For example, for the incomplete penetration defect, the system automatically provides a new process parameter for repair. If the provided process parameter can solve the defect problem, it is considered that the provided repair solution meets the expected results. Otherwise, it is considered that the provided repair solution does not meet the expected results.

[0160] When the repair result obtained by detection meets the expected result, the system can record the parameter data during the repair process and use this data to update the knowledge graph. The updated data may include at least one of: welding process parameters, defect types and characteristics, repair strategies, and process monitoring data.

[0161] For example, welding process parameters may include welding current, welding speed, temperature and other parameters.

[0162] For example, defect types and characteristics may include: defect types (eg, cracks, pores) and their characteristics detected and repaired during the welding process.

[0163] For example, a remediation strategy may include: documenting and optimizing the remediation steps and actions for specific defects.

[0164] For example, process monitoring data may include collecting real-time monitoring data related to the repair process. This process monitoring data can help improve future detection and repair accuracy.

[0165] In step S807, system maintenance and optimization are performed.

[0166] Each repair operation and test result will be recorded and transmitted to the knowledge graph module for analysis. This data will be used to update the knowledge base and optimize the repair strategy. In addition, the system will adjust the working parameters of each module based on historical data and current status to further improve the operating efficiency of the system.

[0167] In some embodiments, the system (eg, a control module) may adjust sensor parameters.

[0168] For example, if historical data indicates that specific welding process parameters (e.g., temperature or current) are repeatedly detected as problematic, the system can automatically adjust the sensitivity of the infrared sensor or the resolution of the vision sensor to more accurately monitor those parameters.

[0169] For another example, in the current state, if the system detects that the ambient light changes significantly, the exposure time of the visual sensor can be adjusted to improve the image quality.

[0170] In some embodiments, the system can optimize the data fusion weights. For example, if historical data shows that acoustic data contributes less to defect detection under certain conditions, the system can reduce its weight and increase the weight of visual and infrared data to improve the detection accuracy after fusion.

[0171] In some embodiments, the system can automatically adjust the repair parameters. For example, based on historical successful repair case data, the speed and path of the welding robot movement are adjusted to perform the repair work more efficiently.

[0172] In some embodiments, the system can perform strategy optimization. For example, if historical data shows that a certain defect often takes a long time to repair, the system can adjust the repair strategy and steps in advance to save time.

[0173] In some embodiments, the system can update the knowledge graph. For example, the system can update the knowledge graph based on the information of common defect types and their effective repair strategies collected from historical data to speed up reasoning and provide more accurate repair solutions.

[0174] In the above steps, the system can perform self-inspection and optimization regularly, continuously improve the knowledge graph by analyzing historical data and repair records, and improve the operating efficiency of each module.

[0175] So far, a control method for a welding process according to other embodiments of the present disclosure is provided. The method can solve the limitation problem of the method of the related art under complex welding environment and high precision requirements. In this method, by introducing multimodal data fusion and knowledge graph technology, higher detection accuracy, real-time performance and automatic repair capability can be achieved in the welding process, thereby improving the overall welding quality, reducing production costs and enhancing production efficiency.

[0176] In addition, during the automatic operation of the system, the operator can monitor the welding process and system status in real time through the user interface. In some cases, the operator can manually adjust some system parameters or intervene in the repair process to deal with special situations or improve the repair effect.

[0177] In the system and method of the embodiments of the present invention, by integrating data from multiple sensors such as vision, acoustics, and infrared, and adopting a weighted fusion algorithm, deep fusion of multi-dimensional information and high-precision defect detection in the welding process are achieved, surpassing the limitations of the single sensor method in related technologies.

[0178] Moreover, in the system and method of the disclosed embodiment, a knowledge graph including welding parameters, defect types and repair strategies is constructed, and the optimal repair plan is generated after the defects are detected in real time by using the reasoning mechanism. The dynamic update and self-learning ability of the knowledge graph enhances the adaptability and intelligence level of the system.

[0179] In addition, in the system and method of the embodiments of the present disclosure, a repair module has been developed that can automatically adjust welding parameters or control the robot to perform repair operations based on the detection results and the inferred repair plan, and ensure continuous optimization of welding quality through a closed-loop control mechanism, reducing human intervention.

[0180] Furthermore, in the system and method of the embodiments of the present disclosure, through the synchronization clock system and real-time processing algorithm, combined with machine learning technology, real-time monitoring of the welding process and high-precision defect identification are achieved, thereby improving the detection efficiency and accuracy of the system.

[0181] Fig. 9 Schematically shows a block diagram of an electronic device according to some embodiments of the present disclosure. The electronic device includes a memory 910 and a processor 920. Among them:

[0182] The memory 910 can be a disk, a flash memory, or any other non-volatile storage medium. The memory is used to store Figure 7 and / or Figure 8 The instructions in the corresponding embodiment.

[0183] The processor 920 is coupled to the memory 910 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 920 is used to execute instructions stored in the memory, realize the automation and intelligence of repairing defects in the welding process, and improve welding quality and production efficiency.

[0184] In some embodiments, it is also possible to Fig.10 As shown, the electronic device 1000 includes a memory 1010 and a processor 1020. The processor 1020 is coupled to the memory 1010 via a BUS 1030. The electronic device 1000 can also be connected to an external storage device 1050 via a storage interface 1040 to call external data, and can also be connected to a network or another computer system (not shown) via a network interface 1060, which will not be described in detail here.

[0185] In this embodiment, the memory stores data instructions, and the processor processes the instructions, thereby achieving automation and intelligence in repairing defects in the welding process, thereby improving welding quality and production efficiency.

[0186] In another embodiment, the present disclosure further provides a computer-readable storage medium (eg, a non-transitory computer-readable storage medium) having computer program instructions stored thereon, which are executed by a processor to implement Figure 7 and / or Figure 8 The steps of the method in the corresponding embodiment. Those skilled in the art should understand that the embodiments of the present disclosure can be provided as methods, devices, or computer program products. Therefore, the present disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure can take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0187] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0188] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0189] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0190] In some embodiments of the present disclosure, a computer program product is further provided. The computer program product includes a computer program or instructions. When the computer program or instructions are executed by a processor, the control method as described above is implemented.

[0191] In some embodiments of the present disclosure, a computer program is further provided, comprising: instructions, which, when executed by a processor, cause the processor to execute the control method as described above.

[0192] So far, the present disclosure has been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0193] Although some specific embodiments of the present disclosure have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A control system for a welding process, comprising: a sensor module configured to collect data during the welding process; A data analysis module is configured to preprocess the data in the welding process, extract key features from the preprocessed data, perform weighted fusion processing on the key features to obtain comprehensive features, and perform defect recognition based on the comprehensive features and through a machine learning model to obtain defect features; A knowledge graph module is configured to obtain a repair strategy corresponding to the defect feature according to the defect feature and using the knowledge graph, wherein the knowledge graph is a knowledge graph including welding parameters, defect features and repair strategies; a control module configured to transmit the data in the welding process received from the sensor module to the data analysis module, receive the repair strategy, and generate a repair instruction based on the repair strategy; and The repair module is configured to automatically adjust welding parameters or control the actuator to perform repair operations based on the repair instructions.

2. The control system according to claim 1, wherein: The knowledge graph is stored in a graph structure form, the nodes of the knowledge graph represent the welding parameters, the defect characteristics or the repair strategy, and the edges of the knowledge graph represent the association relationship between different nodes.

3. The control system according to claim 1 or 2, wherein: The sensor module comprises: a vision sensor configured to collect image data during the welding process; an acoustic sensor configured to collect acoustic wave signal data during the welding process; and An infrared sensor is configured to collect temperature distribution characteristic data during the welding process; The data in the welding process includes the image data, the acoustic wave signal data and the temperature distribution characteristic data.

4. The control system according to claim 3, wherein: The data analysis module is configured to perform denoising, normalization and time alignment processing on the image data, the acoustic wave signal data and the temperature distribution characteristic data.

5. The control system according to claim 4, wherein: The key features include: the key features of the image in the image data , the acoustic spectrum features in the acoustic wave signal data and the temperature gradient characteristics in the temperature distribution characteristic data ; Among them, the comprehensive characteristics for ， in, is the weight of the key features of the image, is the weight of the acoustic spectrum feature, is the weight of the temperature gradient feature.

6. The control system according to claim 1, wherein: The data analysis module is configured to perform defect identification based on the comprehensive features, in combination with the knowledge graph, and through the machine learning model.

7. The control system according to claim 1, wherein: The control module is also configured to dynamically schedule the data acquisition frequency and data transmission rate of the sensor module.

8. The control system of claim 1, wherein: The sensor module is also configured to collect data during the repair process; The data analysis module is further configured to determine whether a repair result meets an expected result based on the data in the repair process, and if the repair result meets the expected result, transmit the data in the repair process to the knowledge graph module; if the repair result does not meet the expected result, send re-repair information to the knowledge graph module; The knowledge graph module is further configured to record the data in the repair process and update the knowledge graph when the repair result meets the expected result, and to send a re-repair strategy different from the current repair strategy to the control module based on the re-repair information when the repair result does not meet the expected result; The control module is further configured to send a re-repair instruction to the repair module based on the re-repair strategy, so that the repair module performs the repair operation again.

9. The control system according to claim 1, wherein: The control module is further configured to detect a current working state of the sensor module, and adjust working parameters of the sensor module according to historical data of the working state of the sensor module and the current working state.

10. A control method for a welding process, comprising: Receive data during welding; Preprocessing the data in the welding process, extracting key features from the preprocessed data, performing weighted fusion processing on the key features to obtain comprehensive features, and performing defect recognition based on the comprehensive features and using a machine learning model to identify defect features; Obtaining a repair strategy corresponding to the defect feature by using a knowledge graph according to the defect feature, wherein the knowledge graph is a knowledge graph including welding parameters, defect features and repair strategies; generating a repair instruction based on the repair strategy; and Based on the repair instruction, the welding parameters are automatically adjusted or the actuator is controlled to perform the repair operation.

11. The control method according to claim 10, wherein: The knowledge graph is stored in a graph structure form, the nodes of the knowledge graph represent the welding parameters, the defect characteristics or the repair strategy, and the edges of the knowledge graph represent the association relationship between different nodes.

12. The control method according to claim 10 or 11, wherein: The data in the welding process includes image data, acoustic wave signal data and temperature distribution characteristic data.

13. The control method according to claim 12, wherein: Preprocessing the data in the welding process includes: The image data, the acoustic wave signal data and the temperature distribution characteristic data are subjected to denoising, normalization and time alignment processing.

14. The control method according to claim 13, wherein: The key features include: the key features of the image in the image data , the acoustic spectrum features in the acoustic wave signal data and the temperature gradient characteristics in the temperature distribution characteristic data ; Among them, the comprehensive characteristics for ， in, is the weight of the key features of the image, is the weight of the acoustic spectrum feature, is the weight of the temperature gradient feature.

15. The control method according to claim 10, wherein: Defect identification is performed based on the comprehensive features and through a machine learning model, including: Based on the comprehensive features, combined with the knowledge graph, and through the machine learning model, defect identification is performed.

16. The control method according to claim 10, further comprising: Dynamically schedule the data acquisition frequency and data transmission rate of the sensor module.

17. The control method according to claim 10, further comprising: Receive data from the repair process; Based on the data in the repair process, determining whether the repair result meets the expected result; When the repair result meets the expected result, record the data in the repair process and update the knowledge graph; When the repair result does not meet the expected result, obtaining a second repair strategy different from the current repair strategy; generating a re-repair instruction based on the re-repair strategy; and The repair operation is performed again based on the re-repair instruction.

18. The control method according to claim 10, further comprising: Detect the current working status of the sensor module; and The working parameters of the sensor module are adjusted according to the historical data of the working state of the sensor module and the current working state.

19. An electronic device comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the control method according to any one of claims 10 to 18 based on instructions stored in the memory.

20. A computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions, when executed by a processor, implement the control method according to any one of claims 10 to 18.

21. A computer program product, comprising a computer program or instructions, which implement the control method according to any one of claims 10 to 18 when executed by a processor.

Images