Robot welding gun current monitoring system of automobile part welding production line

Abstract

The invention discloses a robot welding gun current monitoring system for an automobile part welding production line. The robot welding gun current monitoring system comprises a data acquisition module, a model construction module, a real-time identification module and a data analysis module which are connected in sequence, the data acquisition module is used for inputting real-time welding data of a welding gun into the current anomaly model for current detection to obtain real-time current information and current current characteristics of welding; the model building module is used for building a current anomaly model according to each historical current anomaly feature; the real-time identification module is used for receiving the data transmitted by the data acquisition module in real time, calling a current anomaly model to obtain a welding standard value and judging whether the current characteristics of the real-time data are abnormal or not; and the data analysis module is used for restoring the standard current information of the welding gun according to the historical welding data of the welding gun and identifying current abnormal characteristics contained in the current information. A traditional manual detection mode is replaced, the labor cost is greatly reduced, the detection efficiency is improved, and the welding quality is guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of automobile parts welding production, and in particular to a robot welding gun current monitoring system for an automobile parts welding production line. Background Art

[0002] At present, welding technology is involved in many fields, especially in the production line of automobile parts. The main tool for welding is the welding gun. The welding gun is a tool used to supply welding current to the welding area during welding. During welding, a columnar electrode is used to form a welding spot between the contact surfaces of two overlapping workpieces. During spot welding, pressure is first applied to make the workpieces in close contact, and then the current is turned on. The contact of the workpieces melts under the action of resistance heat, and a welding spot is formed after cooling. Spot welding is mainly used for welding of sheet metal stamping parts with a thickness of less than 4mm, and is particularly suitable for welding of automobile bodies and carriages. The basic function of the welding gun is to provide a stable arc and pressure to ensure the quality of welding. In the production line of automobile parts, welding work is particularly important. In order to improve the production efficiency of the production line, many manufacturers will open multiple production lines at the same time. Although the work efficiency is improved, it requires double the labor to conduct quality verification during the production process, which not only increases the workload of labor, but also easily causes detection omissions. In serious cases, it will cause the suspension of the production line, affecting the production quality and reputation of the factory. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a robot welding gun current monitoring system for an automobile parts welding production line. By introducing machine learning technology, it aims to analyze the complex mapping relationship between the timing characteristics of the welding current (such as the current waveform) and the welding quality, so as to achieve high-precision anomaly detection and defect prediction, thereby replacing the traditional manual detection method, greatly reducing labor costs, improving detection efficiency, and ensuring welding quality.

[0004] The technical solution of the present invention is as follows:

[0005] A robot welding gun current monitoring system for an automobile parts welding production line comprises a data acquisition module, a model building module, a real-time recognition module and a data analysis module connected in sequence;

[0006] The data acquisition module is used to input the real-time welding data of the welding gun into the current anomaly model to perform current detection, thereby obtaining the real-time current information and current current characteristics of the welding;

[0007] The model building module is used to establish a current anomaly model according to each historical current anomaly feature;

[0008] The real-time identification module is used to receive the data transmitted in real time by the data acquisition module, and at the same time call the current abnormality model to obtain the standard value of welding, and determine whether the current characteristics of the real-time data are abnormal;

[0009] The data analysis module is used to restore the standard current information of the welding gun according to the historical welding data of the welding gun, and identify the abnormal current characteristics contained in the current information.

[0010] Furthermore, the data acquisition module includes:

[0011] The data acquisition hardware includes several controllers and detection devices. Each controller is connected to a detection device. Each detection device corresponds to the detection of a welding gun. Each welding gun is equipped with a data acquisition device.

[0012] The network unit includes a data acquisition computer and a data acquisition PLC controller. The data acquisition PLC controller is connected to the data acquisition computer. The data acquisition computer collects data and generates reports through an Ether Net network connection, and checks the current status of the current detection system in real time. The data acquisition PLC controller is also connected to the line body PLC, and performs real-time detection of the welding gun current and caches it through an Ether Net / IP communication interaction.

[0013] A real-time acquisition unit, used to collect real-time welding data generated by the welding gun when the welding gun is in working state;

[0014] A current detection unit is used to input the real-time welding data into the current anomaly model for synchronous detection, obtain the anomaly identification information contained in the real-time welding data, and determine the anomaly attribute corresponding to each anomaly identification information;

[0015] An information reorganization unit is used to obtain the current output information of the current abnormality model, integrate the abnormality identification information into the current output information for abnormality marking, and obtain the real-time current information of welding;

[0016] The feature generation unit is used to construct the current trend of the welding process and the current current of the welding gun based on the corresponding real-time current information at different moments in the welding process, and to construct the current current feature.

[0017] Furthermore, the data acquisition module also includes:

[0018] The data synchronization unit is used to obtain the data stream generated by the welding gun during each welding process, and generate and store the historical welding data of the welding gun after the data stream is updated;

[0019] The data sorting unit is used to obtain the corresponding real-time welding data at different times during the welding process, construct the welding data stream and transmit it to the data synchronization unit for storage.

[0020] Furthermore, the model building module includes:

[0021] An abnormality analysis unit, used to deduce the welding results of the welding gun at the corresponding welding moment based on the abnormal current characteristics, and to construct an abnormality-result correspondence list;

[0022] An abnormality training unit, used for learning the abnormality-result correspondence list by using big data reasoning to construct the current abnormality logic of the welding gun, and converting the current abnormality logic into a model program;

[0023] A framework training unit, used for determining several working parameters of the welding gun based on the machine parameters of the welding gun, and determining a parameter range corresponding to each working parameter, and generating a model framework;

[0024] The model building unit is used for inputting the model program into the model framework to adjust the function of the model framework and generate the current abnormality model of the welding gun.

[0025] Furthermore, the real-time recognition module includes:

[0026] An identification and positioning unit, used to respectively identify the actual welding position corresponding to each abnormal welding position in the welded object, and physically mark the actual welding position using a prescribed marking method;

[0027] An attribute recognition unit is used to collect an actual welding image corresponding to an actual welding position, recognize welding presentation information contained in the actual welding image, and search for abnormal attributes corresponding to the welding presentation information in big data;

[0028] A solution generation unit is used to find the processing method corresponding to the abnormal attribute in the big data and generate an abnormal compensation solution for the welded object in combination with the specification data of the actual welding position;

[0029] The early warning execution unit is used to execute the early warning work after physically marking the actual welding position.

[0030] Furthermore, the real-time recognition module also includes:

[0031] The warning execution unit determines that the current working state of the welding gun is unqualified and performs corresponding warning work when the number of abnormal welding positions corresponding to the current welding object is higher than a first preset number or the number of abnormal attributes corresponding to the current welding object is higher than a second preset number.

[0032] Furthermore, the data analysis module includes:

[0033] The data sampling unit is divided into hardware and software. The collector senses the current signal through the detection coil and converts it into a regular 0-10V voltage signal through the integration circuit. After the signal passes through the data processing system, it obtains the complete welding current data and transmits the data to the PLC module through the 485 interface. The PLC module removes the noise of the real-time data and packs the current, pressure and robot axis values ​​of the same welding point / welding pass into a data packet and stores it in the pre-storage space.

[0034] The real-time identification module is used to monitor in real time whether the PLC stores new data. Once the new data is stored, the data will be taken out immediately and input into the data analysis module. The model will be called to obtain the reasonable parameters to obtain the current current characteristics. If there is an abnormality, the abnormal information will be recorded.

[0035] A data analysis module is used to restore the standard current information of the welding gun based on the historical welding data of the welding gun and identify the abnormal current characteristics contained in the current current information;

[0036] The model building module is used to train the abnormal conditions stored in the real-time recognition module. All parameters recorded in the abnormal conditions, including current, pressure, welding time, and robot axis values, are put into the accumulated data model for retraining to obtain an updated version of the recognition model.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows: a robot welding gun current monitoring system for an automobile parts welding production line provided by the present invention collects and records the data of the robot welding gun when it is working, and then compares the collected data with the standard data output by the model to obtain the welding abnormality result corresponding to each current abnormality feature, thereby establishing a current abnormality model of the welding gun. During operation, the real-time welding data of the robot welding gun is input into the current abnormality model for current detection to obtain the real-time current information and the current current characteristics of this welding. When the current characteristics are abnormal, the welding results are derived according to the real-time current information, the abnormal welding position corresponding to the current current characteristics is located in the welding results, the abnormal attributes corresponding to the abnormal welding position are identified, an abnormal compensation plan is generated and an abnormal warning is issued, which can effectively prevent welding defects. The robot welding gun current monitoring system can monitor the output current of the line welding gun in real time, and the quality of the welds can be obtained without random destructive testing, thereby effectively reducing welding defects and reducing production line downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A system block diagram of a robot welding gun current monitoring system for an automobile parts welding production line provided by the present invention;

[0040] Figure 2 The present invention provides a schematic circuit diagram of a robot welding gun current monitoring system for an automobile parts welding production line. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0042] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0043] Example

[0044] See also Figure 1 , Figure 2 , this embodiment provides a robot welding gun current monitoring system for an automobile parts welding production line, including a data acquisition module, a model building module, a real-time recognition module and a data analysis module connected in sequence.

[0045] in:

[0046] (1) The data acquisition module is used to input the real-time welding data of the welding gun into the current anomaly model for current detection, thereby obtaining the real-time current information and current current characteristics of the welding.

[0047] The data acquisition module specifically includes:

[0048] The data acquisition hardware includes several controllers and detection devices. Each controller is connected to a detection device. Each detection device corresponds to the detection of a welding gun. Each welding gun is equipped with a data acquisition device.

[0049] The network unit includes a data acquisition computer and a data acquisition PLC controller. The data acquisition PLC controller is connected to the data acquisition computer. The data acquisition computer collects data and generates reports through an Ether Net network connection, and checks the current status of the current detection system in real time. The data acquisition PLC controller is also connected to the line body PLC, and performs real-time detection of the welding gun current and caches it through an Ether Net / IP communication interaction.

[0050] A real-time acquisition unit, used to collect real-time welding data generated by the welding gun when the welding gun is in working state;

[0051] A current detection unit is used to input the real-time welding data into the current anomaly model for synchronous detection, obtain the anomaly identification information contained in the real-time welding data, and determine the anomaly attribute corresponding to each anomaly identification information;

[0052] An information reorganization unit is used to obtain the current output information of the current abnormality model, integrate the abnormality identification information into the current output information for abnormality marking, and obtain the real-time current information of welding;

[0053] A feature generation unit, used to construct the current trend of the welding process and the current current of the welding gun based on the corresponding real-time current information at different moments in the welding process, and to construct the current current feature;

[0054] The data synchronization unit is used to obtain the data stream generated by the welding gun during each welding process, and generate and store the historical welding data of the welding gun after the data stream is updated;

[0055] The data sorting unit is used to obtain the corresponding real-time welding data at different times during the welding process, construct the welding data stream and transmit it to the data synchronization unit for storage.

[0056] (2) The model building module is used to establish a current anomaly model according to each historical current anomaly feature.

[0057] The model building blocks include:

[0058] An abnormality analysis unit, used to deduce the welding results of the welding gun at the corresponding welding moment based on the abnormal current characteristics, and to construct an abnormality-result correspondence list;

[0059] An abnormality training unit, used for learning the abnormality-result correspondence list by using big data reasoning to construct the current abnormality logic of the welding gun, and converting the current abnormality logic into a model program;

[0060] A framework training unit, used for determining several working parameters of the welding gun based on the machine parameters of the welding gun, and determining a parameter range corresponding to each working parameter, and generating a model framework;

[0061] The model building unit is used for inputting the model program into the model framework to adjust the function of the model framework and generate the current abnormality model of the welding gun.

[0062] (3) The real-time identification module is used to receive the data transmitted by the data acquisition module in real time, and at the same time call the current anomaly model to obtain the standard value of welding, and determine whether the current characteristics of the real-time data are abnormal.

[0063] The real-time recognition module includes:

[0064] An identification and positioning unit, used to respectively identify the actual welding position corresponding to each abnormal welding position in the welded object, and physically mark the actual welding position using a prescribed marking method;

[0065] An attribute recognition unit is used to collect an actual welding image corresponding to an actual welding position, recognize welding presentation information contained in the actual welding image, and search for abnormal attributes corresponding to the welding presentation information in big data;

[0066] A solution generation unit is used to find the processing method corresponding to the abnormal attribute in the big data and generate an abnormal compensation solution for the welded object in combination with the specification data of the actual welding position;

[0067] An early warning execution unit is used to execute early warning work after physically marking the actual welding position;

[0068] The warning execution unit determines that the current working state of the welding gun is unqualified and performs corresponding warning work when the number of abnormal welding positions corresponding to the current welding object is higher than a first preset number or the number of abnormal attributes corresponding to the current welding object is higher than a second preset number.

[0069] (4) The data analysis module is used to restore the standard current information of the welding gun based on the historical welding data of the welding gun and identify the abnormal current characteristics contained in the current information.

[0070] The data analysis module includes:

[0071] The data sampling unit is divided into hardware and software. The collector senses the current signal through the detection coil and converts it into a regular 0-10V voltage signal through the integration circuit. After the signal passes through the data processing system, it obtains the complete welding current data and transmits the data to the PLC module through the 485 interface. The PLC module removes the noise of the real-time data and packs the current, pressure and robot axis values ​​of the same welding point / welding pass into a data packet and stores it in the pre-storage space.

[0072] The real-time identification module is used to monitor in real time whether the PLC stores new data. Once the new data is stored, the data will be taken out immediately and input into the data analysis module. The model will be called to obtain the reasonable parameters to obtain the current current characteristics. If there is an abnormality, the abnormal information will be recorded.

[0073] A data analysis module is used to restore the standard current information of the welding gun based on the historical welding data of the welding gun and identify the abnormal current characteristics contained in the current current information;

[0074] The model building module is used to train the abnormal conditions stored in the real-time recognition module. All parameters recorded in the abnormal conditions, including current, pressure, welding time, and robot axis values, are put into the accumulated data model for retraining to obtain an updated version of the recognition model.

[0075] To sum up, the robot welding gun current monitoring system collects and records the data of the robot welding gun when it is working, and then compares the collected data with the standard data output by the model to obtain the welding abnormality results corresponding to each current abnormality feature, thereby establishing a current abnormality model of the welding gun. During operation, the real-time welding data of the robot welding gun is input into the current abnormality model for current detection to obtain the real-time current information and current current characteristics of this welding. When the current characteristics are abnormal, the welding results are deduced according to the real-time current information, and the abnormal welding position corresponding to the current current characteristics is located in the welding results. The abnormal attributes corresponding to the abnormal welding position are identified, and an abnormal compensation plan is generated and an abnormal warning is issued, which can effectively prevent welding defects. The robot welding gun current monitoring system can monitor the output current of the line welding gun in real time, and the quality of the welds can be obtained without random destructive testing, which effectively reduces welding defects and reduces production line downtime.

[0076] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A robot welding gun current monitoring system for an automobile parts welding production line, characterized in that: It includes a data acquisition module, a model building module, a real-time recognition module and a data analysis module which are connected in sequence; The data acquisition module is used to input the real-time welding data of the welding gun into the current anomaly model to perform current detection, thereby obtaining the real-time current information and current current characteristics of the welding; The model building module is used to establish a current anomaly model according to each historical current anomaly feature; The real-time identification module is used to receive the data transmitted in real time by the data acquisition module, and at the same time call the current abnormality model to obtain the standard value of welding, and determine whether the current characteristics of the real-time data are abnormal; The data analysis module is used to restore the standard current information of the welding gun based on the historical welding data of the welding gun and identify the abnormal current characteristics contained in the current information.

2. The robot welding gun current monitoring system for the automobile parts welding production line according to claim 1 is characterized in that: The data acquisition module comprises: The data acquisition hardware includes several controllers and detection devices. Each controller is connected to a detection device. Each detection device corresponds to the detection of a welding gun. Each welding gun is equipped with a data acquisition device. The network unit includes a data acquisition computer and a data acquisition PLC controller. The data acquisition PLC controller is connected to the data acquisition computer. The data acquisition computer collects data and generates reports through an Ether Net network connection, and checks the current status of the current detection system in real time. The data acquisition PLC controller is also connected to the line body PLC, and performs real-time detection of the welding gun current and caches it through an Ether Net / IP communication interaction. A real-time acquisition unit, used to collect real-time welding data generated by the welding gun when the welding gun is in working state; A current detection unit is used to input the real-time welding data into the current anomaly model for synchronous detection, obtain the anomaly identification information contained in the real-time welding data, and determine the anomaly attribute corresponding to each anomaly identification information; An information reorganization unit is used to obtain the current output information of the current abnormality model, integrate the abnormality identification information into the current output information for abnormality marking, and obtain the real-time current information of welding; The feature generation unit is used to construct the current trend of the welding process and the current current of the welding gun based on the corresponding real-time current information at different moments in the welding process, and to construct the current current feature.

3. The robot welding gun current monitoring system for the automobile parts welding production line according to claim 2 is characterized in that: The data acquisition module also includes: The data synchronization unit is used to obtain the data stream generated by the welding gun during each welding process, and generate and store the historical welding data of the welding gun after the data stream is updated; The data sorting unit is used to obtain the corresponding real-time welding data at different times during the welding process, construct the welding data stream and transmit it to the data synchronization unit for storage.

4. The robot welding gun current monitoring system for the automobile parts welding production line according to claim 1 is characterized in that: The model building module includes: An abnormality analysis unit, used to deduce the welding results of the welding gun at the corresponding welding moment based on the abnormal current characteristics, and to construct an abnormality-result correspondence list; An abnormality training unit, used for learning the abnormality-result correspondence list by using big data reasoning to construct the current abnormality logic of the welding gun, and converting the current abnormality logic into a model program; A framework training unit, used for determining several working parameters of the welding gun based on the machine parameters of the welding gun, and determining a parameter range corresponding to each working parameter, and generating a model framework; The model building unit is used for inputting the model program into the model framework to adjust the function of the model framework and generate the current abnormality model of the welding gun.

5. The robot welding gun current monitoring system for the automobile parts welding production line according to claim 1 is characterized in that: The real-time recognition module comprises: An identification and positioning unit, used to respectively identify the actual welding position corresponding to each abnormal welding position in the welded object, and physically mark the actual welding position using a prescribed marking method; An attribute recognition unit is used to collect an actual welding image corresponding to an actual welding position, recognize welding presentation information contained in the actual welding image, and search for abnormal attributes corresponding to the welding presentation information in big data; A solution generation unit is used to find the processing method corresponding to the abnormal attribute in the big data and generate an abnormal compensation solution for the welded object in combination with the specification data of the actual welding position; The early warning execution unit is used to execute the early warning work after physically marking the actual welding position.

6. The robot welding gun current monitoring system for the automobile parts welding production line according to claim 5 is characterized in that: The real-time recognition module also includes: The warning execution unit determines that the current working state of the welding gun is unqualified and performs corresponding warning work when the number of abnormal welding positions corresponding to the current welding object is higher than a first preset number or the number of abnormal attributes corresponding to the current welding object is higher than a second preset number.

7. The robot welding gun current monitoring system for the automobile parts welding production line according to claim 2 is characterized in that: The data analysis module includes: The data sampling unit is divided into hardware and software. The collector senses the current signal through the detection coil and converts it into a regular 0-10V voltage signal through the integration circuit. After the signal passes through the data processing system, it obtains the complete welding current data and transmits the data to the PLC module through the 485 interface. The PLC module removes the noise of the real-time data and packs the current, pressure and robot axis values ​​of the same welding point / welding pass into a data packet and stores it in the pre-storage space. The real-time identification module is used to monitor in real time whether the PLC stores new data. Once the new data is stored, the data will be taken out immediately and input into the data analysis module. The model will be called to obtain the reasonable parameters to obtain the current current characteristics. If there is an abnormality, the abnormal information will be recorded. A data analysis module is used to restore the standard current information of the welding gun based on the historical welding data of the welding gun and identify the abnormal current characteristics contained in the current current information; The model building module is used to train the abnormal conditions stored in the real-time recognition module. All parameters recorded in the abnormal conditions, including current, pressure, welding time, and robot axis values, are put into the accumulated data model for retraining to obtain an updated version of the recognition model.

Images