SYSTEM FOR DETECTING NEGATIVE TRENDS IN THE OCCURRING OF DEVIATE WELDING POINTS AND FOR AUTOMATIC ADJUSTING OF WELDING PARAMETERS

Abstract

System (10) for detecting negative trends in the occurrence of deviation weld spots (30) and for automatically adjusting welding parameters, the system (10) comprising: a system controller (12) communicating with a welding controller (14) of a production line (16) for vehicle components (18), the system controller (12) being designed to: monitor multiple weld spots (30) on each of multiple vehicle components (18) as the multiple vehicle components (18) move along the production line (16), with a welding monitoring system (28) communicating with the system controller (12); collect data relating to each of the multiple weld spots (30) on each of the multiple vehicle components (18) via the welding monitoring system (28) with sensors positioned on welding subsystems and associated with the welding monitoring system (28), as well as with external sensors associated with the welding monitoring system (28);Analyzing the data relating to each of the multiple weld points (30) on each of the multiple vehicle components (18) using multiple algorithms employing deep learning techniques; identifying negative trends associated with the occurrence of deviant weld points (30) for each of the multiple weld points (30); and automatically setting welding parameters for at least one of the multiple weld points (30) via communication with the welding controller (14);characterized in that the system controller (12), when collecting data relating to each of the multiple welds (30) on each of the multiple vehicle components (18) via the weld monitoring system (28), is further designed to collect data containing the weld location, the weld plan and its associated variables, a weld identifier (weld ID), vehicle information and sensor-based control data, as well as to collect data for the weld (30) based on the weld ID from a database (46); wherein, when analyzing the data relating to each of the multiple welds (30) on each of the multiple vehicle components (18), the system controller (12) is further designed to detect deviations in welds (30) and to assign a quality measure to each detected deviations in welds (30) and to compare the weld ID and associated weld parameters with computer-aided design (CAD) models;and wherein the system controller (12), when identifying negative trends associated with the occurrence of non-conforming welds (30), is designed for each of the multiple welds (30) to: identify the occurrence of a number of non-conforming welds (30) in a sample set that exceeds a predetermined threshold for any of the multiple welds (30); and identify a trend in which a number of non-conforming welds (30) in a sample set increases for any of the multiple welds (30); and identify a trend in which a quality measure of the identified non-conforming welds (30) deteriorates for any of the multiple welds (30).

Description

INTRODUCTION

[0001] The present invention relates generally to the detection of negative trends in the occurrence of deviating weld points in a welding process and the automatic adjustment of welding parameters to counteract detected negative trends.

[0002] Welding monitoring systems can individually assess the quality of body spot welds and detect non-standard, deviating welds, including so-called "edge welds." Weld quality is critical for the structural integrity of the vehicle, as well as its paint and corrosion resistance. There are several root causes of edge welds, and variations can occur between different batches of parts. Problems with current systems and procedures include the inability to predict deviating welds in advance. Current systems and procedures rely on visual inspection, and manual intervention slows down the process and involves additional work to adjust welding parameters. Furthermore, manual adjustment of welding parameters can introduce shifts that may lead to even higher defect rates.

[0003] Although current systems and procedures fulfill their intended purpose, there is a need for a new and improved system and procedure for automatically detecting negative trends and automatically adjusting welding parameters to avoid deviations in the welds. SUMMARY

[0004] According to several aspects of the present disclosure, a system for detecting negative trends in the occurrence of deviations in welds and for automatically adjusting welding parameters includes a system controller communicating with a welding controller of a vehicle component production line, wherein the system controller monitors multiple welds on each of multiple vehicle components as the multiple vehicle components move along the production line, communicates with a welding monitoring system for collecting data relating to each of the multiple welds on each of the multiple vehicle components via the welding monitoring system, analyzes the data relating to each of the multiple welds on each of the multiple vehicle components using multiple algorithms employing deep learning techniques, and identifies negative trends.which are assigned to the occurrence of deviating weld points, for each of the several weld points, and are designed for the automatic setting of welding parameters for at least one of the several weld points via communication with the welding controller.

[0005] According to another aspect, the system controller is further designed to collect data relating to each of the multiple weld points on each of the multiple vehicle components via the welding monitoring system, using sensors positioned on welding subsystems and associated with the welding monitoring system; and to collect data relating to each of the multiple weld points on each of the multiple vehicle components using external sensors associated with the welding monitoring system.

[0006] According to another aspect, when collecting data relating to each of the multiple weld points on each of the multiple vehicle components via the weld monitoring system, the system controller is further designed to collect data that includes, but is not limited to, the weld location, the weld plan and its associated variables, a weld point identifier (weld point ID), vehicle information and sensor-based control data.

[0007] According to another aspect, the welding monitoring system, when collecting data that includes, but is not limited to, the welding location, the welding plan and its associated variables, the welding point identifier (welding point ID), vehicle information and sensor-based control data, is further designed to communicate with and receive information from at least one human-machine interface (HMI) designed to allow manual data input by a human operator, a high-frequency identification (RFID) reader designed to read data from an RFID tag positioned on each of the vehicle components, and a visual camera designed to read data from a two-dimensional matrix positioned on each of the vehicle components.

[0008] According to another aspect, when collecting data relating to each of the multiple weld points on each of the multiple vehicle components via the weld monitoring system, the system controller is further designed to collect data for the weld point from a database based on the weld point ID.

[0009] According to another aspect, when analyzing the data relating to each of the multiple welds on each of the multiple vehicle components, the system controller is further designed to detect deviating welds and to assign a quality measure to each detected deviating weld and to compare the weld ID and associated welding parameters with computer-aided design (CAD) models.

[0010] According to another aspect, when identifying negative trends associated with the occurrence of non-conforming welds, the system controller is designed for each of the multiple welds to identify the occurrence of a number of non-conforming welds in a sample set that exceeds a predetermined threshold for any of the multiple welds, to identify a trend in which the number of non-conforming welds in a sample set increases for any of the multiple welds, and to identify a trend in which a quality measure of the identified non-conforming welds deteriorates for any of the multiple welds.

[0011] According to another aspect, when automatically setting welding parameters for at least one of the several welding points, the system controller is further designed to set welding parameters that include the location and orientation of a welding gun of a welding arrangement assigned to the at least one of the several welding points, a welding plan for the at least one of the several welding points, and internal logic algorithms in the welding controller, but this was not limited to.

[0012] According to another aspect, after automatic settings have been made to the welding parameters for at least one of the several welding points, the system controller is further designed to collect data relating to the at least one of the several welding points for which settings have been made via the welding monitoring system, to analyze the data relating to the at least one of the several welding points for which settings have been made using several algorithms employing deep learning techniques, and to determine whether the automatic settings have corrected the identified negative trends.

[0013] According to another aspect, if the system controller determines that the automatic settings have not corrected the identified negative trends, it is designed to repeatedly collect data relating to each of the multiple welds on each of the multiple vehicle components via the welding monitoring system until the identified negative trends have been corrected, to analyze the data relating to each of the multiple welds on each of the multiple vehicle components using the multiple algorithms that employ deep learning techniques, to identify negative trends for each of the multiple welds that are associated with the occurrence of deviant welds, and to automatically adjust welding parameters for at least one of the multiple welds via communication with the welding controller.

[0014] According to several aspects of the present disclosure, a method for detecting negative trends in the occurrence of deviation welds and for automatically adjusting welding parameters with a system controller in communication with the welding controller of a vehicle component production line includes monitoring multiple welds on each of multiple vehicle components as the multiple vehicle components move along the production line with a welding monitoring system in communication with the system controller, collecting data relating to each of the multiple welds on each of the multiple vehicle components via the welding monitoring system, analyzing the data relating to each of the multiple welds on each of the multiple vehicle components with multiple algorithms that use deep learning techniques, and identifying negative trends associated with the occurrence of deviation welds.for each of the multiple welding points and the automatic setting of welding parameters for at least one of the multiple welding points via communication with the welding controller.

[0015] According to another aspect, the collection of data relating to each of the multiple weld points on each of the multiple vehicle components via the welding monitoring system further includes the collection of data relating to each of the multiple weld points on each of the multiple vehicle components with sensors positioned on welding subsystems and associated with the welding monitoring system, and the collection of data relating to each of the multiple weld points on each of the multiple vehicle components with external sensors associated with the welding monitoring system.

[0016] According to another aspect, the collection of data relating to each of the multiple weld points on each of the multiple vehicle components via the weld monitoring system further includes the collection of data containing, but not limited to, the weld location, the weld plan and its associated variables, a weld point identifier (weld point ID), vehicle information, and sensor-based control data, by at least one of the communication and reception of data from a human-machine interface (HMI) designed to enable manual data input by a human operator, the communication and reception of data from a high-frequency identification (RFID) reader designed to read data from an RFID tag positioned on each of the vehicle components, and the communication and reception of data from a visual camera designed toTo read data from a two-dimensional matrix positioned at each of the vehicle components.

[0017] According to another aspect, the collection of data relating to each of the multiple weld points on each of the multiple vehicle components via the weld monitoring system further includes communicating and receiving data from a database designed to store information relating to the vehicle body components being welded and the multiple weld points to be applied to them, based on the weld point ID.

[0018] According to another aspect, analyzing the data with respect to each of the multiple welds on each of the multiple vehicle components further includes detecting outlier welds, assigning a quality measure for each detected outlier weld, and comparing the weld ID and associated weld parameters with computer-aided design (CAD) models.

[0019] According to another aspect, identifying negative trends associated with the occurrence of non-conforming welds involves, for each of the multiple welds, identifying the occurrence of a number of non-conforming welds in a sample set that exceeds a predetermined threshold for any of the multiple welds, identifying a trend in which the number of non-conforming welds in a sample set increases for any of the multiple welds, and identifying a trend in which a quality measure of the identified non-conforming welds deteriorates for any of the multiple welds.

[0020] According to another aspect, the automatic setting of welding parameters for at least one of the several welding points further includes the setting of welding parameters that specify the location and orientation of a welding gun of a welding arrangement assigned to the at least one of the several welding points, a welding plan for the at least one of the several welding points, and internal logic algorithms in the welding controller, and was not limited to this.

[0021] According to another aspect, the procedure after automatically setting welding parameters for at least one of the several welding points further includes collecting data relating to the at least one of the several welding points for which settings were made via the welding monitoring system, analyzing the data relating to the at least one of the several welding points for which settings were made using several algorithms employing deep learning techniques, and determining whether the automatic settings have corrected the identified negative trends.

[0022] According to another aspect, the procedure further includes, if the system controller determines that the automatic settings have not corrected the identified negative trends, collecting data relating to each of the multiple welds on each of the multiple vehicle components via the welding monitoring system, analyzing the data relating to each of the multiple welds on each of the multiple vehicle components using the multiple algorithms employing deep learning techniques, identifying negative trends associated with the occurrence of deviant welds for each of the multiple welds, and automatically adjusting welding parameters for at least one of the multiple welds via communication with the welding controller, repeatedly with the system controller, until identified negative trends have been corrected.

[0023] Further areas of applicability are evident from the description given here. It should be understood that the description and the specific examples are for illustrative purposes only and are not intended to limit the scope of protection afforded by this disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described here serve only for illustration and are not intended to limit the scope of protection of the present disclosure in any way; they show: Fig. 1 a schematic view of a production line with a system according to an exemplary embodiment of the present disclosure; Fig. 2 a schematic representation of the system according to an exemplary embodiment; Fig. 3. A perspective view of a scanned vehicle component with two deviating weld points and one non-deviating weld point on it; and Fig. 4 a schematic flow chart representing a procedure according to an exemplary embodiment of the present disclosure.

[0025] The figures are not necessarily to scale, and some features may be exaggerated or reduced, for example, to show details of certain components. In some cases, well-known components, systems, materials, or processes are not described in detail to avoid obscuring the present disclosure. Thus, specific structural and functional details disclosed herein are not to be interpreted as limiting, but merely as a basis for the claims and as a representative basis for teaching the person skilled in the art the various uses of the present disclosure. DETAILED DESCRIPTION

[0026] The following description is by its very nature merely exemplary and is not intended to limit the present disclosure, application, or uses. Furthermore, it is not intended to be limited by any explicit or implicit theory set forth in the preceding technical field, background, summary, or detailed description. It should be understood that corresponding reference numerals throughout the drawings denote identical or corresponding parts and features.As used here, the term "module" refers to any hardware, software, firmware, electronic control component, processing logic, and / or processor device, individually or in any combination, including, without limitation: an application-specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or grouped) and memory executing one or more software or firmware programs, a combination logic circuit, and / or other suitable components providing the described functionality. Although the figures shown here depict an example with specific arrangements of elements, actual embodiments may include additional intervening elements, devices, features, or components. It should be understood that the figures are for illustrative purposes only and may not be to scale.

[0027] The term "vehicle" as used here is not limited to motor vehicles. Although the technology presented here is primarily described in connection with motor vehicles, it is not limited to motor vehicles. The concepts can be used in a wide variety of applications, such as with aircraft, watercraft, other vehicles, and consumer electronics components.

[0028] To ensure that this disclosure is thorough and fully conveys the scope of protection to the person skilled in the art, exemplary embodiments are given. To guarantee a thorough understanding of embodiments of the present disclosure, numerous specific details are presented, such as examples of specific compositions, components, devices, and methods. The person skilled in the art will appreciate that specific details need not be used, that exemplary embodiments can be embodied in many different forms, and that none should be understood as limiting the scope of protection of the disclosure. According to some exemplary embodiments, well-known processes, well-known device structures, and well-known technologies are not described in detail.

[0029] The terminology used here serves only to describe certain exemplary embodiments and is not intended to be limiting. As used here, the singular forms "a," "an," and "the" may be intended to also include the plural forms unless the context unambiguously specifies otherwise. The terms "comprises," "comprehensive," "containing," and "including" are inclusive and thus specify the presence of the mentioned features, elements, compositions, steps, integers, operations, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.Although the open term "comprehensive" is to be understood as a non-restrictive term used to describe and claim various embodiments set forth herein, according to certain aspects the term may alternatively be understood as a more restrictive and limiting term, such as "consisting of" or "essentially consisting of". Thus, for any given embodiment that specifies compositions, materials, components, elements, features, integers, operations and / or process steps, the present disclosure also specifically includes embodiments that consist of or essentially consist of such specified compositions, materials, components, elements, features, integers, operations and / or process steps.In the case of "consisting of", the alternative embodiment excludes any additional compositions, materials, components, elements, features, integers, operations and / or process steps, whereas in the case of "essentially consisting of", any additional compositions, materials, components, elements, features, integers, operations and / or process steps that materially affect the basic and new properties are excluded from such an embodiment, as are any compositions, materials, components, elements, features, integers, operations and / or process steps that do not materially affect the basic and new properties but may be included in the embodiment.

[0030] Unless specifically identified as a sequence of execution, any procedural steps, processes, and operations described herein should not be interpreted as necessarily requiring their execution in the particular order discussed or illustrated. Furthermore, it should be understood that additional or alternative steps may be used unless otherwise indicated.

[0031] When a component, element, or layer is described as being "at," "interacting with," "connected with," or "coupled with" another element or layer, it may be directly at, interacting with, connected with, or coupled to that other component, element, or layer, or there may be intermediate elements or layers. Conversely, when an element is described as being "directly at," "directly interacting with," "directly connected with," or "directly coupled with" another element or layer, there may be no intermediate elements or layers. Other words used to describe the relationship between elements (e.g., "between" versus "directly between," "adjacent" versus "directly adjacent," etc.) are to be interpreted similarly.As the term “and / or” is used here, it includes any combination of one or more of the associated listed items.

[0032] Although the terms first, second, third, etc., may be used here to describe various steps, elements, components, areas, layers, and / or sections, these steps, elements, components, areas, layers, and / or sections are not to be limited by these terms unless otherwise indicated. These terms may only be used to distinguish one step, element, component, area, layer, or section from another. Unless clearly indicated by the context, terms such as "first," "second," and other numerical words, when used here, do not imply any sequence or order.Thus, a first step, a first element, a first component, a first area, a first layer or a first section discussed below could be referred to as a second step, a second element, a second component, a second area, a second layer or a second section without deviating from the teachings of the exemplary embodiments.

[0033] Spatial or temporal relative terms such as "before," "after," "inner," "outer," "below," "under," "lower," "above," "upper," and the like may be used here to facilitate the description of the relationship of one element or feature to other element(s) or feature(s) as depicted in the figures. Spatial or temporal relative terms may also be intended to encompass orientations of the device or system in use or operation other than the orientation shown in the figures.

[0034] Throughout this disclosure, numerical values ​​represent approximate measures or limits for ranges that include small deviations from the given values ​​and embodiments approximately equal to the stated value, as well as those exactly equal to the stated value. Apart from the working examples given at the end of the detailed description, all numerical values ​​of parameters (e.g., of quantities or conditions) in this application, including the appended claims, are to be understood as modified by the term "approximately" in all cases, regardless of whether "approximately" actually precedes the numerical value. "Approximately" indicates that the stated numerical value permits a slight inaccuracy (with a certain approximation to the exactness of the value; approximately or reasonably close to the value; nearly).If the imprecision indicated by "approximately" cannot otherwise be understood in the field with this normal meaning, "approximately," as used here, at least indicates variations that may arise from normal procedures for measuring and using such parameters. For example, with respect to percentages, "approximately" includes a range of ±5%; with respect to temperatures, "approximately" includes a range of ±5 degrees; and with respect to distances, "approximately" includes ±10%. Furthermore, the disclosure of ranges includes the disclosure of all values ​​and further subdivided ranges within the overall range, including endpoints and subranges given for the ranges.

[0035] Exemplary embodiments are now described in more detail with reference to the accompanying drawings. Based on Fig. 1 and Fig. 2 includes a system 10 for detecting negative trends in the occurrence of deviating weld points and automatically adjusting welding parameters according to an exemplary embodiment of the present disclosure, a system controller 12 in communication with a welding controller 14 of a production line 16 for vehicle components 18.

[0036] According to a non-restrictive example, production line 16, as in Fig. 1 and Fig. As shown in Figure 2, the system is designed to apply welds to vehicle components 18, wherein the vehicle components 18 are vehicle body structures. The production line 16 transports several vehicle components 18 sequentially to the vicinity of several welding arrangements 20a-20n, each welding arrangement 20a-20n comprising a robot articulated arm 22 carrying a welding gun 24. As shown, the production line 16 transports the vehicle components 18, as indicated by arrow 26, to a first welding arrangement 20a, a second welding arrangement 20b, a third welding arrangement 20c, and a fourth welding arrangement 20d. A person skilled in the art should understand that the system 10 can include any number of welding arrangements 20a-20n without deviating from the scope of protection of this disclosure.Each welding arrangement 20a-20n is designed to apply at least one weld to each vehicle component 18, according to the control of the system controller 12 and the welding controller 14. Depending on the cycle time and location, any of the multiple welding arrangements 20a-20n can be designed to apply multiple welds to each vehicle component 18.

[0037] The System Controller 12 and the Welding Controller 14 are non-generalised electronic control devices with a pre-programmed digital computer or digital processor, memory or non-transient computer-readable medium used to store data such as control logic, software applications, instructions, computer code, data, lookup tables, etc., and a transceiver [or input / output ports]. Computer-readable medium includes any type of medium accessible by a computer, such as read-only memory (ROM), read / write memory (RAM), a hard disk drive, a compact disc (CD), a digital video disc (DVD), or any other type of storage. "Non-transient" computer-readable medium excludes wired, wireless, optical, or other communication links carrying transient electrical or other signals.A non-transient, computer-readable medium includes media in which data can be permanently stored and media in which data can be stored and later overwritten, such as a rewritable optical disc or an erasable storage device. Computer code includes any type of program code, including source code, object code, and executable code.

[0038] The system controller 12 communicates with a welding monitoring system 28, whereby the system controller 12, using the welding monitoring system 28, monitors several welding points 30 on each of the several vehicle components 18, while the several vehicle components 18 move along the production line 16.

[0039] The welding monitoring system 28 can be a type commonly known and used in the industry. Non-contact and non-destructive automated welding inspection systems improve quality control processes by eliminating manual welding inspections, which are prone to error and operator subjectivity. The welding monitoring system compares weld characteristics against a predefined quality standard based on the intended use and / or industry. According to one exemplary embodiment, real-time welding dimension data generated by high-resolution visual 3D sensors is analyzed for defects by software incorporating algorithms and deep learning techniques.Components of the welding monitoring system 28 can be integrated into the welding setups 20a-20n to monitor the welding process in real time, and / or the welding monitoring system 28 can operate after the welding of the vehicle component 18 is complete. In some cases, sensors and cameras of the welding monitoring system 28 can be positioned on the robot articulated arms 22 of the welding setups 20a-20n, allowing the robot articulated arms 22 to be used for maneuvering and positioning such sensors and cameras. The welding monitoring system 28 can inspect the geometry of each weld 30, including the weld length, weld shape, weld volume, and weld orientation. Furthermore, the welding monitoring system 28 can detect defects that may occur in the weld 30, such as spatter, burn-through, undercutting, porosity, and edge deviations.

[0040] The system controller 12 is designed to collect data relating to each of the multiple weld points 30 on each of the multiple vehicle components 18 via the welding monitoring system 28. According to an exemplary embodiment, when the system controller 12 collects data relating to each of the multiple weld points 30 on each of the multiple vehicle components 10 via the welding monitoring system 28, it is further designed to collect data relating to each of the multiple weld points 30 on each of the multiple vehicle components 18 using sensors 32, which are positioned on welding subsystems such as the robot articulated arms 22 and the welding guns 24 of the welding arrangements 20a-20n and which are associated with the welding monitoring system 28, and furthermore to collect data relating to each of the multiple weld points 30 on each of the multiple vehicle components 18 using external sensors 34, which are associated with the welding monitoring system 28.The data includes, but is not limited to, the welding location, the welding plan and its associated variables, a welding point identifier (welding point ID), vehicle information and sensor-based control data.

[0041] The sensors 32 positioned on the welding subsystems and the external sensors 34 can be cameras designed to visually "look" at the weld points 30 during and after the weld points 30 have been formed. The sensors 32 positioned on the welding subsystems can include sensors attached to the welding gun 24, designed to monitor the operating behavior of the welding gun 24 (voltage, temperature, etc.) as a measure of the quality of the weld points 30. As in Fig. As shown in Figure 1, a sensor 32A is positioned on the robot articulated arm 22 of the second welding assembly 20b, and is designed to monitor the weld point formed by the welding gun 24 attached to the robot articulated arm 22 of the second welding assembly 20b. Furthermore, a sensor 32B is positioned on the welding gun 24 of the first welding assembly 20a. External sensors 34 are mounted near the welding assemblies 20a-20n, and are capable of collecting data and monitoring the weld points 30 formed by the multiple welding assemblies 20a-20n.

[0042] According to an exemplary embodiment, when collecting data relating to the weld points 30, the system controller 12 is further configured, via the welding monitoring system 28, to communicate with and receive information from at least one human-machine interface (HMI) 36, designed to allow manual data input by a human operator and to enable information to be shared between the system controller 12 and an operator of the system 10; one high-frequency identification (RFID) reader 38, designed to read data from an RFID tag 40 positioned on each of the vehicle components 18; and one visual camera 42, designed to read data from a two-dimensional matrix 44 positioned on each of the vehicle components 18. As in Fig. As shown in Figure 1, a general camera attached to the first welding arrangement 20a can be either an RFID reader 38 or a visual camera 42, and a label attached to the vehicle component 18 can be either an RFID label 40 or a two-dimensional matrix 44. A person skilled in the art should understand that the system 10 can receive data relating to the welds 30, such as the weld ID or specific parameters assigned to the welds 30, from any or any combination of the HMI 36, the RFID reader 38, or the visual camera 42.

[0043] According to an exemplary embodiment, the system controller 12 retrieves data for weld 30 from a database 46 via the welding monitoring system 14, based on the weld ID. For example, if the system controller 12 knows the weld ID for weld 30, as obtained from the HMI 36, the RFID tag 40, or the two-dimensional matrix 44 attached to the vehicle component 18, it can access the database 46, which contains detailed data such as welding parameters, dimensional and location parameters, and other data associated with the weld ID. Thus, using the weld ID, the system controller 12 can access much more data relating to each of the multiple welds 30 on each of the vehicle components 18. Each weld 30 on each of the vehicle components 18 has a unique weld ID.Specifications and parameters for each weld point 30 are based on requirements for the operational behavior of the weld point 30. The database 46 can be located in the system controller 12 or can be a remote database that communicates directly or wirelessly with the system controller 12 or with the welding monitoring system 28.

[0044] The system controller 12 is designed to analyze, using data collected or received by the welding monitoring system 28, the data relating to each of the multiple weld points 30 on each of the multiple vehicle components 18, using several algorithms 48 stored within it that employ deep learning techniques. Furthermore, the system controller 12 is designed to detect deviations in the weld points 30D and to assign a quality measure to each detected deviation in the weld point 30D, and to compare the weld point ID and associated welding parameters for each weld point 30 with computer-aided design (CAD) models stored in the system controller 12 or in the database 46.

[0045] Based on Fig. 3 is an edge weld, a weld formed at or near an edge of the vehicle component 18. An edge weld 30 has welding parameters such as a centerline distance 50 from the edge of component 18 and a target distance 52 from the edge of component 18 to ensure the correct quality and structural integrity of the weld 30. A non-standard weld 30D is a weld that does not meet these parameters. As in Fig. As shown in Figure 3, for example, the acceptable edge weld 30 has a correct location such that the centerline distance 50 and the target distance 52 satisfy the parameters for the edge weld 30. The values ​​for the centerline distance 50 and the target distance 52, and the relationship between them, depend on many factors, including, but not limited to, the size of the electrode of the welding gun 24, the material from which the vehicle component 18 is manufactured, and structural strength requirements for the vehicle component 18.

[0046] A non-compliant edge weld 30D is a weld that does not meet one or more of the welding parameters. For example, the upper weld 30D-1, again based on... Fig. Figure 3 is an example of a deviation edge weld with bulging, where the weld has actually bulged out the edge of the vehicle component 18. The middle weld 30D-2 is an example of a deviation edge weld where the edge is bulged out because weld 30D-2 is not correctly spaced from it.

[0047] Furthermore, the system controller 12 is designed to assign a quality measure for each detected non-conforming weld 30D. By evaluating the extent to which an identified non-conforming weld does not meet the welding parameters for that weld 30, and by comparing the non-conforming weld with CAD models of a correct weld 30 and modeling the sensors 32, 34, the system controller 12 establishes a quality measure for each identified non-conforming weld.

[0048] Furthermore, the system controller 12 is designed to identify negative trends associated with the occurrence of non-conforming welds for each of the multiple welds 30 by identifying the occurrence of a number of non-conforming welds in a sample set that exceeds a predefined threshold. For example, during a production run, sample sets of ten vehicle components are periodically taken, evaluating each of four welds 301, 302, 303, 304. As an illustrative example, the predefined threshold is three, and the system continues without adjustment as long as three or fewer non-conforming welds are identified in a sample set of ten vehicle components.

[0049] During the first set of samples, the system controller 12 identifies two out of ten vehicle components where a first weld 301 is non-compliant, zero out of ten vehicle components where a second weld 302 is non-compliant, one out of ten vehicle components where a third weld 303 is non-compliant, and one out of ten vehicle components where a fourth weld 304 is non-compliant. After the first set of samples, the system controller does not take any corrective action.

[0050] Later during the production run, during a second set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld 301 is defective, zero out of ten vehicle components where the second weld 302 is defective, four out of ten vehicle components where the third weld 303 is defective, and one out of ten vehicle components where the fourth weld 304 is defective. The number of vehicle components in the sample set of ten where the third weld 303 was identified as defective exceeds the predefined threshold of three, so the system controller 12 automatically adjusts the welding parameters for the third weld 303 via communication with the welding controller 14.

[0051] Furthermore, the system controller 12 is designed to identify negative trends associated with the occurrence of non-conforming welds for each of the multiple weld locations 30 by identifying a trend in which the number of non-conforming weld locations in a sample set increases for any of the multiple weld locations 301, 302, 303, 304. For example, during a production series, sample sets from ten vehicle components are taken periodically, evaluating each of the four weld locations 301, 302, 303, 304. The predefined threshold is three non-conforming weld locations.

[0052] During a first set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld 301 is non-standard, zero out of ten vehicle components where the second weld 302 is non-standard, one out of ten vehicle components where the third weld 303 is non-standard, and one out of ten vehicle components where the fourth weld 304 is non-standard.

[0053] Later, during a second set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld 301 is non-standard, one out of ten vehicle components where the second weld 302 is non-standard, one out of ten vehicle components where the third weld 303 is non-standard, and one out of ten vehicle components where the fourth weld 304 is non-standard.

[0054] Later, during a third set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld 301 is non-standard, two out of ten vehicle components where the second weld 302 is non-standard, one out of ten vehicle components where the third weld 303 is non-standard, and one out of ten vehicle components where the fourth weld 304 is non-standard.

[0055] Finally, during a fourth set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld 301 is non-standard, three out of ten vehicle components where the second weld 302 is non-standard, one out of ten vehicle components where the third weld 303 is non-standard, and one out of ten vehicle components where the fourth weld 304 is non-standard.

[0056] Although none of the welds 301, 302, 303, 304 showed a number of deviations exceeding the predefined threshold, the second weld 302 showed a consistent trend of increasing numbers of deviations (increasing by one in each successive set of samples). This indicates a trend that the system controller 12 identifies and uses to predict an increase in deviations for the second weld 302, enabling it to preemptively and automatically adjust welding parameters for the second weld 302 via communication with the welding controller 14.

[0057] Furthermore, the system controller 12 is designed to identify negative trends associated with the occurrence of non-conforming welds for each of the multiple weld points 30 by identifying a trend in which a quality measure of any of the multiple weld points 301, 302, 303, 304 deteriorates. For example, during a production series, sample sets of ten vehicle components are periodically taken, with each of the four weld points 301, 302, 303, 304 being evaluated. The predefined threshold is three non-conforming weld points.

[0058] During an initial set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld (301) is non-compliant, zero out of ten vehicle components where the second weld (302) is non-compliant, one out of ten vehicle components where the third weld (303) is non-compliant, and one out of ten vehicle components where the fourth weld (304) is non-compliant. The quality measure assigned to each of the non-compliant welds (301, 303, and 304) indicates that the welds are only slightly outside the conformity with welding parameters.

[0059] Later, during a second set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld 301 is non-compliant, one out of ten vehicle components where the second weld 302 is non-compliant, two out of ten vehicle components where the third weld 303 is non-compliant, and two out of ten vehicle components where the fourth weld 304 is non-compliant. The quality measure assigned to each of the non-compliant welds—for the first, third, and fourth welds 301, 303, and 304, respectively—indicates that the welds are still slightly out of compliance with welding parameters. However, the quality measure assigned to the non-compliant weld 302 indicates that the non-compliant second weld 302 was substantially out of compliance with welding parameters.

[0060] Later, during a third set of samples, the system controller 12 identifies two out of ten vehicle components where the first weld (301) is non-compliant, two out of ten vehicle components where the second weld (302) is non-compliant, one out of ten vehicle components where the third weld (303) is non-compliant, and one out of ten vehicle components where the fourth weld (304) is non-compliant. The quality measure assigned to each of the non-compliant welds (301, 303, and 304) indicates that the welds are still only slightly out of compliance with welding parameters, while the quality measure assigned to the non-compliant welds (302) indicates that the non-compliant second welds (302) are drastically out of compliance with welding parameters.

[0061] The deteriorating quality measure of the deviating welds for the second weld 302 indicates that the quality of the second weld 302 is deteriorating while the specified threshold is not exceeded, which can be an early indicator that progressively more of the second weld 302 are becoming deviant, enabling the system controller 12 to preemptively and automatically adjust welding parameters for the second weld 302 via communication with the welding controller 14.

[0062] Thus, the system controller 12 can automatically make adjustments by identifying negative trends in the welding parameters, including, but not limited to, the location and orientation of the welding gun 24 of the welding arrangement 20a-20n, which is assigned to at least one of the several welding points 301, 302, 303, 304 for which the trend has been identified, a welding plan for at least one of the several welding points 301, 302, 303, 304 for which the trend has been identified, and internal logic algorithms in the welding controller 14, which control the movements of the robot articulated arms 22 and welding guns 24 of the welding arrangements 20a-20n.

[0063] The system controller 12 determines and calculates parameter settings based on the analysis of the collected data, such as the trends in quality measurements, and identifies a suitable weld location, taking into account the electrode size of the welding gun 24 and the target weld location relative to the edge of the vehicle component 18. For example, with an electrode diameter of eight millimeters (mm) and a target weld location five mm from the edge of the vehicle component 18, the welding monitoring system 28 either provides a precise weld center location using a visual camera or provides a classification window using sensor 32 and 34 modeling. Thus, the minimum setting must be 8 mm / 2 + 5 mm = 9 mm.

[0064] The system controller 12 makes individual and independent adjustments to the welding parameters for each welding arrangement 20a-20n, sets the placement of the welding point 30, the power for the welding gun 24, the orientation of the welding gun 24 to the vehicle component 18 during welding, the overall welding plan (the sequence of welding, which welding arrangement 20a-20n applies which of the multiple welding points 301, 301, 303, 304, etc.) in order to adjust each of the multiple resulting welding points 301, 302, 303, 304, comparing each of the multiple welding points 301, 302, 303, 304 with CAD models that are either stored in the system controller 12, stored in the database 46, or accessed wirelessly from a remote location, in order to increase the overall quality of the welding process for the vehicle component 18.

[0065] After these settings have been made, the system controller is designed to verify that the settings have actually improved the quality of the multiple welds 30. The system controller 12 is designed, after automatic adjustments have been made to the welding parameters for at least one of the multiple welds 30, to collect data via the welding monitoring system 28 relating to the at least one of the multiple welds 301, 302, 303, 304 for which settings were made, to analyze the data relating to the at least one of the multiple welds 301, 302, 303, 304 for which settings were made using the multiple algorithms 48 that employ deep learning techniques, and to determine whether the automatic adjustments have corrected the identified negative trends.

[0066] As described above, the system controller considers successive sets of samples to determine whether the negative trends originally identified by the system controller 12 have been corrected, improved, or whether the settings had no effect at all.

[0067] If the system controller 12 determines that the automatic settings have not corrected the identified negative trends, the system controller 12 is designed to repeatedly collect data via the welding monitoring system 28 relating to each of the multiple weld points 301, 302, 303, 304 on each of the multiple vehicle components 18, to analyze the data relating to each of the multiple weld points 301, 302, 303, 304 on each of the multiple vehicle components 18 using the multiple algorithms 48 that employ deep learning techniques, to identify negative trends associated with the occurrence of deviant weld points for each of the multiple weld points 301, 302, 303, 304, and to automatically adjust welding parameters for at least one of the multiple weld points 301, 302, 303, 304 via communication with the welding controller 14 until the identified negative trends have been corrected.

[0068] Based on Fig.4 contains a method 200 for detecting the occurrence of negative trends in a deviating weld and for automatically adjusting welding parameters, starting in block 202 with a system controller 12 in communication with a welding controller 14 of a production line 16 for vehicle components 18, transitioning to block 204 for monitoring multiple welds 30 on each of the multiple vehicle components 18 while the multiple vehicle components 18 move along the production line 16, with a welding monitoring system 28 in communication with the system controller 12, transitioning to block 206 for collecting data relating to each of the multiple welds 30 on each of the multiple vehicle components 18 via the welding monitoring system 28, transitioning to block 208 for analyzing the data relating to each of the multiple welds 30 on each of the multiple vehicle components 18 with multiple algorithms 48.using deep learning techniques, transitioning to block 210, identifying negative trends associated with the occurrence of deviating weld spots for each of the multiple weld spots 30, and transitioning to block 212, automatically adjusting welding parameters of at least one of the multiple weld spots 30 via communication with the welding controller 14.

[0069] According to an exemplary embodiment, the acquisition of data relating to each of the multiple weld points 30 on each of the multiple vehicle components 18 in block 206 via the welding monitoring system 28 further comprises the acquisition of data relating to each of the multiple weld points 30 on each of the multiple vehicle components 18 with sensors 32 that are positioned on welding subsystems and are associated with the welding monitoring system 28, and the acquisition of data relating to each of the multiple weld points 30 on each of the multiple vehicle components 18 with external sensors 34 that are associated with the welding monitoring system 28.

[0070] According to another exemplary embodiment, the acquisition of data relating to each of the multiple weld spots 30 on each of the multiple vehicle components 18 via the weld monitoring system 28 in block 206 further comprises the acquisition of data including, but not limited to, the weld location, the weld plan and its associated variables, a weld spot identifier (weld spot ID), vehicle information and sensor-based control data, by at least one of the transmission and reception of data from a human-machine interface (HMI) 36 designed to enable manual data input by a human operator, the transmission and reception of data from a high-frequency identification (RFID) reader 38 designed to read data from an RFID tag 40 positioned on each of the vehicle components 18, and the transmission and reception of data from a visual camera 42 designed toTo read data from a two-dimensional matrix 44, which is positioned at each of the vehicle components 18.

[0071] According to another exemplary embodiment, the acquisition of data relating to each of the multiple weld spots 30 on each of the multiple vehicle components 18 via the weld monitoring system 28 in block 206 further includes the transmission and receipt of data from a database 46 designed to store information relating to the vehicle body components 18 being welded and the multiple weld spots 30 applied thereto, based on the weld spot ID.

[0072] According to another exemplary embodiment, analyzing the data relating to each of the multiple weld spots 30 on each of the multiple vehicle components 18 in block 208 further includes detecting outlier weld spots, assigning a quality measure for each detected outlier weld spot, and comparing the weld spot ID and associated welding parameters with computer-aided design (CAD) models.

[0073] According to another exemplary embodiment, identifying negative trends associated with the occurrence of non-conforming welds for each of the multiple welds 30 in block 210 further includes identifying the occurrence of a number of non-conforming welds in a sample set that exceeds a predetermined threshold for any of the multiple welds 30, identifying a trend in which a number of non-conforming welds in a sample set increases for any of the multiple welds 30, and identifying a trend in which a quality measure of the identified non-conforming welds deteriorates for any of the multiple welds 30.

[0074] According to another exemplary embodiment, the automatic setting of welding parameters for at least one of the several welding points 30 in block 212 further includes the setting of welding parameters that include the location and orientation of a welding gun 24 of a welding arrangement 20a-20n, which is assigned to the at least one of the several welding points 30, a welding plan for the at least one of the several welding points and an internal logic algorithm in the welding controller 14, but is not limited thereto.

[0075] According to another exemplary embodiment, the method 200, after automatically setting welding parameters for at least one of the several welding points 30 in block 212, further comprises, transitioning to block 214, collecting data relating to the at least one of the several welding points 30 for which settings were made, via the welding monitoring system 28, transitioning to block 216, analyzing the data relating to the at least one of the several welding points 30 for which settings were made, using several algorithms 48 employing deep learning techniques, and transitioning to block 218, determining whether the automatic settings have corrected the identified negative trends.

[0076] If the system controller 12 determines in block 218 that the automatic settings have corrected the identified negative trends, the procedure proceeds to block 220 and ends.

[0077] If the system controller subsequently determines in block 218 that the automatic settings have not corrected the identified negative trends, the procedure 200 includes, with the system controller 12, repeatedly, until the identified negative trends have been corrected, returning to block 206, collecting data relating to each of the multiple welds 30 on each of the multiple vehicle body components 18 via the weld monitoring system 28, transitioning to block 208, analyzing the data relating to each of the multiple welds 30 on each of the multiple vehicle components 18 with the multiple algorithms 48 that use deep learning techniques, transitioning to block 210, identifying negative trends that are associated with the occurrence of deviating welds.for each of the multiple welding points 30 and transitioning to block 212, the setting of welding parameters for at least one of the multiple welding points 30 automatically via communication with the welding controller 14.,

[0078] System 10 and Method 200 of the present disclosure automatically detect negative trends with respect to deviations in welds, including that the number of deviations in welds exceeds a predetermined threshold, that the number of deviations in welds steadily increases, and that the associated quality measure of identified deviations in welds deteriorates, in order to identify the trend and predict future deviations in welds and preemptively adjust welding parameters of welding arrangements to prevent such future deviations in welds.

[0079] The description in this disclosure is essentially only exemplary, and modifications that do not deviate from the main point of this disclosure shall remain within the scope of protection of this disclosure. Such modifications shall not be considered a deviation from the inventive concept and scope of protection of this disclosure.

Claims

System for detecting negative trends in the occurrence of deviating welds and for automatically adjusting welding parameters, the system comprising: a system controller communicating with a welding controller of a vehicle component manufacturing line, the system controller being designed to: monitor multiple welds on each of multiple vehicle components as the multiple vehicle components move along the manufacturing line, with a welding monitoring system communicating with the system controller; collect data relating to each of the multiple welds on each of the multiple vehicle components via the welding monitoring system; analyze the data relating to each of the multiple welds on each of the multiple vehicle components using multiple algorithms employing deep learning techniques;Identifying negative trends associated with the occurrence of deviating welds for each of the multiple welds; and automatically adjusting welding parameters for at least one of the multiple welds via communication with the welding controller. System according to claim 1, wherein the system controller, when collecting data relating to each of the multiple weld points on each of the multiple vehicle components via the welding monitoring system, is further configured to: collect data relating to each of the multiple weld points on each of the multiple vehicle components with sensors positioned on welding subsystems and associated with the welding monitoring system; and collect data relating to each of the multiple weld points on each of the multiple vehicle components with external sensors associated with the welding monitoring system. System according to claim 2, wherein the system controller, when collecting data relating to each of the multiple weld points on each of the multiple vehicle components via the weld monitoring system, is further configured to collect data that includes, but is not limited to, the weld location, the weld plan and variables associated therewith, a weld point identifier (weld point ID), vehicle information and sensor-based control data. System according to claim 3, wherein the welding monitoring system, when collecting data including, but not limited to, the welding location, the welding plan and its associated variables, the welding point identifier (weld point ID), vehicle information and sensor-based control data, is further configured to communicate with and receive information from at least one of the following: a human-machine interface (HMI) designed to enable manual data input by a human operator; a high-frequency identification (RFID) reader designed to read data from an RFID tag positioned on each of the vehicle components; and a visual camera designed to read data from a two-dimensional matrix positioned on each of the vehicle components. System according to claim 3, wherein the system controller, when collecting data relating to each of the multiple weld points on each of the multiple vehicle components via the weld monitoring system, is further configured to collect data for the weld point from a database based on the weld point ID. System according to claim 5, wherein the system controller, when analyzing the data relating to each of the multiple weld points on each of the multiple vehicle components, is further configured to detect deviating weld points and to assign a quality measure for each detected deviating weld point and to compare the weld point ID and associated welding parameters with computer-aided design (CAD) models. System according to claim 6, wherein the system controller, when identifying negative trends associated with the occurrence of non-conforming welds, is designed for each of the multiple welds to: identify the occurrence of a number of non-conforming welds in a sample set that exceeds a predetermined threshold for any of the multiple welds; and identify a trend in which a number of non-conforming welds in a sample set increases for any of the multiple welds; and identify a trend in which a quality measure of the identified non-conforming welds deteriorates for any of the multiple welds. System according to claim 7, wherein the system controller, when automatically setting welding parameters for at least one of the several welding points, is further configured to set welding parameters that include the location and orientation of a welding gun assigned to the at least one of the several welding points, a welding plan for the at least one of the several welding points, and internal logic algorithms in the welding controller, which was not limited thereto. System according to claim 8, wherein the system controller, after automatic settings have been made to the welding parameters for at least one of the multiple welding points, is further configured to: collect data relating to the at least one of the multiple welding points for which settings have been made, via the welding monitoring system; analyze the data relating to the at least one of the multiple welding points for which settings have been made, using multiple algorithms employing deep learning techniques; and determine whether the automatic settings have corrected the identified negative trends. System according to claim 9, wherein the system controller, when the system controller determines that the automatic settings have not corrected the identified negative trends, is designed to repeatedly: collect data relating to each of the multiple welds on each of the multiple vehicle components via the weld monitoring system; analyze the data relating to each of the multiple welds on each of the multiple vehicle components using the multiple algorithms employing deep learning techniques; identify, for each of the multiple welds, negative trends associated with the occurrence of deviant welds; and automatically adjust welding parameters for at least one of the multiple welds via communication with the weld controller.

Images