A method and system for digital traceability of welding process quality

Abstract

This invention relates to the field of data traceability technology, and more particularly to a digital traceability method and system for welding process quality. The method includes: constructing a digital traceability base map based on the current product; determining the process segment number and re-welding number of the welding part, constructing process segment records, calculating hierarchical position identifiers, and writing the hierarchical position identifiers into the process segment records to form a digital traceability chain; reading the quality inspection record corresponding to the product number, mapping the quality inspection record to the process segment record corresponding to the welding part, calculating the original impact value, calculating the mapping weight based on the original impact value, and writing the mapping weight and the number of the quality inspection record into the process segment record to form a digital quality mapping chain; within the analysis set, aggregating the mapping weights using the welding part number as an index to generate the overall impact intensity, calculating the anomaly concentration location, determining the key process record segment, and writing it into the digital quality mapping chain to generate a digital traceability result.

Description

Technical Field

[0001] This invention relates to the field of data traceability technology, and more particularly to a digital traceability method and system for welding process quality. Background Technology

[0002] The plate heat exchanger inlet and outlet water pipe assemblies in heat pump units are subjected to the combined effects of alternating hot and cold temperatures, water circulation, pressure pulsation, equipment vibration, and humid environments over long periods. Their welded connections must not only meet flow guidance and assembly requirements but also directly affect sealing reliability, pressure resistance, corrosion resistance, and overall appearance consistency. In current engineering practice, the industry has gradually shifted from manual brazing of dissimilar metals to all-stainless steel connections and automated welding. These improvements have shown significant effectiveness in preventing verdigris, reducing surface blackening, improving weld consistency, and increasing manufacturing efficiency. Furthermore, with the addition of visual inspection, pressure testing, and salt spray testing, it is possible to determine the quality of components. However, for plate heat exchanger stainless steel inlet and outlet water pipe assemblies, which have multiple welding points, a defined welding sequence, and verifiable subsequent quality results, the challenges remain.

[0003] Current technologies primarily focus on manufacturing completion and inspection, lacking a unified data organization method for "digital traceability of welding process quality." Specifically, there's a lack of seamless digital correlation between product number, welding location, equipment execution sequence, internal welding process segments, and quality results such as appearance, pressure resistance, and salt spray. This means that while it's possible to identify an anomaly in appearance, pressure resistance, or corrosion resistance, it's difficult to pinpoint the specific welding location, welding process segment, or whether it's related to repair welding, continued welding, or subsequent processes. Due to the lack of a continuous mapping from product structure to welding execution and quality results, on-site analysis typically relies on manual record review, batch comparison, or experience-based judgment, limiting both the efficiency and accuracy of problem localization.

[0004] Therefore, based on the fully automated stainless steel welding process and the existing testing system, there is an urgent need to establish a digital traceability method that is compatible with the welding scenarios of this type of component. This method would enable the definition of welding locations, execution sequence, process segment records, and quality results to enter the same chain structure, and allow abnormal results to converge to the corresponding welding locations and their key process segments. This would transform the scattered manufacturing and inspection information in the existing technology into a unified basis that can be used for quality accountability, process verification, and process optimization. Summary of the Invention

[0005] The purpose of this invention is to provide a digital traceability method and system for welding process quality to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, this invention proposes a digital traceability method for welding process quality, comprising: Read the process configuration record corresponding to the current product and extract the structural connection record related to welding. Convert the structural connection record into a welding part number, obtain the first execution trigger value and product number corresponding to each welding part, calculate the execution order of each welding part based on the first execution trigger value, and combine the welding part number, product number and execution order into a digital traceability base map of the current product. The process segment number of the welding part is determined based on the arc start / stop signal, the welding current stability, and the program segment start / stop signal. The number of re-welding times is determined by comparing the consistency of the corresponding numbers of adjacent program segments, and a process segment record is constructed. The hierarchical position identifier is calculated based on the process segment record, and the hierarchical position identifier is written into the process segment record to form a digital traceability chain. Read the quality inspection record corresponding to the product number, map the quality inspection record to the process segment record corresponding to the welding part through a preset mapping relationship, calculate the original influence value through the original influence function, calculate the mapping weight based on the original influence value, write the mapping weight and the number of the quality inspection record into the process segment record, and arrange the process segment records according to the actual writing order in the digital traceability chain to form a digital quality mapping chain; Using the product number as an index, the quality inspection records of the current product are extracted from the digital quality mapping chain and grouped according to the inspection category to form an analysis set. Within the analysis set, the mapping weights are aggregated using the welding part number as an index to generate the overall influence intensity. Based on the overall influence intensity, the anomaly concentration location is calculated. By comparing the distance between the anomaly concentration location and the hierarchical location identifier, the key process record segment is determined and written into the digital quality mapping chain to generate digital traceability results.

[0007] In some embodiments, determining the process segment number of the welding location based on the arc start / stop signal, the welding current stability, and the program segment start / stop signal specifically includes: The starting point of the process segment is determined by detecting the arc start signal, the moment when the welding current reaches a stable threshold, or the program segment start signal. The endpoint of the process segment is determined by detecting the arc extinguishing signal or the program segment end signal; The process segment number is determined based on the starting point and the ending point.

[0008] In some embodiments, before mapping the quality inspection record to the process segment record corresponding to the welding part through a preset mapping relationship, the method further includes: By establishing a mapping relationship between the structural region and the welding location during the debugging phase, the quality inspection records are mapped to the welding location.

[0009] In some embodiments, the parameters of the original influence function include the detection category, the position adjustment coefficient corresponding to the detection category, the order adjustment coefficient corresponding to the detection category and the process segment, the hierarchical position identifier, the execution order, the execution sequence number, and the maximum process segment sequence number in which the welding part appears in the digital traceability chain.

[0010] In some embodiments, the first execution trigger value is obtained through the first setting time of the PLC program segment, the first start time of the robot task node, or the first action timestamp in the controller event log.

[0011] In some embodiments, the analysis set includes an appearance analysis set, a pressure resistance analysis set, and a salt spray analysis set.

[0012] In some embodiments, the location of the anomaly cluster is calculated and generated by an anomaly discrimination function, the parameters of which include hierarchical location identifiers and mapping weights.

[0013] In some embodiments, converting the structural connection record into a weld location number specifically includes: The parsing program assigns values ​​to each welded part in the same product in sequence according to the pre-fixed coding rules, generating welded part numbers.

[0014] In some embodiments, the hierarchical position identifier is generated by an identifier function, the parameters of which include the execution order of the welding part, the process segment number inside the welding part, the number of re-welding times, and the maximum number of process segments allowed for the welding part.

[0015] To achieve the above objectives, another aspect of the present invention proposes a digital traceability system for welding process quality, comprising: The traceability base map construction module is used to read the process configuration record corresponding to the current product, extract the structural connection record related to welding, convert the structural connection record into welding part number, obtain the first execution trigger value and product number corresponding to each welding part, calculate the execution order of each welding part based on the first execution trigger value, and combine the welding part number, product number and execution order into the digital traceability base map of the current product. The traceability chain construction module is used to determine the process segment number of the welding part based on the arc start-stop signal, welding current stability and program segment start-stop signal, determine the number of re-welding by comparing the consistency of the corresponding numbers of adjacent program segments, construct process segment records, calculate the hierarchical position identifier based on the process segment records, and write the hierarchical position identifier into the process segment records to form a digital traceability chain. The quality mapping chain construction module is used to read the quality inspection record corresponding to the product number, map the quality inspection record to the process segment record corresponding to the welding part through a preset mapping relationship, calculate the original influence value through the original influence function, calculate the mapping weight based on the original influence value, write the mapping weight and the number of the quality inspection record into the process segment record, and arrange the process segment records according to the actual writing order in the digital traceability chain to form a digital quality mapping chain. The traceability result generation module is used to extract the quality inspection records of the current product from the digital quality mapping chain using the product number as an index, and group them according to the inspection category to form an analysis set. Within the analysis set, the mapping weights are aggregated using the welding part number as an index to generate the overall influence intensity. Based on the overall influence intensity, the anomaly concentration location is calculated. By comparing the distance between the anomaly concentration location and the hierarchical location identifier, the key process record segment is determined and written into the digital quality mapping chain to generate digital traceability results.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This application uses the product number as the primary index and the welding location and welding execution sequence as the basic framework. First, it forms a product-level digital traceability base map of the welding process quality. Then, it generates a process segment traceability chain by combining the actual execution process of the automatic welding equipment at each welding location. This ensures that each main weld, continuation weld, or repair weld can obtain a clear hierarchical position within its respective welding location. Subsequently, the appearance inspection results, pressure resistance test results, and salt spray test results are mapped to the process chain according to the correspondence between the product, welding location, and process segment, forming a unified digital quality mapping relationship between quality results and welding execution process. On this basis, the quality mapping weights of different process segments under the same welding location are converged, and the output is a digital traceability result that can directly indicate abnormal welding locations and key process segments.

[0017] Through the above technical solution, this invention organizes information that was originally scattered in process configuration, equipment execution, and quality inspection into a data chain that runs through the entire process. This allows quality anomalies to be judged from the product level down to the welding part level and process segment level, enabling quantitative location of the source of the problem. As a result, the quality analysis of the stainless steel inlet and outlet water pipe components of the plate heat exchanger of the heat pump unit after automatic welding is upgraded from result judgment to process traceability and cause location, which enhances the pertinence of quality management, process correction, and batch manufacturing consistency control. Attached Figure Description

[0018] Figure 1 A flowchart of a digital traceability method for welding process quality provided in this application embodiment; Figure 2 A block diagram of a welding process quality digital traceability system provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] refer to Figure 1 As shown, one embodiment of this application proposes a method for digital traceability of welding process quality, including: S1: Read the process configuration record corresponding to the current product, extract the structural connection record related to welding, convert the structural connection record into a welding part number, obtain the first execution trigger value and product number corresponding to each welding part, calculate the execution order of each welding part based on the first execution trigger value, and combine the welding part number, product number, and execution order into a digital traceability base map of the current product, specifically including: After the stainless steel inlet and outlet water pipe assemblies of the plate heat exchanger enter the automatic welding station, the system first reads the process configuration record corresponding to the current product and organizes the structural connection information related to welding into a set of welding positions for the current product. The process configuration record usually already exists in the MES distribution file, PLC recipe table, or robot task file. In engineering, the connection name, connection position, and welding execution segment number can be directly read from the corresponding fields. Taking a common inlet and outlet water pipe assembly as an example, the configuration file will contain structural connection records such as plate heat exchanger interface connections, pull-out hole connections, and joint connections. After reading these structural connection records, the system converts each connection record into a stable welding position number. This number is generated by the process configuration parsing program, which assigns values ​​to each welding position in the same product sequentially according to pre-defined coding rules. For example, for interface-type connections... , Hole-type connection generation Connector type connection generation Meanwhile, the current product number The work order barcode is read by the production line barcode scanner and transmitted to the equipment controller, or it is directly written to the task buffer of this product by the host computer when the work order is issued. In this way, the product number... and welding part number to They then enter the same data link, laying the foundation for subsequent sequential positioning.

[0021] After numbering the welding parts, the execution order of each welding part is further determined based on the actual execution sequence of the automatic welding equipment. This employs the basic approach of discrete mathematics' order relation and sequence statistics: first, the initial execution trigger value corresponding to each welding part is obtained, and then the execution order is formed according to the trigger values ​​from smallest to largest. In engineering implementation, this initial execution trigger value can directly come from the first set time of the PLC program segment, the first start time of the robot task node, or the first action timestamp in the controller event log. If the first... The initial execution trigger value for each welding location is denoted as: Then its execution order Determine using the following formula: ; in, Indicates the first The execution sequence of each welding part in the current product welding process; This indicates the total number of welded parts in the current product, obtained after analyzing the process configuration. Indicates the first The initial execution trigger value for each welding part comes from the PLC event table, robot event log, or controller task record; This indicates the initial execution trigger value for the remaining welded parts; Indicates the judgment result, when Earlier The value is 1 if the condition is met, and 0 otherwise. This expression originates from the rank definition of discrete sequences, and its original form is used to determine the order of an element in a finite set. Here, it is rewritten as a calculation form for the execution sequence of welding parts. The purpose is to directly convert the original trigger values ​​in the equipment log into the execution order required for subsequent tracing, without relying on manual comparison of program segments. If the trigger values ​​output by the field controller are the same, the program segment number inherent in the equipment program can be directly used as the basis for determining the order when values ​​are the same, and the program segment executed first can be written into the smaller execution order.

[0022] Execution order After receiving it, then add the product number. Welding part number and the corresponding execution order Combined into a digital traceability base map of the welding process quality of the current product Its organizational form uses set representation, derived from the conventional notation of "ordered identification of elements" in discrete structures, specifically written as... ; in, This represents the digital traceability base map of the current product's welding process quality. Indicates the current product number; Indicates the first Each welding part number is generated by the process configuration parsing program; Indicates the first The execution sequence of each welding part is calculated from the previous formula; This indicates the total number of welded parts. The first formula converts the equipment execution record into an execution sequence, while the second formula links the execution sequence to the welded parts of the product. Therefore, they are sequential: only the first formula can be used to determine the total number of welded parts. Only then can a base map be formed. When the product number is The welding parts are as follows: , , , When the corresponding initial execution trigger values ​​are recorded as 12, 27, 41, and 56 respectively, substituting them into the previous formula yields: For In fact, there is no earlier trigger value than 12, therefore ;right In this regard, only 12 is earlier than 27, therefore ;right In other words, 12 and 27 are earlier than 41, therefore ;right In terms of 12, 27, and 41, they are earlier than 56, therefore After substituting the values ​​into the formula, the traceability base map of the current product is obtained as follows: ; ; This set of results can be directly written to the MES traceability table, or stored in a PLC data block or robot task context, so that subsequent welding process segments can be recorded according to the execution sequence. Directly index to the corresponding welding part This unifies the connection relationships in the process configuration, the sequence of actions in equipment execution, and the current product number into the same traceability base map.

[0023] S2: Determine the process segment number of the welding part based on the arc start / stop signal, welding current stability and program segment start / stop signal, determine the number of re-welding times by comparing the consistency of the corresponding numbers of adjacent program segments, construct process segment records, calculate the hierarchical position identifier based on the process segment records, and write the hierarchical position identifier into the process segment records to form a digital traceability chain. A digital traceability base map of welding process quality has been obtained in S1. Next, this step uses the base map as input to map the discrete structural information within it to the actual welding process, thus forming a digital traceability chain for welding process quality that can be used for subsequent quality mapping. In practice, the automatic welding equipment continuously outputs program segment numbers or welding task node numbers during execution. These numbers were already matched with the welding position numbers in the process configuration during the equipment debugging phase. A mapping relationship has been established, so whenever the device enters a certain program segment, the corresponding welding part can be determined. And its execution order in the base map. Meanwhile, the control system uses the arc start signal, the moment when the welding current reaches a stable threshold, or the program segment start signal as the starting point of a process segment, and determines the end point of the process segment when the arc is extinguished or the program segment ends, thereby generating the process segment sequence number within the welding area. When the same welding location is called again after welding is completed, the system checks the "location number corresponding to the current program segment". Returning to the previous location with a different end number The event "" will be counted as one reflow and a corresponding reflow count will be generated. Therefore, each actual welding action is transformed into a process containing... Information process segment record.

[0024] To organize the aforementioned discrete records into a continuous, ordered execution chain, it is necessary to compress the "welding position sequence," "internal process segment sequence," and "reflow level" into a single monotonically increasing identifier. This construction originates from the lexicographical embedding method in discrete mathematics, which maps multiple ordered variables to a real number interval to maintain their original ordering. In this step, we first use S1... As the main order, then introduce As a subdivision of order, and on this basis, add As a sub-item re-entering the same part, a labeling function is constructed, the expression of which is: ; in, Indicates the first Each process segment is marked with its hierarchical position within its respective welding area; This indicates the execution order of the welding parts to which this process segment belongs in the base map, and its value is determined by S1 through the equipment execution order; This indicates the process segment number within the welding area, generated by the arc start signal or the program segment start signal. This indicates the number of times the welding part has been re-welded before the current process segment, which is obtained by the controller from the statistics of the part switching history; This indicates the maximum number of process segments allowed for the welding location, and its value is derived from the process configuration or welding procedure parameter settings. The construction of this formula follows the following derivation logic: First, use an integer... Identify the welding location to which this process segment belongs; then introduce... This indicates the sequence of process segments within the welding area; further, it introduces... This indicates a more detailed layer when re-entering the same welding area. Because... and ,therefore Always falls within the range This allows it to serve as a hierarchical coordinate system for different process segments within the same welding area, providing a stable positional reference for mapping subsequent quality results to process segments.

[0025] Explain based on actual working conditions, assuming a certain product number. its welding parts The execution order in the base map is as follows: The process configuration specifies that this part is allowed to have a maximum of [number] [units]. This is a process segment. In the actual welding process, the equipment first performs the main welding of this part, at which point... and Substituting into the above formula, we get The equipment was then moved to another location to complete the welding before returning. Perform repair welding at this time and Substituting into the above formula, we get Therefore, two conditions can be met. And all are located in the interval The ordered identifiers within the chain thus maintain order throughout the entire chain. The location of the second welding point clearly distinguishes between its main welding and repair welding processes. Repeating the above calculations for all welding points yields a set of strictly monotonically increasing... sequence.

[0026] After obtaining all process segments Afterwards, the control system writes the process segment records according to the actual start order of each process segment, and records the corresponding product number. Welding part number Execution order and process segment number This information is written into the data structure to form a digital traceability chain for welding process quality. This chain can be represented as a set of tuples arranged in the actual triggering order, where each tuple corresponds to one actual welding action. This is used to identify the layer position of the action within the corresponding welding area. This can be used later when mapping quality inspection results. and Determine the welding location, and then... This distinguishes the specific process segments within the welding area, thereby achieving a precise correlation between the quality result and the welding execution process. Throughout the process, the base map provided by S1... In It is fully inherited and participates in the calculation, while and The structure is generated by the device's real-time signals and embedded in it, so that static structural information and dynamic execution information are expressed in a unified manner in the same chain.

[0027] S3: Read the quality inspection record corresponding to the product number, map the quality inspection record to the process segment record corresponding to the welding part through a preset mapping relationship, calculate the original influence value through the original influence function, calculate the mapping weight based on the original influence value, write the mapping weight and the number of the quality inspection record into the process segment record, and arrange the process segment records according to the actual writing order in the digital traceability chain to form a digital quality mapping chain, specifically including: A digital traceability chain for welding process quality has been established in S2. Subsequently, based on this chain, this step maps the actual detected appearance, pressure resistance, and salt spray results step by step to specific welding process segments, thereby generating a digital quality mapping chain for welding process quality. In practice, the current product number is first read from the detection system. All corresponding quality inspection records are available. These records originate from the visual inspection station on the production line (manual entry of defect areas or output of defect location coordinates from industrial cameras), pressure testing equipment (output of whether there is leakage and the corresponding interface label for leakage), and salt spray test records (output of corrosion areas or failure indicators). All test data are organized by product number. It is stored as an index, so it can be directly linked to the traceability chain. The association is then established. Subsequently, based on the defect area or interface identifier given in the inspection record, the "structural area—welding location number" established during the commissioning phase is used to... "The mapping relationship pinpoints each quality record to a specific welding location." This allows quality information to be precisely transferred from the product level down to the welding component level.

[0028] After completing the part-level matching, each quality record is further mapped to the set of all process segments corresponding to that welding part in S2. Because... Each process segment has passed Comparable hierarchical positional relationships were established within the same welded area, and each With the only The combinations correspond, so they can be directly filtered by the criteria. Obtain all process segments under this location and retrieve the corresponding... and Building upon this, to reflect the varying degrees of influence of different process stages on the quality results at the same welding location, a method for constructing influence values ​​based on sequence position weighting is introduced. This method originates from the position weighting model in sequence analysis, and its basic idea is to express the relative influence of each element in an ordered sequence through position offsets. This step improves upon this, transforming the globally ordered positions in S2... Starting position of the part in S1 Combined, a local position quantity that only applies within the welded area is constructed. and the process segment number Together they constitute the influence value of the process segment The expression for the original influence function is: ; in, Indicates the first The quality record for the first The original impact values ​​of each process segment; This indicates the testing category of the quality record. Its value comes from the output of the testing equipment and can be appearance, pressure resistance, or salt spray. Indicates the category of detection The corresponding position adjustment coefficient is obtained from process debugging records or historical data statistics; Indicates the category of detection The corresponding process segment sequence adjustment coefficient also comes from the process parameter settings; Indicates the first The hierarchical position identifiers calculated in S2 for each process segment; This indicates the execution order of the welding parts to which this process segment belongs, as determined in S1. This indicates the execution sequence number of the process segment within its respective welding location, generated by a device event. Indicates the current welding position In the traceability chain The largest process segment number that appears in the sequence is obtained by traversing all processes in that part. The formula adds a baseline term "1" and a double-coefficient term related to the detection category to the original sequence weighted model, making the distribution of different quality types within the same welded part different. This is more in line with the failure characteristics of stainless steel inlet and outlet water pipe assemblies of plate heat exchangers in actual welding. For example, appearance abnormalities are more concentrated in the later arc termination and repair welding positions, while pressure resistance abnormalities depend more on the continuity of the main weld section.

[0029] After obtaining the original impact values ​​of all relevant process segments Next, normalization is used to distribute the influence of the same quality record within the weld area. This process is derived from the classic weighted normalization method, the purpose of which is to ensure that the sum of the weights of the same quality record across all relevant process segments is 1, thus facilitating direct comparison of the influence of each process segment. The specific calculation formula is as follows: ; in, Indicates the first Quality record mapped to the first The final weights of each process segment; denominator This indicates that the quality record is at the corresponding welding location. The sum of the original impact values ​​of all process segments. The first formula is used to calculate the original impact strength of each process segment, and the second formula is used to convert these impact strengths into standardized allocation weights. The two are logically connected in a continuous derivation relationship. A specific example will be used to illustrate this, assuming a product number... its welding parts In the traceability chain It contains two process segments, corresponding to They are 2.25 and 2.5625 respectively, and Process segment number They are 1 and 2 respectively. If a certain appearance inspection record corresponds to For appearance, and set during debugging. , Substituting this into the previous equation, we obtain the original impact value of the first process segment. The original impact value of the second process segment is The sum of the two is 3.0875. Substituting these values ​​into the normalization formula, we can obtain the weight of the first process segment as follows: The weight of the second process segment is Through this calculation process, the quality record is refined from a single "location anomaly" to "the distribution of that location across different process segments," and the distribution ratio has a clear numerical expression.

[0030] After completing the above calculations, the product number will be... Welding part number Execution order Process segment number Hierarchical position identifier and quality record number , testing categories and mapping weights They are all written to the same data record and traced according to the process segments. The actual writing order is arranged to form a digital quality mapping chain for the welding process. This chain can be stored as an ordered table structure in a database, or it can be cached on the device side for direct use by subsequent steps. In this way, the structural relationships in S1, the execution process in S2, and the quality results in this step are unified into the same data chain, allowing for direct reference during anomaly localization. The process segments with higher weights are read from the data, thereby enabling precise traceability from quality results to the welding process.

[0031] S4: Using the product number as an index, extract the quality inspection records of the current product from the digital quality mapping chain, and group them according to the inspection category to form an analysis set. Within the analysis set, aggregate the mapping weights using the weld part number as an index to generate the overall influence intensity. Calculate the anomaly concentration location based on the overall influence intensity. Determine the key process record segment by comparing the distance between the anomaly concentration location and the hierarchical position identifier, and write it into the digital quality mapping chain to generate digital traceability results. Specifically, this includes: In S3, a digital quality mapping chain for welding process quality has been obtained. After that, this step directly utilizes The "product number" system already established in China —Welding area —Execution order —Process segment number —Global Location —Quality mapping weights This complete correlation structure quantitatively converges on quality anomalies and generates digital traceability results for welding process quality that can be directly used for process decision-making. In specific implementation, it first uses the product number... For indexing, from Extract all quality mapping records for the product and categorize them according to testing category. The data were grouped so that the appearance, pressure resistance, and salt spray data formed separate analysis sets. Within each set, the analysis was based on the weld location. As an index, the mapping weights of all process segments under this part. Aggregation is performed to obtain the overall influence intensity of the region under the detection category, and the corresponding values ​​for each process segment are retained. The value is used for subsequent positioning.

[0032] After completing the part-level aggregation, it is necessary to compress the weight information distributed across multiple process segments into a single index that can reflect the location of anomaly concentration. This process uses the calculation approach of discrete probability distribution expectation to transform the weight information obtained in S3. Treating them as discrete weights, Treating it as a location variable, the location of the anomaly cluster within the current quality category is obtained by using a weighted average through an anomaly discriminant function. The calculation expression for the anomaly discriminant function is as follows: ; in, Indicates the welding area The location of the anomaly concentration under the current quality category; where in the formula... The summation range is the current product Current detection category And the current welding location is All quality records, for The summation range is the entire process segment corresponding to each of the above quality records at that welding location; Indicates the first welded part The hierarchical position identifier obtained for each process segment in S2; This represents the quality mapping weights calculated in S3, derived from the allocation calculation between quality inspection results and process segments. This expression originates from the calculation form of the expected value of discrete random variables in probability and statistics; its original form is... In this scenario, by The quality mapping weights are replaced and double summation is used to ensure that multiple quality records under the same welding location and the same inspection category can be uniformly converged to a single location index.

[0033] In obtaining Then, through comparison With each of the welded parts The distance between them is used to determine the process segment closest to the center of gravity as the critical process segment. In specific implementation, all process segments within that area are calculated. Choose the minimum value. This serves as the process segment number most relevant to the anomaly. Taking specific data as an example, when the product... Welding parts When there is an appearance anomaly record, S3 obtains two process segments. The values ​​are 2.25 and 2.5625 respectively, corresponding to weights. , Since there is only one quality record, the above formula can be directly degenerated into: Further calculate the distances between the two process segments and this value, respectively. and Therefore, the second process segment is determined to be... This is a critical process segment. The results indicate that within this welding area, the subsequent execution process has a more significant impact on appearance anomalies.

[0034] After the above calculations are completed, the product number will be... Welding part number Execution order Key process segment number and the corresponding This information is included in the traceability results structure to form a digital traceability result for the welding process quality. This result can be directly used for production line display or process adjustment, for example, to indicate "a certain product is at a certain welding position on the [number]th ... "There are abnormal risks in the welding process," which guides engineers to optimize parameters or modify processes for specific welding stages. Through this step, the quality weights scattered across multiple process segments in S3 are compressed into clear location results, making the entire solution form a continuous and closed data processing path from base map construction, process chain generation, quality mapping to anomaly location.

[0035] refer to Figure 2 As shown, in another aspect of this application embodiment, a digital traceability system for welding process quality is also proposed, including: The traceability base map construction module is used to read the process configuration record corresponding to the current product, extract the structural connection record related to welding, convert the structural connection record into welding part number, obtain the first execution trigger value and product number corresponding to each welding part, calculate the execution order of each welding part based on the first execution trigger value, and combine the welding part number, product number and execution order into the digital traceability base map of the current product. The traceability chain construction module is used to determine the process segment number of the welding part based on the arc start-stop signal, welding current stability and program segment start-stop signal, determine the number of re-welding by comparing the consistency of the corresponding numbers of adjacent program segments, construct process segment records, calculate the hierarchical position identifier based on the process segment records, and write the hierarchical position identifier into the process segment records to form a digital traceability chain. The quality mapping chain construction module is used to read the quality inspection record corresponding to the product number, map the quality inspection record to the process segment record corresponding to the welding part through a preset mapping relationship, calculate the original influence value through the original influence function, calculate the mapping weight based on the original influence value, write the mapping weight and the number of the quality inspection record into the process segment record, and arrange the process segment records according to the actual writing order in the digital traceability chain to form a digital quality mapping chain. The traceability result generation module is used to extract the quality inspection records of the current product from the digital quality mapping chain using the product number as an index, and group them according to the inspection category to form an analysis set. Within the analysis set, the mapping weights are aggregated using the welding part number as an index to generate the overall influence intensity. Based on the overall influence intensity, the anomaly concentration location is calculated. By comparing the distance between the anomaly concentration location and the hierarchical location identifier, the key process record segment is determined and written into the digital quality mapping chain to generate digital traceability results.

[0036] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method of digital traceability of welding process quality, characterized in that, include: Read the process configuration record corresponding to the current product and extract the structural connection record related to welding. Convert the structural connection record into a welding part number, obtain the first execution trigger value and product number corresponding to each welding part, calculate the execution order of each welding part based on the first execution trigger value, and combine the welding part number, product number and execution order into a digital traceability base map of the current product. The process segment number of the welding part is determined based on the arc start / stop signal, the welding current stability, and the program segment start / stop signal. The number of re-welding times is determined by comparing the consistency of the corresponding numbers of adjacent program segments, and a process segment record is constructed. The hierarchical position identifier is calculated based on the process segment record, and the hierarchical position identifier is written into the process segment record to form a digital traceability chain. Read the quality inspection record corresponding to the product number, map the quality inspection record to the process segment record corresponding to the welding part through a preset mapping relationship, calculate the original influence value through the original influence function, calculate the mapping weight based on the original influence value, write the mapping weight and the number of the quality inspection record into the process segment record, and arrange the process segment records according to the actual writing order in the digital traceability chain to form a digital quality mapping chain; Using the product number as an index, the quality inspection records of the current product are extracted from the digital quality mapping chain and grouped according to the inspection category to form an analysis set. Within the analysis set, the mapping weights are aggregated using the welding part number as an index to generate the overall influence intensity. Based on the overall influence intensity, the anomaly concentration location is calculated. By comparing the distance between the anomaly concentration location and the hierarchical location identifier, the key process record segment is determined and written into the digital quality mapping chain to generate digital traceability results.

2. The method of claim 1, wherein, The process segment number for determining the welding location based on the arc start / stop signal, welding current stability, and program segment start / stop signal specifically includes: The starting point of the process segment is determined by detecting the arc start signal, the moment when the welding current reaches a stable threshold, or the program segment start signal. The endpoint of the process segment is determined by detecting the arc extinguishing signal or the program segment end signal; The process segment number is determined based on the starting point and the ending point.

3. The method of claim 1, wherein, Before mapping the quality inspection records to the process segment records corresponding to the welding parts through a preset mapping relationship, the method further includes: By establishing a mapping relationship between the structural region and the welding location during the debugging phase, the quality inspection records are mapped to the welding location.

4. The method of claim 1, wherein, The parameters of the original influence function include the detection category, the position adjustment coefficient corresponding to the detection category, the order adjustment coefficient corresponding to the detection category and the process segment, the hierarchical position identifier, the execution order, the execution sequence number, and the maximum process segment sequence number in which the welding part appears in the digital traceability chain.

5. The method of claim 1, wherein, The initial execution trigger value is obtained from the first setting time of the PLC program segment, the first start time of the robot task node, or the first action timestamp in the controller event log.

6. The weld procedure quality digital traceability method of claim 1, wherein, The analysis set includes the appearance analysis set, the pressure resistance analysis set, and the salt spray analysis set.

7. The weld procedure quality digital traceability method of claim 1, wherein, The locations of the anomalies are calculated using an anomaly discrimination function, whose parameters include hierarchical location identifiers and mapping weights.

8. The weld procedure quality digital traceability method of claim 1, wherein, The process of converting the structural connection record into a welding part number specifically includes: The parsing program assigns values ​​to each welded part in the same product in sequence according to the pre-fixed coding rules, generating welded part numbers.

9. The digital traceability method for welding process quality according to claim 1, characterized in that, The hierarchical position identifier is generated by an identifier function. The parameters of the identifier function include the execution order of the welding part, the process segment number inside the welding part, the number of re-welding times, and the maximum number of process segments allowed for the welding part.

10. A digital traceability system for welding process quality, characterized in that, include: The traceability base map construction module is used to read the process configuration record corresponding to the current product, extract the structural connection record related to welding, convert the structural connection record into welding part number, obtain the first execution trigger value and product number corresponding to each welding part, calculate the execution order of each welding part based on the first execution trigger value, and combine the welding part number, product number and execution order into the digital traceability base map of the current product. The traceability chain construction module is used to determine the process segment number of the welding part based on the arc start-stop signal, welding current stability and program segment start-stop signal, determine the number of re-welding by comparing the consistency of the corresponding numbers of adjacent program segments, construct process segment records, calculate the hierarchical position identifier based on the process segment records, and write the hierarchical position identifier into the process segment records to form a digital traceability chain. The quality mapping chain construction module is used to read the quality inspection record corresponding to the product number, map the quality inspection record to the process segment record corresponding to the welding part through a preset mapping relationship, calculate the original influence value through the original influence function, calculate the mapping weight based on the original influence value, write the mapping weight and the number of the quality inspection record into the process segment record, and arrange the process segment records according to the actual writing order in the digital traceability chain to form a digital quality mapping chain. The traceability result generation module is used to extract the quality inspection records of the current product from the digital quality mapping chain using the product number as an index, and group them according to the inspection category to form an analysis set. Within the analysis set, the mapping weights are aggregated using the welding part number as an index to generate the overall influence intensity. Based on the overall influence intensity, the anomaly concentration location is calculated. By comparing the distance between the anomaly concentration location and the hierarchical location identifier, the key process record segment is determined and written into the digital quality mapping chain to generate digital traceability results.

Images