Method and system for defect traceability in watch key processing process

Abstract

The invention discloses a defect traceability method and system in a watch key processing process. The method comprises the following steps: acquiring a process value and a quality value; preprocessing the quality value to obtain a performance data set; performing performance evaluation on the performance data set based on a multiple regression analysis method to obtain an evaluation result; performing node construction and traceability analysis according to the process value and the quality value to obtain a node graph and a mapping relation table; performing model construction according to the node graph, the mapping relation table and the evaluation result to obtain a prediction model; performing defect prediction according to the prediction model to obtain a prediction result; classifying defect types in the prediction result based on a clustering algorithm to obtain a standard defect library; performing feature identification according to a standard defect library to obtain a traceability identification code; and performing association mapping on the real-time process quality data and the traceability identification code to obtain defect traceability information. According to the method, the processing efficiency of the defect problem can be improved by constructing the node graph and the corresponding relation between the process and the quality index.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent manufacturing and quality control, and particularly to a method and system for defect traceability in the processing of watch buttons. Background Art

[0002] Various defects are likely to occur during the processing of watch buttons, which will seriously affect the user experience of the buttons and the product quality. At present, there are many technical problems in defect traceability. First of all, there are no scientific quantitative standards and testing methods for the performance indicators of buttons such as response speed and durability, resulting in the inability to accurately evaluate the button quality. Secondly, the button processing involves multiple processes and there are many nodes where defects occur. Existing quality management means are difficult to accurately locate the specific links and reasons that cause defects. Moreover, the types of defects are diverse and the severity levels are different. There is an urgent need to establish a defect classification system and design a reasonable sorting algorithm to clarify the priority order of defect handling. Finally, a large amount of defect data records are scattered, lacking a unified management platform and traceability mechanism, which affects the rapid response and closed-loop processing of defect problems. There is an urgent need to build a complete management system for defect information collection, storage, analysis, and traceability, realize the refined control of the entire life cycle of defects, and use traceability codes to achieve accurate retrieval of defect information and rapid problem positioning, so as to continuously improve the watch button processing technology and comprehensively guarantee the product quality.

[0003] In an existing technology, the specific implementation includes: sampling and inspecting buttons through traditional quality inspection means to detect their performance indicators such as response speed and durability. These inspection means will include manual pressing tests, simple mechanical test equipment, etc. The inspection results are usually recorded in paper or spreadsheets for subsequent analysis and processing.

[0004] However, in the existing technology, the quality management means are difficult to accurately locate the specific links and reasons that cause defects. The types of defects are diverse and the severity levels are different. There is a lack of a unified defect classification system and priority sorting algorithm, resulting in a low response speed and processing efficiency for defect problems. Summary of the Invention

[0005] The present invention provides a method and system for defect traceability in the processing of watch buttons to solve the problems in the existing technology that the quality management means are difficult to accurately locate the specific links and reasons that cause defects, the types of defects are diverse and the severity levels are different, and there is a lack of a unified defect classification system and priority sorting algorithm, resulting in a low response speed and processing efficiency for defect problems.

[0006] In the first aspect, to solve the above technical problems, the present invention provides a method for defect traceability in the processing of watch buttons, including: Obtaining process values and quality values; wherein, the quality values include force values, displacement amounts, and resistance values; Preprocess the force value, the displacement amount, and the resistance value to obtain a performance data set; Perform performance evaluation on the performance data set based on the multiple regression analysis method to obtain an evaluation result; Perform node construction and traceability analysis according to the process value and the quality value to obtain a node diagram and a mapping relationship table; Construct a model according to the node diagram, the mapping relationship table, and the evaluation result to obtain a prediction model; Perform defect prediction according to the prediction model to obtain a prediction result; Classify the defect types in the prediction result based on the clustering algorithm to obtain a standard defect library; perform feature identification according to the standard defect library to obtain a traceability identification code; Perform associated mapping on the real-time process quality data and the traceability identification code to obtain defect traceability information.

[0007] In an implementable manner of the first aspect, the preprocessing the force value, the displacement amount, and the resistance value to obtain a performance data set includes: Perform synchronization calibration according to the force value, the displacement amount, and the resistance value to obtain a calibration data set; Align the calibration data set in time series to obtain a sequence data set; Extract features according to the sequence data set to obtain a multi-dimensional feature data set; Classify the multi-dimensional feature data based on the clustering algorithm to obtain a performance data set.

[0008] In an implementable manner of the first aspect, the performing performance evaluation on the performance data set based on the multiple regression analysis method to obtain an evaluation result includes: Construct a model for the performance data set based on the multiple regression analysis method to obtain an initial association model; Perform dimension expansion on the initial association model by incorporating dynamic factors and interaction factors to obtain an extended association model; Perform feature analysis on the real-time process quality data according to the extended association model to obtain a feature value; Calculate a weight coefficient according to the feature value; Optimize the performance data set according to the weight coefficient to obtain an optimized process parameter set; Perform performance evaluation according to the optimized process parameter set to obtain an evaluation result.

[0009] In an implementable manner of the first aspect, the calculating a weight coefficient according to the feature value includes: The weight coefficient is calculated by the following formula: Wherein, represents the weight coefficient of the th index, represents the eigenvalue of the th index, is the total number of items of the index.

[0010] In an implementable manner of the first aspect, the node construction and traceability analysis are performed according to the process value and the quality value to obtain a node graph and a mapping relation table, including: Performing data cleaning and format conversion according to the process value and the quality value to obtain standardized data; Storing the standardized data into a distributed repository based on a preset node relationship model to obtain a node graph; Performing anomaly analysis on the node graph based on an anomaly detection analyzer. If there are anomaly data points in the node graph, quality tracing and root cause analysis are performed on the anomaly data points to obtain a mapping relation table.

[0011] In an implementable manner of the first aspect, the feature identification is performed according to the standard defect library to obtain a traceability identification code, including: Performing weight calculation on the standard defect library based on the analytic hierarchy process to obtain an index weight value; Calculating the priority score of the defect according to the index weight value to obtain a score value; Sorting the score values from high to low to obtain a defect processing sequence; Encoding the defect processing sequence based on a pre-stored defect coding rule to obtain a traceability identification code.

[0012] In an implementable manner of the first aspect, the real-time process quality data is associated and mapped with the traceability identification code to obtain defect traceability information, including: Judging according to the real-time process quality data and a preset threshold range; if the real-time process quality data exceeds the preset threshold range, it is determined that there is a defect source; Searching for the defect source according to the traceability identification code to obtain the storage location and timestamp of the defect information; Outputting the storage location of the defect information and the timestamp to obtain defect traceability information.

[0013] In a second aspect, the present invention provides a system for defect traceability in the processing of a watch button, including: A data acquisition module for acquiring a process value and a quality value; wherein, the quality value includes a force value, a displacement amount, and a resistance value; A data cleaning module, configured to preprocess the force value, the displacement amount, and the resistance value to obtain a performance data set; A performance evaluation module, configured to perform performance evaluation on the performance data set based on a multiple regression analysis method to obtain an evaluation result; A node construction and analysis module, configured to perform node construction and traceability analysis according to the process value and the quality value to obtain a node diagram and a mapping relation table; A model construction module, configured to construct a model according to the node diagram, the mapping relation table, and the evaluation result to obtain a prediction model; A defect prediction module, configured to perform defect prediction according to the prediction model to obtain a prediction result; A defect classification module, configured to classify the defect types in the prediction result based on a clustering algorithm to obtain a standard defect library; A defect coding module, configured to perform feature identification according to the standard defect library to obtain a traceability identification code; An output module, configured to perform associated mapping on real-time process quality data and the traceability identification code to obtain defect traceability information.

[0014] In an implementable manner of the second aspect, the preprocessing of the force value, the displacement amount, and the resistance value to obtain a performance data set includes: Performing synchronization calibration according to the force value, the displacement amount, and the resistance value to obtain a calibration data set; Performing time series alignment on the calibration data set to obtain a sequence data set; Performing feature extraction according to the sequence data set to obtain a multi-dimensional feature data set; Classifying the multi-dimensional feature data based on a clustering algorithm to obtain a performance data set.

[0015] In an implementable manner of the second aspect, the performing performance evaluation on the performance data set based on a multiple regression analysis method to obtain an evaluation result includes: Constructing a model for the performance data set based on a multiple regression analysis method to obtain an initial association model; Performing dimension expansion on the initial association model by incorporating dynamic factors and interaction factors to obtain an extended association model; Performing feature analysis on real-time process quality data according to the extended association model to obtain a feature value; Calculating a weight coefficient according to the feature value; Optimizing the performance data set according to the weight coefficient to obtain an optimized process parameter set; Performing performance evaluation according to the optimized process parameter set to obtain an evaluation result.

[0016] In an implementable manner of the second aspect, the calculating of the weight coefficient according to the eigenvalue includes: The weight coefficient is calculated by the following formula: where, represents the weight coefficient of the th index, represents the eigenvalue of the th index, is the total number of items of the index.

[0017] In an implementable manner of the second aspect, the constructing of nodes and traceability analysis according to the process value and quality value to obtain a node graph and a mapping relation table includes: Performing data cleaning and format conversion according to the process value and quality value to obtain standardized data; Storing the standardized data into a distributed repository based on a preset node relationship model to obtain a node graph; Performing anomaly analysis on the node graph based on an anomaly detection analyzer. If there are abnormal data points in the node graph, performing quality traceability and root cause analysis on the abnormal data points to obtain a mapping relation table.

[0018] In an implementable manner of the second aspect, the obtaining of a traceability identification code by performing feature identification according to the standard defect library includes: Calculating the weight of the standard defect library based on the analytic hierarchy process to obtain an index weight value; Calculating the priority score of the defect according to the index weight value to obtain a score value; Sorting the score values from high to low to obtain a defect processing sequence; Encoding the defect processing sequence based on a pre-stored defect coding rule to obtain a traceability identification code.

[0019] In an implementable manner of the second aspect, the associative mapping of real-time process quality data with the traceability identification code to obtain defect traceability information includes: Judging according to the real-time process quality data and a preset threshold range; if the real-time process quality data exceeds the preset threshold range, it is determined that there is a defect source; Searching for the defect source according to the traceability identification code to obtain the storage location and timestamp of the defect information; Outputting the storage location of the defect information and the timestamp to obtain defect traceability information.

[0020] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the method for defect traceability during the processing of a watch button described in any one of the above is implemented.

[0021] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program. Wherein, when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method for defect traceability during the processing of a watch button described in any one of the above.

[0022] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a method for defect traceability during the processing of a watch button, including obtaining process values and quality values; wherein, the quality values include force values, displacement amounts, and resistance values; preprocessing the force values, the displacement amounts, and the resistance values to obtain a performance data set; performing performance evaluation on the performance data set based on a multiple regression analysis method to obtain an evaluation result; constructing nodes and performing traceability analysis according to the process values and the quality values to obtain a node graph and a mapping relationship table; constructing a prediction model according to the node graph, the mapping relationship table, and the evaluation result; performing defect prediction according to the prediction model to obtain a prediction result; classifying the defect types in the prediction result based on a clustering algorithm to obtain a standard defect library; performing feature identification according to the standard defect library to obtain a traceability identification code; and performing associated mapping between the real-time process quality data and the traceability identification code to obtain defect traceability information. The present invention establishes a performance index quantization model based on physical characteristics, collects key response time and pressure change data, and constructs a multi-dimensional performance index data set; adopts multiple regression analysis to establish an association model between key performance and process parameters, and determines key influencing factors; the present invention deploys online detection devices at each processing node, collects process parameters and quality data in real time, and constructs a node graph and the corresponding relationship between processes and quality indicators. Based on the detection results, defects are classified and a standard defect library is established; a defect priority evaluation model is constructed through the analytic hierarchy process to generate a list of disposal sequences. The present invention also designs a defect tracking mechanism based on a traceability code to realize the information chain of the entire life cycle of defects, and implements a quality data analysis and early warning system to provide decision support for quality improvement; the present invention can effectively improve key performance, reduce the defect rate, and realize the full-process control of product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a schematic flowchart of the method for defect traceability during the processing of a watch button provided by the first embodiment of the present invention; Figure 2It is a schematic structural diagram of a system for defect traceability in the processing of watch buttons provided by the second embodiment of the present invention. Detailed implementation manners

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0025] Refer to Figure 1 , the first embodiment of the present invention provides a method for defect traceability in the processing of watch buttons, including the following steps: S1. Obtain process values and quality values; wherein, the quality values include force values, displacement amounts, and resistance values; S2. Preprocess the force value, the displacement amount, and the resistance value to obtain a performance data set; S3. Perform performance evaluation on the performance data set based on the multiple regression analysis method to obtain an evaluation result; S4. Perform node construction and traceability analysis according to the process value and the quality value to obtain a node diagram and a mapping relationship table; S5. Perform model construction according to the node diagram, the mapping relationship table, and the evaluation result to obtain a prediction model; S6. Perform defect prediction according to the prediction model to obtain a prediction result; S7. Classify the defect types in the prediction result based on the clustering algorithm to obtain a standard defect library; S8. Perform feature identification according to the standard defect library to obtain a traceability identification code; S9. Perform associated mapping on the real-time process quality data and the traceability identification code to obtain defect traceability information.

[0026] In step S1, obtain process values and quality values; wherein, the quality values include force values, displacement amounts, and resistance values.

[0027] Exemplarily, the response time and pressure change data of the key are obtained through sensors to obtain the original acquisition data. The multi-channel data acquisition system is the basis for evaluating the key performance. Such a system typically includes a force sensor, a displacement sensor, and a resistance measurement device, which are respectively used to obtain the force value, displacement amount, and resistance value during the key pressing process. For example, a piezoelectric force sensor can be used to measure the pressing force, an optical encoder to measure the displacement, and a high-precision resistance measurement circuit to measure the key resistance change. Data synchronization calibration is a key step to ensure the accuracy of the acquired data. There will be slight differences in the response times of different sensors, and calibration needs to be carried out through methods such as timestamp alignment or interpolation. For instance, assuming the sampling rate of the force sensor is 1000Hz and that of the displacement sensor is 500Hz, the displacement data can be matched to the time points of the force data through linear interpolation. Time series alignment is to unify data of different dimensions onto the same time axis. This can be achieved through resampling or interpolation. For example, all data can be unified to a time interval of 10ms for subsequent analysis and feature extraction. Multi-dimensional feature extraction is to extract meaningful features from the original data. For key performance, the following features can be extracted: maximum force, force rise time, force fall time, displacement curve slope, resistance change rate, etc. These features can comprehensively reflect the touch and response characteristics of the key. Constructing a performance index evaluation system is to convert the extracted features into quantifiable performance indexes.

[0028] S2. Preprocess the force value, the displacement amount, and the resistance value to obtain a performance data set.

[0029] In step S2, preprocess the force value, the displacement amount, and the resistance value to obtain a performance data set.

[0030] In the above step S2, the preprocessing of the force value, the displacement amount, and the resistance value to obtain a performance data set specifically further includes the following steps: S21. Perform synchronization calibration according to the force value, the displacement amount, and the resistance value to obtain a calibrated data set; S22. Align the time series of the calibrated data set to obtain a sequence data set; S23. Extract features according to the sequence data set to obtain a multi-dimensional feature data set; S24. Classify the multi-dimensional feature data based on a clustering algorithm to obtain a performance data set.

[0031] It should be noted that in the above steps S21 to S24, the specific implementation process includes: using a multi-channel data acquisition system to obtain the force value, displacement amount, and resistance value during the key pressing process; performing synchronization calibration on the collected force value, displacement amount, and resistance value; aligning the calibrated force value, displacement amount, and resistance value in time series; extracting multi-dimensional features from the data after time series alignment; constructing a performance index evaluation system based on the extracted multi-dimensional features; using a clustering algorithm to classify the data in the performance index evaluation system; if the classification result meets the preset threshold, it is determined that the multi-dimensional performance index data set is constructed successfully.

[0032] Exemplarily, a multi-channel data acquisition system is used to obtain the force value, displacement amount, and resistance value of the key during the pressing process. Since there are differences in the response times of different sensors, in order to ensure the accuracy of the collected data, synchronization calibration is required. For example, the sampling rates of a piezoelectric force sensor, an optical encoder, and a high-precision resistance measurement circuit may be different. At this time, methods such as timestamp alignment or interpolation can be used to calibrate the data collected by different sensors, thereby obtaining a calibrated data set. Then enter step S22, and align the time series of the calibrated data set. In order to facilitate subsequent analysis and feature extraction, the data in different dimensions need to be unified to the same time axis. For example, all data can be unified to a time interval of 10 ms by resampling or interpolation, so as to obtain a sequence data set. Subsequently, perform step S23, and extract features based on the sequence data set. From the data of the key pressing process, multi-dimensional features that can comprehensively reflect the key touch and response characteristics are extracted. Such as the maximum force, force rise time, force fall time, displacement curve slope, resistance change rate, etc. These features constitute a multi-dimensional feature data set.

[0033] Finally, it is step S24. Based on a clustering algorithm, the multi-dimensional feature data is classified. Algorithms such as K-means or hierarchical clustering can be selected. Assuming 10 performance indicators are extracted, through clustering analysis, the key performance may be classified into different categories such as "light touch type", "hard type", and "balanced type", and then a performance data set is obtained.

[0034] In step S3, based on the multiple regression analysis method, the performance of the performance data set is evaluated to obtain an evaluation result.

[0035] In the above step S3, the performance of the performance data set is evaluated based on the multiple regression analysis method to obtain an evaluation result, which specifically further includes the following steps: S31, based on the multiple regression analysis method, construct a model for the performance data set to obtain an initial correlation model; S32. Expand the dimensions based on the initial association model by incorporating dynamic factors and interaction factors to obtain an extended association model; S33. Analyze the characteristics of the real-time process quality data according to the extended association model to obtain characteristic values; S34. Calculate the weights based on the characteristic values to obtain weight coefficients; It should be noted that the weight coefficients are calculated through the following formula: Where, represents the weight coefficient of the th index, represents the characteristic value of the th index, is the total number of items of the index.

[0036] S35. Optimize the performance data set according to the weight coefficients to obtain an optimized process parameter set; S36. Conduct performance evaluation according to the optimized process parameter set to obtain an evaluation result.

[0037] In a specific embodiment, in the above steps S31 to S36, the specific implementation process includes: using the multiple regression analysis method to obtain data from the key performance indicators and processing process parameters to establish an initial association model; in the initial association model, incorporate dynamic factors and interaction factors to expand the model dimensions and obtain an extended association model; through the extended association model, analyze the relationship between the performance indicators and process parameters to determine the key influencing factors; according to the key influencing factors, calculate the weight coefficients of each factor to obtain a weight distribution result. If the weight coefficient is greater than the preset threshold, then mark this factor as a significant influencing factor; according to the significant influencing factors, optimize the processing process parameters to obtain an optimized process parameter set. Use the optimized process parameter set to re-evaluate the key performance indicators to obtain the final performance evaluation result.

[0038] In this embodiment, the so-called dimension expansion specifically means that in order to make the model closer to the actual situation and improve the model accuracy, it is necessary to incorporate dynamic factors and interaction factors for dimension expansion. Dynamic factors, such as the influence of environmental temperature changes on material properties. Under different environmental temperatures, even if the materials and processing technologies are the same, the performance of the keys may vary. For example, in a high-temperature environment, the flexibility of the material may increase, thereby affecting the response speed and force of the keys. Interaction factors consider the combined effects between multiple process parameters, such as the interaction between material hardness and processing pressure. When the material hardness is high, the influence of processing pressure on the key performance may be different from when the material hardness is low, and there is a complex interaction relationship between the two. After incorporating these dynamic factors and interaction factors into the initial association model, the extended association model is obtained.

[0039] Exemplarily, the deployment of the distributed collector can cover key detection points on the production line, such as raw material feeding, processing, and finished product inspection. Taking automobile manufacturing as an example, in the body welding process, sensors can be installed on the robot welding torch, welding power source, and workpiece fixture to collect process values such as welding current, voltage, and speed in real time. At the same time, weld quality data can be collected at the weld inspection station. These data are transmitted in real-time streams through industrial Ethernet or 5G networks to ensure the timeliness and integrity of the data. Data cleaning and format conversion are key steps to ensure data quality. In the automobile manufacturing scenario, data problems that may be encountered include missing values, outliers, and inconsistent formats. For example, the welding current data may have zero values or outliers outside the normal range due to sensor failures, and these data need to be processed by methods such as interpolation or elimination. At the same time, the data output by different devices has different formats and needs to be uniformly converted into a standard format, such as JSON or CSV, for subsequent analysis. The construction of the node relationship model is the basis for understanding the process flow. In automobile manufacturing, each processing step can be regarded as a node, and the material flow and information transfer between the steps are used as the connections between the nodes. For example, for the whole vehicle painting process after body welding, a node chain of "welding - grinding - pretreatment - primer - topcoat - drying" can be established, and each node contains corresponding process parameters and quality indicators. The mapping relationship table between the process and quality is the core of quality traceability. Taking engine manufacturing as an example, a mapping relationship can be established between the process parameters of the cylinder block processing (such as cutting speed, feed rate) and the quality indicators of the cylinder block (such as surface roughness, dimensional accuracy). This mapping relationship helps to quickly locate the cause of quality problems and improve production efficiency. The application of the anomaly detection analyzer can timely detect and solve production problems. In electronic product manufacturing, if it is found that the welding quality of the PCB board is abnormal, it can be found through retrospective analysis whether it is caused by insufficient solder paste printing thickness or improper reflow soldering temperature curve, etc. This analysis can guide process improvement and prevent similar problems from occurring again.

[0040] In step S4, node construction and traceability analysis are performed based on the process value and the quality value to obtain a node graph and a mapping relationship table.

[0041] In the above step S4, the node construction and traceability analysis based on the process value and the quality value to obtain a node graph and a mapping relationship table specifically further includes the following steps: S41, perform data cleaning and format conversion based on the process value and the quality value to obtain standardized data; S42, store the standardized data in the distributed repository based on the preset node relationship model to obtain a node graph; S43. Based on the anomaly detection analyzer, perform anomaly analysis on the node graph. If there are abnormal data points in the node graph, perform quality traceability and root cause analysis on the abnormal data points to obtain a mapping relationship table.

[0042] In a specific embodiment, in the above steps S41 to S43, the specific implementation process includes: using a distributed collector to obtain process values and quality values from each detection point, and transmitting them to the processor in real-time stream for data cleaning and format conversion. According to the preset node relationship model, store the cleaned process values and quality values in a distributed repository to construct a complete node graph. Extract the process parameter set and quality index set from the repository, and establish a mapping relationship table between the process and quality based on the preset mapping rules. If there are abnormal data points in the node graph, start the anomaly detection analyzer to perform quality traceability and root cause analysis on the abnormal data. For example, in a process quality monitoring system, when abnormal data points appear in the node graph, the anomaly detection analyzer is immediately started to carry out quality traceability and root cause analysis work. The system will first classify the abnormal data according to the pre-set rules, such as determining whether it belongs to process parameter anomalies or product quality index anomalies. If it is a process parameter anomaly, based on the previously constructed mapping relationship table between the process and quality, combined with the process data stored in the node graph, trace back reversely along the production process to check the changes in process parameters before the abnormal data point. Exemplarily, obtaining defect data by the quality inspection system is a key link in product quality management. Taking the automotive manufacturing industry as an example, in the body welding process, solder joint defect information can be captured through a laser scanner and an image recognition system. These raw data usually contain a large amount of noise and redundant information and need to be preprocessed. The preprocessing process involves steps such as data cleaning, denoising, and standardization to ensure the accuracy of subsequent analysis. Extracting defect features is the basis for identifying and classifying defects. The features in welding defects include the size, shape, depth, etc. of the solder joints. The solder joint contour can be extracted through an edge detection algorithm, and the area and perimeter can be calculated; the depth of the solder joint can be measured through depth image analysis. These features constitute a multi-dimensional vector describing the defect. The clustering algorithm can group the defects according to the similarity of the defect features. The commonly used K-means algorithm can classify the solder joint defects into several main types, such as pores, cracks, incomplete penetration, etc. By analyzing the clustering results, the distribution of various defects can be obtained, such as pores accounting for 30%, cracks accounting for 15%, incomplete penetration accounting for 20%, etc. The determination of the defect severity is usually based on a preset standard. For example, for a solder joint crack, a length less than 1 mm is considered minor, 1 - 3 mm is medium, and more than 3 mm is severe. This grading helps to optimize resource allocation and prioritize the handling of the most serious problems.

[0043] In step S5, construct a model based on the node graph, the mapping relationship table, and the evaluation result to obtain a prediction model.

[0044] In a specific implementation, according to the mapping relation table and the node graph, a prediction model of process parameters and quality indicators is established by using the regression analysis method. Based on the prediction model and the process values collected in real time, the quality indicators are predicted online to generate a quality early warning report. The prediction results are compared and analyzed with the measured quality values, and the accuracy of the prediction model is continuously improved through the iterative optimization method.

[0045] Exemplarily, the establishment and application of the prediction model are effective means to realize the pre-positioning of quality control. In the process of steel smelting, a regression model of process parameters such as furnace temperature and feeding amount and quality indicators such as steel strength and purity can be established based on historical data.

[0046] In step S6, defect prediction is performed according to the prediction model to obtain prediction results.

[0047] Exemplarily, by inputting the current process parameters in real time, the model can predict the quality of the final product. If the prediction results do not meet the standards, the process parameters can be adjusted in time to avoid the production of unqualified products. The continuous optimization of the prediction model is the key to maintaining the effectiveness of the model. By comparing the prediction results with the actual quality inspection results, the prediction error of the model can be calculated. If the error exceeds the preset threshold, the model can be retrained using the newly added data to continuously improve the accuracy of the model. This dynamic optimization mechanism can enable the model to adapt to changes in production conditions, such as the influence of factors such as equipment wear and raw material batch changes.

[0048] In step S7, the defect types in the prediction results are classified based on the clustering algorithm to obtain a standard defect library.

[0049] In an implementable manner, the classification of the defect types in the prediction results based on the clustering algorithm to obtain a standard defect library specifically includes: using the clustering algorithm to classify the extracted defect features to obtain the defect type distribution. According to the defect type distribution and the preset severity standard, the severity levels of various defects are determined. For the defect types with different severity levels, the preset treatment schemes are matched. The defect features, defect types, severity levels, and treatment schemes are integrated into structured data and stored in the standard defect library. Through the data in the standard defect library, the parameter settings of the clustering algorithm are optimized. Based on the optimized clustering algorithm, the defect classification model is updated to improve the classification accuracy.

[0050] In step S8, feature identification is performed according to the standard defect library to obtain a traceability identification code.

[0051] In the above step S8, the feature identification according to the standard defect library to obtain a traceability identification code specifically further includes the following steps: S81, calculating the index weight value based on the analytic hierarchy process for the standard defect library; S82. Calculate the priority score of the defect according to the index weight value to obtain a score value; S83. Sort the score values from high to low to obtain a defect handling sequence; S84. Encode the defect handling sequence based on the pre-stored defect coding rule to obtain a traceability identification code.

[0052] It should be noted that in the processes of the above steps S81 to S84, the specific implementation includes: obtaining the original defect data according to the standard defect library, including the defect occurrence frequency, impact degree rating, and repair cost estimation, constructing a third-order weight matrix according to the analytic hierarchy process, and calculating the weight values of each index. Using the weight values to calculate the priority score of each defect, sorting them from high to low according to the score values, and generating a defect handling sequence. For the quality data of different processes, formulate a unified defect coding rule, and map the defect features to a unique identification code. Classify and store the quality data according to the defect coding, establish a structured defect information library, and set up an effective indexing mechanism. For the retrieval requirements of defect information, adopt the inverted index technology to construct the mapping relationship between defect features and storage locations. During the retrieval process, if the input defect features are provided, the storage location of the defect information can be quickly located through the inverted index to query the relevant data. For the newly generated defect data, encode it according to the defect coding rule and update it to the defect information library to maintain data consistency. Exemplarily, deploying data acquisition devices on the production line is the key to realizing quality control. Taking automobile manufacturing as an example, installing sensors and scanning devices at key process points such as body welding, painting, and final assembly can collect quality data such as vehicle identification numbers, the number of weld points, and coating thickness in real time. These data are associated with the process points to form a product quality file. When judging the defect source through a preset threshold, a welding strength lower than 800N or a coating thickness deviation exceeding ±0.1mm can be defined as a defect. Once an anomaly is detected, the system immediately records the defect information and the time stamp, providing a basis for subsequent analysis. Defect transmission chain analysis is crucial for improving production efficiency. Taking body welding defects as an example, if they are not discovered and corrected in time, it will lead to chain reactions such as uneven paint surfaces in the subsequent painting process and abnormal door noises in the final assembly process. By establishing the mapping relationship between the defect transmission chain and the treatment points, timely intervention can be carried out before the problem expands. The application of association rule algorithms helps to optimize defect treatment strategies. For example, it is found through analysis that there is a strong correlation between welding current and weld point strength, and the defect rate can be effectively reduced by adjusting the welding parameters. This method can not only improve the pertinence of treatment but also achieve defect traceability and prevention. Constructing a time series data model for the entire life cycle of defects can clearly show the complete process of defects from generation to resolution. For example, recording the discovery time, rework time, and re-inspection time of welding defects to ensure the integrity of the information chain, which helps to analyze defect treatment efficiency and identifying. Defect classification and statistics are important means to optimize the production process. Classification can be carried out according to defect types (such as welding, painting, assembly, etc.) and severity levels (minor, general, serious) to generate a priority list. This helps to reasonably allocate resources and prioritize the treatment of defects with greater impacts. The application of multi-source data fusion technology makes the information chain of the entire life cycle of defects more complete.Integrate process point data (such as welding station number), quality data (weld strength), defect source information (welding defects), transfer chain data (affecting subsequent painting quality), etc. to form a comprehensive quality control system.

[0053] In step S9, associate and map the real-time process quality data with the traceability identification code to obtain defect traceability information.

[0054] In the above step S9, the operation of associating and mapping the real-time process quality data with the traceability identification code to obtain defect traceability information specifically further includes the following steps: S91, judge according to the real-time process quality data and the preset threshold range; if the real-time process quality data exceeds the preset threshold range, it is determined that there is a defect source; S92, retrieve the defect source according to the traceability identification code to obtain the storage location and time stamp of the defect information; S93, output the storage location of the defect information and the time stamp to obtain defect traceability information.

[0055] Exemplarily, taking automobile manufacturing as an example, sensors and scanning devices are installed at key process points such as body welding, painting, and final assembly to collect quality data such as vehicle identification numbers, the number of weld points, and coating thickness in real time. These data are associated with the process points to form a product quality file. When judging the defect source through a preset threshold, a welding strength lower than 800 N or a coating thickness deviation exceeding ±0.1 mm can be defined as a defect. Once an anomaly is detected, the system immediately records the defect information and the time stamp to provide a basis for subsequent analysis. Defect transfer chain analysis is crucial for improving production efficiency. Taking body welding defects as an example, if they are not discovered and corrected in time, it will lead to chain reactions such as uneven paint surfaces in the subsequent painting process and abnormal door noises in the final assembly process. By establishing the mapping relationship between the defect transfer chain and the governance points, timely intervention can be carried out before the problem expands. The application of association rule algorithms helps optimize defect governance strategies. For example, it is found through analysis that there is a strong correlation between welding current and weld point strength, and the defect rate can be effectively reduced by adjusting welding parameters. This method can not only improve the pertinence of governance but also achieve defect traceability and prevention. Constructing a time-series data model for the entire life cycle of defects can clearly show the complete process from the generation to the resolution of defects. For example, recording the discovery time, rework time, and re-inspection time of welding defects to ensure the integrity of the information chain helps analyze defect handling efficiency and identifying. Defect classification and statistics are important means to optimize the production process. Classification can be carried out according to defect types (such as welding, painting, assembly, etc.) and severity levels (minor, general, serious) to generate a priority list. This helps allocate resources reasonably and prioritize the handling of defects with greater impacts. The application of multi-source data fusion technology makes the information chain of the entire life cycle of defects more complete. Integrating process point data (such as welding station numbers), quality data (weld point strength), defect source information (welding defects), transfer chain data (affecting subsequent painting quality), etc. forms a comprehensive quality control system.

[0056] In summary, the present invention discloses a method for defect traceability in the processing of watch buttons, which includes obtaining process values and quality values; wherein, the quality values include force values, displacement amounts, and resistance values; preprocessing the force values, the displacement amounts, and the resistance values to obtain a performance data set; performing performance evaluation on the performance data set based on the multiple regression analysis method to obtain an evaluation result; performing node construction and traceability analysis according to the process values and the quality values to obtain a node diagram and a mapping relation table; constructing a prediction model according to the node diagram, the mapping relation table, and the evaluation result; performing defect prediction according to the prediction model to obtain a prediction result; classifying the defect types in the prediction result based on a clustering algorithm to obtain a standard defect library; performing feature identification according to the standard defect library to obtain a traceability identification code; and performing associated mapping on the real-time process quality data and the traceability identification code to obtain defect traceability information. The present invention establishes a performance index quantization model based on physical characteristics, collects button response time and pressure change data, and constructs a multi-dimensional performance index data set; adopts multiple regression analysis to establish an association model between button performance and process parameters, and determines key influencing factors; the present invention deploys on-line detection devices at each processing node, collects process parameters and quality data in real time, and constructs a node diagram and the corresponding relationship between the process and quality indexes. Based on the detection results, defects are classified and a standard defect library is established; a defect priority evaluation model is constructed through the analytic hierarchy process, and a disposal sequence list is generated. The present invention also designs a defect tracking mechanism based on the traceability code to realize the information chain of the entire life cycle of defects, and implements a quality data analysis and early warning system to provide decision-making support for quality improvement; the present invention can effectively improve the button performance, reduce the defect rate, and realize the whole-process control of product quality.

[0057] Referring to Figure 2 , the second embodiment of the present invention provides a system for defect traceability in the processing of watch buttons, including: A data acquisition module 101, configured to acquire process values and quality values; wherein, the quality values include force values, displacement amounts, and resistance values; A data cleaning module 102, configured to preprocess the force values, the displacement amounts, and the resistance values to obtain a performance data set; A performance evaluation module 103, configured to perform performance evaluation on the performance data set based on the multiple regression analysis method to obtain an evaluation result; A node construction and analysis module 104, configured to perform node construction and traceability analysis according to the process values and quality values to obtain a node diagram and a mapping relation table; A model construction module 105, configured to construct a prediction model according to the node diagram, the mapping relation table, and the evaluation result; A defect prediction module 106, configured to perform defect prediction according to the prediction model to obtain a prediction result; A defect classification module 107, configured to classify the defect types in the prediction result based on a clustering algorithm to obtain a standard defect library; A defect coding module 108, configured to perform feature identification according to the standard defect library to obtain a traceability identification code; An output module 109, configured to perform an association mapping between the real-time process quality data and the traceability identification code to obtain defect traceability information.

[0058] It should be noted that the system for defect traceability in the processing of watch buttons provided in the embodiment of the present invention is used to execute all the process steps of the method for defect traceability in the processing of watch buttons in the above embodiment. The working principles and beneficial effects of the two correspond one by one, so they will not be elaborated here.

[0059] The embodiment of the present invention also provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a program for defect traceability in the processing of watch buttons. When the processor executes the computer program, the steps in the above-mentioned method embodiments for defect traceability in the processing of watch buttons are implemented, such as Figure 1 the step S1 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented, such as the data acquisition module.

[0060] Exemplarily, the computer program can be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0061] The electronic device can be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device, and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0062] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and circuits.

[0063] The memory can be used to store the computer program and / or module. The processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0064] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0065] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative work.

[0066] The above-described specific embodiments have further elaborated on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for tracing defects in the process of watch button processing, characterized in that: Executed by a computer, including: Obtaining process values ​​and quality values; wherein the quality values ​​include force values, displacement values ​​and resistance values; Preprocessing the force value, the displacement value and the resistance value to obtain a performance data set; Performing performance evaluation on the performance data set based on a multiple regression analysis method to obtain an evaluation result; Perform node construction and traceability analysis according to the process value and the quality value to obtain a node graph and a mapping relationship table; Constructing a model according to the node graph, the mapping relationship table and the evaluation result to obtain a prediction model; Perform defect prediction according to the prediction model to obtain a prediction result; Classifying the defect types in the prediction results based on a clustering algorithm to obtain a standard defect library; performing feature identification according to the standard defect library to obtain a traceability identification code; The real-time process quality data is associated and mapped with the traceability identification code to obtain defect traceability information.

2. The method for tracing defects in the watch button processing process according to claim 1, characterized in that: The preprocessing of the force value, the displacement value and the resistance value to obtain a performance data set includes: Perform synchronization calibration according to the force value, the displacement and the resistance value to obtain a calibration data set; Performing time series alignment on the calibration data set to obtain a sequence data set; Perform feature extraction based on the sequence data set to obtain a multi-dimensional feature data set; The multi-dimensional feature data is classified based on a clustering algorithm to obtain a performance data set.

3. The method for tracing defects in the watch button processing process according to claim 1, characterized in that: The performance evaluation of the performance data set is performed based on the multivariate regression analysis method to obtain the evaluation result, including: Building a model for the performance data set based on a multiple regression analysis method to obtain an initial correlation model; Based on the initial correlation model, dynamic factors and interactive factors are incorporated to perform dimension expansion to obtain an extended correlation model; Performing feature analysis on real-time process quality data according to the extended association model to obtain feature values; Perform weight calculation according to the characteristic value to obtain a weight coefficient; Optimizing the performance data set according to the weight coefficient to obtain an optimized process parameter set; A performance evaluation is performed according to the optimized process parameter set to obtain an evaluation result.

4. The method for tracing defects in the watch button processing process according to claim 3, characterized in that: The weight calculation is performed according to the characteristic value to obtain the weight coefficient, including: The weight coefficient is calculated by the following formula: in, Indicates The weight coefficient of each indicator, Indicates The characteristic value of the indicator, is the total number of indicators.

5. The method for tracing defects in the watch button processing process according to claim 1, characterized in that: The node construction and traceability analysis are performed according to the process value and the quality value to obtain a node diagram and a mapping relationship table, including: Perform data cleaning and format conversion according to the process value and quality value to obtain standardized data; Based on a preset node relationship model, the standardized data is stored in a distributed repository to obtain a node graph; An anomaly detection analyzer is used to perform an anomaly analysis on the node graph. If an abnormal data point exists in the node graph, quality tracing and root cause analysis are performed on the abnormal data point to obtain a mapping relationship table.

6. The method for defect tracing in the watch button processing process according to claim 1, characterized in that: The step of performing feature identification according to the standard defect library to obtain a traceability identification code includes: Based on the hierarchical analysis method, weight calculation is performed on the standard defect library to obtain the indicator weight value; Calculate the priority score of the defects according to the indicator weight value to obtain a score value; Sort the score values ​​from high to low to obtain a defect processing sequence; The defect processing sequence is encoded based on a pre-stored defect encoding rule to obtain a traceability identification code.

7. The method for defect tracing in the watch button processing process according to claim 1, characterized in that: The real-time process quality data is associated and mapped with the traceability identification code to obtain defect traceability information, including: Making a judgment based on the real-time process quality data and the preset threshold range; if the real-time process quality data exceeds the preset threshold range, it is determined that a defect source exists; Retrieve the defect source according to the traceability identification code to obtain the storage location and timestamp of the defect information; The defect information storage location and the timestamp are output to obtain defect tracing information.

8. A diesel generator set fault prediction model, characterized in that: include: A data acquisition module, used to acquire process values ​​and quality values; wherein the quality values ​​include force values, displacement values ​​and resistance values; A data cleaning module, used for preprocessing the force value, the displacement value and the resistance value to obtain a performance data set; A performance evaluation module, used to perform performance evaluation on the performance data set based on a multiple regression analysis method to obtain an evaluation result; A node construction and analysis module is used to perform node construction and traceability analysis according to the process value and quality value to obtain a node diagram and a mapping relationship table; A model building module, used to build a model according to the node graph, the mapping relationship table and the evaluation result to obtain a prediction model; A defect prediction module, used to perform defect prediction according to the prediction model to obtain a prediction result; A defect classification module, used to classify the defect types in the prediction results based on a clustering algorithm to obtain a standard defect library; A defect coding module, used to perform feature identification according to the standard defect library to obtain a traceability identification code; The output module is used to associate and map the real-time process quality data with the traceability identification code to obtain defect traceability information.

9. An electronic device, characterized in that: It comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the method for defect tracing in the watch button processing process as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for defect tracing in the watch button processing process as described in any one of claims 1 to 7.

Images