Digital visual management system and method for accessory production and processing

Abstract

The invention relates to the technical field of intelligent manufacturing, in particular to a digital visual management system and method for accessory production and processing. The method comprises the following steps: acquiring accessory production and processing data, and performing production process feature extraction according to the accessory production and processing data so as to obtain production process data; performing accessory quality defect analysis on the production process data to obtain accessory quality defect data; carrying out production equipment state feature extraction based on the accessory production processing data so as to obtain equipment state data; performing association analysis on the accessory quality defect data according to the equipment state data to obtain equipment state-accessory defect association data; and constructing a quality defect prediction model according to the equipment state-accessory defect associated data so as to obtain the quality defect prediction model. Based on the intelligent manufacturing technology, the accessory production efficiency and the quality qualification rate are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a digital visualization management system and method for parts production and processing. Background Art

[0002] Traditional systems usually have data islands, and data from different production links are difficult to share and communicate, resulting in poor information flow and delayed decision-making. Traditional methods rely on manual experience and rules to make decisions, lack intelligent analysis and real-time feedback, and are difficult to respond quickly to changes in production, resulting in low decision-making efficiency. Traditional methods are unable to cope with complex data associations, and it is difficult to conduct in-depth correlation analysis of multi-dimensional data such as equipment status, production processes, and raw materials, and it is impossible to fully identify potential problems. Traditional systems lack predictive capabilities and can usually only handle problems that have already occurred. They cannot provide early warnings or predictive maintenance, thus missing the best opportunity for improvement. The traditional system's level of refined management is insufficient, and it is difficult to accurately track every piece of data in the production process, resulting in unreasonable resource allocation and difficulty in achieving maximum optimization effects. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a digital visualization management system and method for accessory production and processing to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above purpose, a digital visualization management method for parts production and processing includes the following steps:

[0005] Step S1: acquiring accessory production and processing data, and extracting production process features based on the accessory production and processing data, thereby obtaining production process data; performing accessory quality defect analysis on the production process data, thereby obtaining accessory quality defect data;

[0006] Step S2: extracting production equipment status features based on accessory production and processing data to obtain equipment status data; performing correlation analysis on accessory quality defect data based on the equipment status data to obtain equipment status-accessory defect correlation data;

[0007] Step S3: construct a quality defect prediction model based on the equipment status-accessory defect association data, thereby obtaining a quality defect prediction model; perform intelligent warning of accessory abnormalities on the accessory production and processing data based on the quality defect prediction model, thereby obtaining intelligent warning data of accessory abnormalities;

[0008] Step S4: Optimize the production scheduling of the parts production and processing data according to the intelligent warning data of the parts abnormality to obtain the production scheduling optimization data, and update the parts production and processing data according to the production scheduling optimization data to obtain the production process update data; upload the production process update data to the digital production management platform to perform digital visualization management tasks.

[0009] The present invention can accurately reflect the key process parameters in the production process by acquiring the production and processing data of the accessories and extracting the production process characteristics, and provide a solid data foundation for subsequent quality control and production optimization. When analyzing the quality defects of accessories, it can fully identify the quality problems that occur in the production process, discover potential defects in advance and provide data support for optimization, avoiding the blindness of traditional methods that rely only on manual experience and rules. In addition, by extracting the characteristics of the equipment status and performing correlation analysis with the quality defects of the accessories, the inherent relationship between equipment operation and product quality can be fully explored, solving the problems of data island phenomenon and poor information flow in traditional methods, so that data from different production links can be effectively integrated and interoperable, and the timeliness and accuracy of production decisions are improved. In the construction and application of the quality defect prediction model, combined with the equipment status and production and processing data, the intelligent early warning function can be realized, and potential problems in the production of accessories can be predicted in advance, avoiding the shortcomings of the traditional system that cannot perform predictive maintenance. This intelligent early warning system can dynamically respond to production changes, effectively reduce delays and resource waste in the production process, and improve production efficiency and equipment utilization. Combined with production scheduling optimization, it can perform intelligent scheduling based on real-time data, adjust production tasks and resource allocation, and avoid the inefficiency of traditional methods that lack real-time feedback and scheduling optimization. By uploading the updated production process data to the digital production management platform, it is possible to achieve centralized management and visual display of data on the platform, ensuring that every data in the production process is accurately tracked, promoting the rational allocation and refined management of resources, and thus achieving the maximum optimization effect. Through these beneficial effects, the entire production process has achieved a comprehensive improvement from information flow, decision-making efficiency, data correlation analysis, early warning capabilities to refined management, greatly improving the intelligence level of the production system and resource utilization efficiency.

[0010] Preferably, step S1 specifically comprises:

[0011] Step S11: acquiring accessory production and processing data, and extracting production process features according to the accessory production and processing data, thereby obtaining production process data;

[0012] Step S12: Analyze the surface defects of the parts on the production process data, so as to obtain the surface defect data of the parts;

[0013] Step S13: performing accessory size deviation defect analysis on the production process data, thereby obtaining accessory size deviation defect data;

[0014] Step S14: Integrate the accessory quality defects according to the accessory surface defect data and the accessory size deviation defect data, so as to obtain the accessory quality defect data.

[0015] The present invention can accurately capture key process information in the production process by acquiring accessory production and processing data and extracting production process characteristics, eliminate the data island phenomenon in the traditional system, and realize data sharing and intercommunication between production links. This process makes data flow smooth, can improve the speed of information transmission, and reduce decision-making delays caused by data lag. In the analysis of accessory surface defects and dimensional deviation defects, accurate analysis based on process data can timely discover quality problems of accessories in the production process, such as surface flaws or dimensional non-compliance, avoid the limitations of quality inspection relying on manual experience in traditional methods, and improve the accuracy and efficiency of quality control. By integrating surface defects and dimensional deviation defect data, potential problems of accessory quality can be fully identified, providing strong data support for subsequent quality improvement and production optimization. Not only does it optimize the bottlenecks of low decision-making efficiency and lack of real-time feedback in traditional methods, but it also solves the problem that traditional systems cannot perform deep correlation analysis on multi-dimensional data such as equipment status, production process and raw materials, so as to fully identify potential production problems and equipment failure risks. Through the integrated analysis of quality defect data, early warning can be given before problems occur, and production plans or processes can be adjusted in time, thereby reducing quality problems in the production process and improving product qualification rates. This intelligent management method can track every data in the production process in real time, ensure the rational allocation of resources, optimize the production process through data-driven decision-making, and maximize production efficiency and resource utilization. Through these beneficial effects, each link in the production process is precisely controlled, and the intelligence level of the overall production system is significantly improved.

[0016] Preferably, step S12 is specifically as follows:

[0017] Step S121: Obtain industrial accessories;

[0018] Step S122: dividing the industrial accessories based on the production process data, thereby obtaining fine process industrial accessories and coarse process industrial accessories;

[0019] Step S123: performing crack defect detection on the fine craft industrial accessories to obtain crack defect data;

[0020] Step S124: performing burr defect detection on rough process industrial accessories to obtain burr defect data;

[0021] Step S125: integrating the surface defects of the accessory according to the crack defect data and the burr defect data, thereby obtaining the surface defect data of the accessory.

[0022] The present invention can finely divide accessories into fine process and rough process industrial accessories by acquiring industrial accessories and dividing them based on production process data. This process breaks the data island phenomenon in the traditional system, enables the accessory data of different process types to be clearly classified, promotes data sharing and circulation, thereby improving the speed of information flow and reducing decision-making delays caused by data lag. Crack defect detection is performed on fine process industrial accessories and burr defect detection is performed on rough process industrial accessories. Through accurate defect detection methods, subtle problems that affect the quality of accessories in the production process can be discovered in time. This intelligent defect detection method replaces the traditional detection method that relies on manual experience and rules, can effectively improve the accuracy and efficiency of detection, and ensure that the produced accessories meet strict quality standards. The integration of crack defect data and burr defect data further enhances the comprehensiveness of quality analysis and helps identify all surface defects that occur in accessories during the production process. Deep data analysis and intelligent processing are realized, which not only overcomes the bottleneck of lack of real-time feedback and inability to accurately identify potential problems in traditional systems, but also provides detailed support for the optimization of production processes and quality management. Through comprehensive defect detection and integration, problems can be discovered and adjusted in a timely manner, thereby reducing production stagnation and resource waste caused by quality defects and improving production efficiency. Overall, after implementing these steps, data sharing and accurate decision-making capabilities in the production process have been significantly improved, the intelligence level of the system has been improved, and the optimization of production management efficiency and resource allocation has also been achieved.

[0023] Preferably, step S123 is specifically as follows:

[0024] Apply penetrant to fine craft industrial accessories to obtain penetrant application industrial accessories;

[0025] Surface penetrant removal is performed according to the penetrant application industrial accessories to obtain surface penetrant removal industrial accessories;

[0026] Evenly spray the developer on the surface penetrant removal industrial accessories and collect data to obtain the developer spraying data;

[0027] Based on the developer spraying data, the crack appearance area of ​​the surface penetrant removal industrial accessories is identified to obtain the crack appearance area data;

[0028] Perform crack length statistics on the crack appearance area data to obtain long crack data;

[0029] Based on the long crack data, stress concentration analysis is performed on precision industrial parts to obtain long crack stress concentration data;

[0030] Based on the long crack data, fatigue crack growth simulation is performed on the precision industrial parts to obtain fatigue crack growth data;

[0031] The influence of long crack defects on fatigue crack growth data is analyzed based on long crack stress concentration data to obtain crack defect data.

[0032] The present invention applies a penetrant to the accessories of the fine craft industry, and the sensitivity of crack detection can be effectively enhanced by applying the penetrant. Next, the surface penetrant is removed from the accessories after the penetrant is applied to ensure the cleanliness and accuracy of the detection area. After the developer is evenly sprayed and data collection is performed, the tiny cracks on the surface of the accessories can be revealed through the action of the developer, thereby providing accurate data support for further crack detection. In this process, through data collection, key parameters in the spraying process, such as spraying amount, time, etc., can be accurately recorded to provide an accurate data source for subsequent analysis. Based on the developer spraying data, the crack appearance area is identified, and the location and distribution of the crack can be accurately calibrated, which provides accurate basic data for subsequent crack length statistics. Crack length statistics are performed on the crack appearance area data, which can effectively screen out long cracks, and then stress concentration analysis is performed on these cracks. By acquiring stress concentration data, the potential danger area of ​​the crack can be identified, providing key parameters for subsequent fatigue crack extension simulation. Using long crack data to simulate fatigue crack growth of precision industrial parts can simulate how cracks grow during actual use and identify potential problems that lead to major failures in advance. This simulation analysis can provide a scientific basis for preventive maintenance and process optimization. By combining long crack stress concentration data with fatigue crack growth data, the impact of crack defects on the performance of parts can be accurately evaluated, and more refined production process adjustments and quality control strategies can be adopted. Through precise detection, data collection and simulation analysis, the bottlenecks of data islands and insufficient real-time feedback in traditional parts production management have been effectively broken through, and comprehensive identification, prediction and response to parts defects have been achieved. Through this intelligent management method, not only production efficiency is improved, but also potential quality problems can be discovered in advance, which greatly reduces production stagnation and resource waste, and improves the level of refined management of the production process.

[0033] Preferably, step S124 is specifically as follows:

[0034] Capturing images of crude industrial parts, thereby obtaining images of crude industrial parts;

[0035] Performing image enhancement on the rough process industrial parts image, thereby obtaining a rough process industrial parts enhanced image;

[0036] Binarization conversion is performed according to the rough process industrial parts enhanced image, so as to obtain a binary industrial parts enhanced image;

[0037] Based on the preset burr threshold data, the burr area is identified on the binary industrial parts enhanced image, so as to obtain the burr area image;

[0038] Performing burr contour detection on the burr area image to obtain burr contour data;

[0039] Determine the burr position of the burr area image according to the burr contour data, thereby obtaining the burr position data;

[0040] According to the burr position data, the rough process industrial accessories are assembled and simulated to obtain assembly simulation data;

[0041] The burr defect impact assessment is performed based on assembly simulation data to obtain burr defect data.

[0042] The present invention can accurately obtain the surface features of the accessories by performing image acquisition on the rough process industrial accessories, and provide key data for subsequent quality analysis. Image enhancement of the rough process industrial accessories image can improve the clarity and details of the image, making the subsequent image processing more accurate. By performing binary conversion on the enhanced image, the key features of the accessory surface can be effectively highlighted, unnecessary noise can be eliminated, and the foundation for the accurate identification of the burr area can be laid. The burr area identification of the binary image based on the preset burr threshold data can accurately locate the area where the burr exists, thereby providing an important basis for the subsequent burr contour detection. The extraction of burr contour data can clarify the shape, size and position characteristics of the burr on the accessory surface, thereby providing accurate data for the refined analysis of the burr. By determining the burr position data, the specific position of the burr on the accessory can be further confirmed, providing reliable input for assembly simulation. Assembly simulation based on the burr position data can predict the interference and influence caused by the burr in the actual assembly process, and identify potential accessory assembly problems in advance. This simulation analysis provides a scientific basis for optimizing production processes and improving product quality. At the same time, combining assembly simulation data to evaluate the impact of burr defects can help to comprehensively evaluate the impact of burrs on the quality and performance of accessory assembly, thereby providing decision support for subsequent quality improvement and production scheduling. It can not only improve the image processing technology in accessory production, but also effectively solve the problems of inaccurate image processing, lack of real-time feedback and early warning capabilities in traditional methods. The implementation of this series of steps can strengthen the refined management of the production process, identify and solve potential defects in advance, improve production efficiency and product quality, and thus reduce waste and failure rates in production.

[0043] Preferably, step S25 is specifically as follows:

[0044] Step S131: extracting aperture process features from production process data to obtain aperture process data;

[0045] Step S132: obtaining standard aperture size data;

[0046] Step S133: Calculating the size deviation of the aperture process data according to the standard aperture size data, thereby obtaining size deviation data;

[0047] Step S134: identifying the size deviation location of the fine craft industrial accessories according to the size deviation data, thereby obtaining the size deviation location of the accessories;

[0048] Step S135: performing assembly simulation according to the size deviation part of the accessory, thereby obtaining assembly simulation data of the size deviation part of the accessory;

[0049] Step S136: evaluating the strength defect of the accessory according to the assembly simulation data of the accessory size deviation position, thereby obtaining the strength defect data of the accessory;

[0050] Step S137: performing accessory durability prediction according to the accessory size deviation position assembly simulation data, thereby obtaining accessory durability data;

[0051] Step S138: judging the impact of the size deviation defect on the accessory durability data based on the accessory strength defect data, thereby obtaining the accessory size deviation defect data.

[0052] The present invention can accurately obtain detailed data of the accessories in terms of aperture process by extracting aperture process characteristics from production process data, providing a basis for subsequent quality control. After obtaining the standard aperture size data, it can be compared with the aperture process data to calculate the size deviation. This process helps to accurately identify the size deviation of the accessories in aperture processing, and provides data support for optimizing the process and quality control. By analyzing the size deviation data, the size deviation part of the accessories can be accurately identified, providing valuable information for accurate quality management and subsequent processing. The identification of the size deviation part of the accessories provides more detailed data support for assembly simulation, ensuring that errors in the assembly process can be discovered and adjusted in time. Assembly simulation can predict the performance of accessories in the assembly process, thereby identifying existing assembly problems and formulating solutions in advance for problems on the production line. According to the assembly simulation data of the size deviation part of the accessories, the strength defect assessment of the accessories is carried out, which helps to analyze the structural problems of the accessories in the case of size deviation. This assessment can clarify whether there is a potential risk of insufficient strength of the accessories, and then provide a decision-making basis for production adjustment. The durability prediction of accessories can evaluate the long-term performance of accessories based on these data, judge the failure and life cycle of accessories during long-term use in advance, and provide more accurate early warning for production and maintenance. By combining the analysis of durability data with the strength defect data of accessories, the specific impact of dimensional deviation on the long-term use effect of accessories can be further evaluated. This judgment provides a more refined basis for quality control in production, so that potential defects can be reduced by adjusting the production process in a timely manner. The implementation of this series of steps can solve the problems of data silos and lack of real-time feedback in traditional methods, and effectively improve production efficiency, product quality and the timeliness of production decisions through precise multi-dimensional data analysis and predictive maintenance, thereby achieving maximum resource optimization and defect prevention in the production process.

[0053] Preferably, step S2 specifically comprises:

[0054] Step S21: extracting production equipment status features based on accessory production and processing data, thereby obtaining equipment status data;

[0055] Step S22: Identify the device abnormality code according to the device status data, thereby obtaining the device abnormality code data;

[0056] Step S23: identifying mechanical abnormal equipment according to the equipment abnormality code data, thereby obtaining mechanical abnormal equipment;

[0057] Step S24: performing abnormal state time statistics on mechanical abnormal equipment, thereby obtaining equipment abnormal state time data;

[0058] Step S25: performing defect detection time statistics on the accessory quality defect data, thereby obtaining defect detection time data;

[0059] Step S26: Perform time-related intersection operations based on the equipment abnormal state time data and the defect detection time data to obtain equipment state-accessory defect correlation data.

[0060] The present invention can fully understand the current working state of the equipment by extracting the state features of the production equipment based on the production and processing data of the accessories, and provide a data basis for further maintenance and optimization. The acquisition of equipment state data provides real real-time data support for subsequent abnormal detection and fault analysis. The equipment abnormal code recognition process helps to quickly capture the preliminary signal of equipment failure, thereby effectively identifying the potential abnormal problems of the equipment and providing a basis for troubleshooting and timely intervention. The identification of mechanical abnormal equipment can accurately locate the specific equipment with faults based on the abnormal code data, provide detailed fault information for maintenance personnel, help improve maintenance efficiency, and reduce unnecessary downtime. The statistics of abnormal state time not only help evaluate the frequency of equipment abnormalities, but also analyze the working time of the equipment in abnormal state, providing a reliable basis for the maintenance and repair plan of the equipment. Through the statistics of defect detection time data, accurate time tracking can be provided for accessory quality control, helping to identify the specific links where quality defects occur. This statistical data supports the rapid discovery of quality problems, accelerates the implementation of corrective measures, and improves the product qualification rate. By performing time-related intersection operations on equipment abnormal state time data and defect detection time data, the intrinsic connection between equipment failure and quality problems can be deeply excavated. This correlation data can reveal the potential impact of equipment failure on parts quality, provide strong support for optimizing production processes and maintenance plans, and further improve production efficiency and product quality. It also helps to provide early warning of equipment failures and quality defects, reduce downtime, and avoid production delays, thereby achieving more intelligent and refined management.

[0061] Preferably, step S3 specifically comprises:

[0062] Step S31: constructing a quality defect prediction model according to the equipment status-accessory defect association data, thereby obtaining a quality defect prediction model;

[0063] Step S32: Predicting quality defects of the parts production and processing data according to the quality defect prediction model, thereby obtaining quality defect prediction data;

[0064] Step S33: performing confusion matrix analysis based on the quality defect prediction data, thereby obtaining quality defect prediction confusion matrix data;

[0065] Step S34: performing prediction error analysis on the quality defect prediction data according to the quality defect prediction confusion matrix data, thereby obtaining prediction error data;

[0066] Step S35: Correcting the quality defect prediction model according to the prediction error data, thereby obtaining a quality defect prediction correction model;

[0067] Step S36: Perform intelligent warning of accessory abnormality on accessory production and processing data according to the quality defect prediction and correction model, thereby obtaining intelligent warning data of accessory abnormality.

[0068] The present invention constructs a quality defect prediction model based on the equipment status-accessory defect association data, and can use deep learning and data mining technology to provide early warning for potential quality problems in the production process to avoid further expansion of the problem. The quality defect prediction model provides intelligent support for production decisions, so that potential quality defects can be identified in advance during the production process, resource allocation can be optimized, and overall production efficiency can be improved. Using this model to predict quality defects in accessory production and processing data helps to improve production quality and pass rate, reduce the flow of unqualified products into the market, and reduce the cost of rework and maintenance. By performing confusion matrix analysis, the classification effect of the quality defect prediction model can be understood in detail, and indicators such as the accuracy, recall rate, precision and F1 score of the model can be evaluated to help identify potential deviations or deficiencies in the model prediction, thereby providing a basis for model optimization. Based on prediction error analysis, it is possible to find out the links where the model prediction is inaccurate, specifically analyze the error type, identify the cause and prescribe the right medicine, effectively improve the accuracy of the model, reduce the error in the prediction, and improve the overall performance of the system. By correcting the quality defect prediction model based on the prediction error data, the accuracy of the prediction model can be further improved by adjusting the model parameters, retraining or introducing new feature data, ensuring that it can run stably and provide reliable early warnings during the production process. Finally, intelligent early warning of the production and processing data of accessories based on the corrected model can timely detect quality problems and provide targeted improvement suggestions, prevent abnormal situations in production in advance, and improve the response speed and flexibility of the production line, thereby improving product quality, reducing resource waste, and optimizing the entire production process.

[0069] Preferably, step S4 is specifically:

[0070] Step S41: performing abnormal warning classification according to the abnormal intelligent warning data of the accessories, thereby obtaining abnormal surface defect warning data and abnormal dimension deviation warning data;

[0071] Step S42: Perform quality control processing based on the surface defect abnormality warning data to obtain quality control data;

[0072] Step S43: dynamically allocating production tasks based on the abnormal dimension deviation warning data to obtain dynamic production task allocation data;

[0073] Step S44: updating the accessory production and processing data according to the quality control data and the production task dynamic allocation data to obtain the production process update data;

[0074] Step S45: Upload the production process update data to the digital production management platform to perform digital visualization management tasks.

[0075] The present invention can effectively distinguish different types of production anomalies, such as surface defects and dimensional deviations, by classifying abnormal warning data of accessory abnormal intelligent warning data, so that managers can intervene and deal with different problems more accurately and avoid waste of resources. The distinction between surface defect abnormal warning data and dimensional deviation abnormal warning data helps to optimize subsequent production decisions and ensure that the most appropriate measures are taken for each type of defect in the production process. Quality control processing based on surface defect abnormal warning data can timely identify quality problems in the production process and carry out targeted intervention. This can not only improve the qualified rate of products, but also reduce production costs and avoid waste of time and resources caused by rework of defective products. At the same time, the generation of quality control data also provides data support for future quality improvement and helps to form a closed-loop quality management system. Based on the dimensional deviation abnormal warning data, production tasks are dynamically allocated, and production tasks can be flexibly adjusted according to abnormal conditions in different links of production, making the production process more efficient. By dynamically allocating tasks, it can ensure that key process links are given priority, thereby reducing production delays caused by dimensional deviations and ensuring production progress and delivery period. Dynamically allocated production task data provides a strong guarantee for improving production efficiency and reduces the burden of manual decision-making to a certain extent. Updating the production and processing data of accessories according to the quality control data and the dynamic allocation data of production tasks will help to achieve continuous optimization of the production process. The updated production process data can not only adjust the production parameters according to the real-time feedback, but also further improve the production efficiency, reduce the waste of resources, and enhance the responsiveness of the production system to changes. Through process updates, bottlenecks and potential problems in the production process can be solved in advance, thereby maintaining the smooth operation of the production line. Uploading the production process update data to the digital production management platform not only makes the information flow smoother, but also makes each production link more transparent. By integrating real-time data, the digital management platform can respond quickly and achieve collaborative optimization of the production links. At the same time, the platform's execution of digital visualization management tasks provides technical support for the monitoring, adjustment and optimization of each process in the production process, greatly improving the accuracy, flexibility and efficiency of production.

[0076] Preferably, the present specification also provides a digital visualization management system for parts production and processing, which is used to execute the digital visualization management method for parts production and processing as described above, and the digital visualization management system for parts production and processing includes:

[0077] The accessory quality defect analysis module is used to obtain accessory production and processing data, and extract production process features based on the accessory production and processing data to obtain production process data; perform accessory quality defect analysis on the production process data to obtain accessory quality defect data;

[0078] The correlation analysis module is used to extract the production equipment status features based on the parts production and processing data, so as to obtain the equipment status data; the correlation analysis of the parts quality defect data is performed based on the equipment status data to obtain the equipment status-parts defect correlation data;

[0079] The accessory abnormality intelligent early warning module is used to construct a quality defect prediction model based on the equipment status-accessory defect correlation data, thereby obtaining a quality defect prediction model; perform accessory abnormality intelligent early warning on the accessory production and processing data based on the quality defect prediction model, thereby obtaining accessory abnormality intelligent early warning data;

[0080] The digital visualization management module is used to optimize the production scheduling of the parts production and processing data according to the intelligent warning data of the parts abnormality, obtain the production scheduling optimization data, and update the parts production and processing data according to the production scheduling optimization data to obtain the production process update data; upload the production process update data to the digital production management platform to perform digital visualization management tasks.

[0081] The digital visualization management system for parts production and processing of the present invention can implement any one of the digital visualization management methods for parts production and processing of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the digital visualization management method for parts production and processing. The internal modules of the system cooperate with each other, which significantly improves the production efficiency and the quality qualification rate of parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0083] Figure 1 A schematic diagram of the steps of the digital visualization management method for production and processing of accessories of the present invention;

[0084] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0085] Figure 3 Detailed step flow diagram of step S12 in the present invention;

[0086] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0087] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0088] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0089] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0090] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for digital visualization management of parts production and processing, the method comprising the following steps:

[0091] Step S1: acquiring accessory production and processing data, and extracting production process features based on the accessory production and processing data, thereby obtaining production process data; performing accessory quality defect analysis on the production process data, thereby obtaining accessory quality defect data;

[0092] In this embodiment, the original data of the production and processing of accessories is obtained from the database of the production line. These data include key factors such as production process parameters, processing time, temperature, pressure, and rotation speed, and SQL database query technology is used to extract data. On this basis, feature extraction is performed on the production process data of accessories. Taking "process parameters" as an example, specific process conditions such as welding current, welding voltage, and welding speed are extracted. In order to ensure the accuracy of process feature extraction, the data is normalized using a standardized method so that the dimensions of different process parameters are unified. For example, the welding current range is adjusted to between 0 and 1, and the Z-score standardized method is used for processing to ensure that the data can be compared. The production process data is analyzed for accessory quality defects. According to the variables that cause defects in the production process (such as too high current or too low voltage), a statistical method is used to perform regression analysis on the data to identify process factors related to quality problems such as surface defects, dimensional deviations, and cracks. For example, through the regression analysis method, it is identified that when the welding current is greater than 150A, the probability of cracks on the surface of the accessory increases. Finally, based on these analysis results, accessory quality defect data is generated, and the relevant process parameters and quality problem types are recorded.

[0093] Step S2: extracting production equipment status features based on accessory production and processing data to obtain equipment status data; performing correlation analysis on accessory quality defect data based on the equipment status data to obtain equipment status-accessory defect correlation data;

[0094] In this embodiment, the equipment status data is obtained from the sensor system of the production equipment. The data contains information such as the temperature, vibration, load, and running time of the equipment. The equipment status data is monitored in real time by acquisition devices such as acceleration sensors and temperature sensors, and specific values ​​such as vibration data are collected by sensors with a sampling frequency of 50Hz. The data are preprocessed using a data processing platform, such as smoothing the data by a sliding window method to eliminate noise and ensure data accuracy. Feature extraction is performed based on the equipment status data. Taking the equipment vibration data as an example, the acceleration signal and spectrum characteristics of the equipment, such as the maximum amplitude, the center frequency of the spectrum, etc., are extracted to calculate the feature vector. These features can be used to reflect the working status of the equipment. For example, a high vibration frequency indicates that the equipment has mechanical problems and affects the production quality. Next, the correlation analysis of the equipment status and the quality defect data of the accessories is performed. The Pearson correlation coefficient is used to calculate the relationship between the equipment status characteristics and the quality defects, and the equipment status-accessory defect correlation data is obtained. For example, the peak value of the vibration data is highly correlated with the occurrence frequency of cracks in certain accessories, and this information can be further used for subsequent prediction and scheduling optimization.

[0095] Step S3: construct a quality defect prediction model based on the equipment status-accessory defect association data, thereby obtaining a quality defect prediction model; perform intelligent warning of accessory abnormalities on the accessory production and processing data based on the quality defect prediction model, thereby obtaining intelligent warning data of accessory abnormalities;

[0096] In this embodiment, a linear regression method is used, in which the quality defect data of accessories is used as the dependent variable and the equipment status characteristics are used as the independent variable. Using annotated historical data sets (such as data related to equipment status and quality defects), select features with high correlation, such as equipment temperature, vibration frequency, and equipment load, to build a prediction model. The amount of data used for model training is no less than 3,000 sets of sample data, and the data is cross-validated to ensure that the model has good generalization ability. Intelligent warning of accessory abnormalities is performed based on the quality defect prediction model. By inputting the equipment status data collected in real time, the prediction model is applied to obtain the type of defective accessories. For example, when the vibration frequency of the equipment exceeds the set threshold (such as exceeding 5Hz), and the prediction model calculates that the probability of defect is above 80%, the system issues an abnormal warning signal to remind the operator to perform maintenance or adjustments.

[0097] Step S4: Optimize the production scheduling of the parts production and processing data according to the intelligent warning data of the parts abnormality to obtain the production scheduling optimization data, and update the parts production and processing data according to the production scheduling optimization data to obtain the production process update data; upload the production process update data to the digital production management platform to perform digital visualization management tasks.

[0098] In this embodiment, the production tasks that need to be processed first are determined according to the early warning data. For example, if the abnormal early warning signal of a certain equipment indicates the risk of quality defects (such as cracks or dimensional deviations), the production scheduling system will automatically adjust the production plan and postpone or reallocate the task to other equipment for processing. At the same time, the scheduling tasks are optimized using production process parameters (such as welding time, temperature, pressure, etc.) so that resources are effectively configured. The scheduling optimization algorithm uses priority-based task scheduling rules to dynamically adjust the order and time of production tasks according to the severity of the quality defects of the accessories and the status of the equipment. The accessory production and processing data is updated according to the production scheduling optimization data and reflected in the production process update data. Specifically, the optimized scheduling information will be transmitted to the production management system to update the production progress, equipment allocation and other data. For example, the welding process of a certain accessory is adjusted to a lower temperature to avoid cracks, and the system will update the operating parameters of the welding machine accordingly. Finally, the updated production process is uploaded to the digital production management platform, which displays the current production status through a visual interface, automatically executes the new scheduling plan, and ensures real-time monitoring and management.

[0099] Preferably, step S1 specifically comprises:

[0100] Step S11: acquiring accessory production and processing data, and extracting production process features according to the accessory production and processing data, thereby obtaining production process data;

[0101] In this embodiment, the accessory production and processing data is obtained from the real-time data acquisition system at the production site, including parameters such as welding current, voltage, welding time, and welding speed. The data is saved in CSV format and imported into the data analysis platform. These data are sorted and counted through the Pandas library in Python or the data processing function in Excel. The mean, standard deviation, maximum and minimum values ​​of the welding current and voltage are calculated, and these statistical characteristics can reflect the stability and fluctuation of the welding process during the production process. The data of welding time and welding speed are further processed, and the standard deviation and fluctuation range are used to reflect the changing trend of the process. According to the welding time, the allowable range of process fluctuation is set. For example, if the welding time deviation exceeds ±5%, it is marked as abnormal, and if the welding speed deviation exceeds ±10%, a warning is required. At the same time, the temperature variation range is extracted, and the difference between it and the standard process temperature is calculated as an important indicator of process characteristics.

[0102] Step S12: Analyze the surface defects of the parts on the production process data, so as to obtain the surface defect data of the parts;

[0103] In this embodiment, a laser scanner and a visual sensor are used to detect surface defects of accessories. The laser scanner is used to obtain surface height data, and the visual sensor (such as a high-resolution camera) obtains image information of the accessory surface. The image data is imported into the image processing platform, and the surface image is preprocessed using the Canny edge detection algorithm to extract the edge information of the surface defects. The low threshold of the Canny algorithm is set to 100 and the high threshold is set to 200 to ensure that small surface defects are not ignored. For more complex surface defects, such as cracks and pores, morphological operations (expansion, corrosion) are used to denoise and correct the boundaries of the defects. The size, position and shape of the defects are extracted through regional calibration technology, and the defects are classified according to the defect area and shape irregularity. When recording defect data, the defect type, location coordinates, size (for example, the crack length is 5mm) and shape characteristics are included.

[0104] Step S13: performing accessory size deviation defect analysis on the production process data, thereby obtaining accessory size deviation defect data;

[0105] In this embodiment, for the dimensional deviation defect analysis of accessories, a three-coordinate measuring machine (CMM) is used for precise measurement. The key dimensions of the accessories (such as diameter, thickness, and length) are collected at the measuring points, and the standard dimensions are compared with the actual measured values. First, set the tolerance range, for example, the tolerance of the diameter is ±0.2mm, and the tolerance of the length is ±0.5mm. The dimensional deviation value of each measuring point is calculated. If the deviation value exceeds the tolerance range, it is considered that there is a dimensional deviation at this point. During specific operations, error measurement standards, such as the dimensional accuracy standards of ISO 9001, are used to determine the dimensional error. These dimensional data are stored in the database, and the accessories with deviations exceeding the tolerance range are retrieved through the SQL query system to automatically generate a deviation report. For parts produced in large quantities, the statistical process control (SPC) method is used to analyze the dimensional deviation trend, and the mean and standard deviation of the dimensional error are analyzed to monitor the dimensional stability during the production process.

[0106] Step S14: Integrate the accessory quality defects according to the accessory surface defect data and the accessory size deviation defect data, so as to obtain the accessory quality defect data.

[0107] In this embodiment, the integration operation of the quality defects of the accessories is first based on the surface defect data obtained in step S12 and the dimensional deviation data in step S13, combined with the severity of the surface defects and the degree of dimensional deviation for comprehensive evaluation. Each data is assigned different weights according to the defect type (surface crack, dimensional deviation) and its severity (less than ± 0.2mm, ± 0.5mm, etc.). For example, when the severity of the surface crack is high, the weight is set to 0.7, and the weight of the dimensional deviation is set to 0.3. After each feature of the surface defect (such as crack length, pore position) and dimensional deviation data (such as dimensional error 0.3mm) are combined, the total quality defect data is formed. The integration method adopts a weighted average algorithm to multiply the weight of each defect and the defect value to calculate the comprehensive defect score. By setting a threshold, if the total score of the quality defect exceeds a preset value (such as 80 points), it is marked as an unqualified product and sent for inspection. This integrated data is stored in the database and serves as the basis for subsequent production scheduling and quality improvement.

[0108] Preferably, step S12 is specifically as follows:

[0109] Step S121: Obtain industrial accessories;

[0110] In this embodiment, industrial accessories are obtained through a production line or inventory system. Accessories can be mechanical parts made of metal, plastic or composite materials, depending on the process requirements. The acquisition of accessories relies on automated warehousing systems or manually input order data, using barcode scanners or RFID technology to identify the unique ID of each accessory. Before production, accessories are preliminarily classified according to demand and type to ensure that the size and shape of the accessories meet the processing requirements. The acquired data includes the material, size, production batch, etc. of the accessories, which are stored in the system for subsequent processing and inspection.

[0111] Step S122: dividing the industrial accessories based on the production process data, thereby obtaining fine process industrial accessories and coarse process industrial accessories;

[0112] In this embodiment, according to the production process data, the production information of all accessories is first queried through the process management system or database. By comparing the process parameters (such as welding time, welding current, voltage) and the type of processing equipment, the accessories are divided into fine process and rough process categories. Fine process industrial accessories usually require higher processing accuracy, and the control of process parameters is relatively strict. They are suitable for accessories with high requirements for surface quality and size. Rough process industrial accessories are relatively rough and allow a certain error. They are mainly used for basic processing or accessories that are not directly exposed. The classification standard is set as: if the standard deviation of the welding time exceeds 5% or the current fluctuation is greater than ±10%, it is regarded as a rough process accessory. The system automatically classifies according to these standards and records the process type of each accessory.

[0113] Step S123: performing crack defect detection on the fine craft industrial accessories to obtain crack defect data;

[0114] In this embodiment, the fine craft industrial accessories are detected for crack defects through an automatic visual inspection system. The system uses a high-resolution camera to capture the surface of the accessories, and with LED lighting and appropriate angles, enhances the clarity of surface details. The image is preprocessed by image processing software (such as OpenCV), and an adaptive threshold method is used to highlight the crack features. When setting the threshold parameters, edge detection (such as the Canny algorithm) is used to process the image. The low threshold of the Canny algorithm is set to 50 and the high threshold is set to 150 to ensure that the cracks are clearly visible. The length, width and depth information of the cracks are obtained through contour extraction technology. The severity of the cracks is marked as defects when the length exceeds 3mm or the width exceeds 0.2mm, and is stored in the defect database.

[0115] Step S124: performing burr defect detection on rough process industrial accessories to obtain burr defect data;

[0116] In this embodiment, burr defect detection of rough process industrial accessories is performed by an automated deburring system. The system uses a laser sensor and a visual sensor to perform online real-time detection. The laser sensor performs a three-dimensional scan of the surface of the accessory, accurately measures the height change of each point, and uses edge detection technology to identify the burr area in combination with the accessory image taken by the visual sensor. The burr detection standard is: the part with a height exceeding 0.2mm and a burr length exceeding 5mm is judged as a defect. The burr defect detection data includes the length, position and size of the burr, and the data is input into the control system for further processing and reporting. The accuracy of burr detection is optimized by adjusting the position and angle of the laser probe.

[0117] Step S125: integrating the surface defects of the accessory according to the crack defect data and the burr defect data, thereby obtaining the surface defect data of the accessory.

[0118] In this embodiment, the integration of accessory surface defect data first summarizes the crack defect data and burr defect data. Crack defect data includes crack location, length and depth, and burr defect data includes burr location, length and height. When integrating data, the weighted average method is used to evaluate the severity of each defect. The weight of the crack defect is set to 0.7, and the weight of the burr defect is set to 0.3. For each accessory, the crack and burr defects are weighted according to their respective severity and standards to obtain a comprehensive defect score. For example, if the crack length is 4mm and the burr height is 0.25mm, the corresponding defect data are multiplied by the weights and then summed. In this way, the defect data is integrated and provides a basis for subsequent quality assessment and production scheduling.

[0119] Preferably, step S123 is specifically as follows:

[0120] Apply penetrant to fine craft industrial accessories to obtain penetrant application industrial accessories;

[0121] In this embodiment, the fine craft industrial accessories are coated with penetrant after surface cleaning. The penetrant used must comply with ISO 3452 standards to ensure that it is suitable for metal surfaces and will not corrode the material of the accessories. The coating method adopts the spraying method, and the spray gun is used to set the spraying pressure to 2.5 bar and the nozzle size to 0.8 mm to ensure that the penetrant is evenly covered on the surface of the accessories. The coated accessories need to stand at room temperature for 15 to 30 minutes to wait for the penetrant to fully penetrate the tiny cracks and defects on the surface of the accessories. During the coating process, the coating thickness of each accessory is monitored, and a laser thickness gauge is used to ensure that the thickness of the penetrant layer is controlled between 0.1 mm and 0.2 mm. After the coating is completed, the accessories enter the next process for penetrant removal.

[0122] Surface penetrant removal is performed according to the penetrant application industrial accessories to obtain surface penetrant removal industrial accessories;

[0123] In this embodiment, the surface penetrant is removed from the industrial accessories after the penetrant is coated, and a high-pressure water gun or a special solvent cleaning agent is used for cleaning. In the specific operation, the water gun spray pressure is 15MPa, and the nozzle is kept at about 20cm from the surface of the accessory to ensure that the penetrant is completely removed from the surface. When cleaning, a solvent mixture of 60% alcohol and 40% deionized water is used to spray the difficult-to-clean areas locally. During the cleaning process, ensure that there is no penetrant residue on the surface of each accessory. After cleaning, use a hair dryer to blow dry the surface of the accessory and perform a drying process. The temperature is set to 50°C and the drying time is 15 minutes. The surface of the accessory after cleaning is in a state of no penetrant residue, ready for developer spraying.

[0124] Evenly spray the developer on the surface penetrant removal industrial accessories and collect data to obtain the developer spraying data;

[0125] In this embodiment, after the penetrant on the surface of the accessory is removed, the developer is evenly sprayed. The developer is a visual developer that meets the ASTM E165 standard, which can show the crack area under ultraviolet irradiation. The developer is sprayed through an automatic spraying system, the nozzle size of the spraying system is set to 1.0mm, and the spraying pressure is 2bar, ensuring that the developer evenly covers the surface of the accessory. After the developer is sprayed, the accessory is exposed under a UV lamp, the exposure intensity is set to 300W, and the exposure time is 5 minutes. After the developer is evenly coated on the surface of the accessory, a high-resolution camera is used to capture images of it. During the acquisition process, the camera maintains a fixed distance of 50cm from the surface of the accessory, and the resolution is set to 1920x1080 to obtain clear crack image data.

[0126] Based on the developer spraying data, the crack appearance area of ​​the surface penetrant removal industrial accessories is identified to obtain the crack appearance area data;

[0127] In this embodiment, image processing software is used to extract crack regions. First, image binarization is performed and the threshold is set to 250 so that the crack region is clearly visible in the image. Then, a morphological processing method is used to remove noise and retain the crack morphology through corrosion and expansion operations. For the identified crack region, its aspect ratio is calculated. The standard for a long crack is that the aspect ratio is greater than 2 and the crack region area exceeds 0.2 cm 2 The part is determined as the crack-appearing area. The boundary of the crack-appearing area is extracted by contour detection technology, and the position and shape data of the crack are obtained to form the crack-appearing area data, which is stored for subsequent analysis.

[0128] Perform crack length statistics on the crack appearance area data to obtain long crack data;

[0129] In this embodiment, the crack length information contained in the crack appearance area data is counted by an image processing system. When counting, the crack length standard is set to calculate the length by the Euclidean distance of the crack boundary contour points. Cracks with a length greater than 2 mm are classified as long cracks. During the statistical process, a statistical analysis tool is used to summarize the lengths of all cracks to ensure that the crack length data of each accessory is accurately recorded. For the parts with a crack length greater than 2 mm, the number, distribution area and total length of the long cracks are counted. The results are summarized as long crack data, including the number and location of long cracks on each accessory, and the data is stored for subsequent stress analysis.

[0130] Based on the long crack data, stress concentration analysis is performed on precision industrial parts to obtain long crack stress concentration data;

[0131] In the present embodiment, long crack data is used as input to perform stress concentration analysis of fine process industrial accessories. Stress analysis is performed by finite element analysis (FEA) method, and commercial software (such as ANSYS) is used for modeling. The geometry and material properties (such as Young's modulus, Poisson's ratio, etc.) of the accessories are input into the FEA software through the CAD model. The long crack position and geometry are input into the analysis as load boundary conditions to calculate the stress value of the stress concentration zone. The length and shape of the crack are set as important parameters for stress concentration analysis. During the simulation process, the stress concentration intensity at the end of the crack is positively correlated with the crack length. The maximum stress value in the stress concentration area is recorded and compared with the material bearing limit value of the accessories to obtain stress concentration data.

[0132] Based on the long crack data, fatigue crack growth simulation is performed on the precision industrial parts to obtain fatigue crack growth data;

[0133] In this embodiment, the long crack data is combined with the stress concentration data to simulate the fatigue crack growth of the precision industrial parts. The fatigue crack growth model based on the Paris law is used to set the initial conditions and loading cycles of the crack growth. The crack growth rate is calculated based on the stress intensity factor (K value). During the simulation, the parameters of crack growth include the loading cycle, crack depth and growth rate. During the simulation, each loading cycle is set to 10,000 times, and the crack growth rate is 10 -7 The fatigue crack growth data records the crack growth length under different loading cycles, outputs the crack length change data for each loading cycle, and provides it for subsequent defect impact analysis.

[0134] The influence of long crack defects on fatigue crack growth data is analyzed based on long crack stress concentration data to obtain crack defect data.

[0135] In this embodiment, the influence of long crack extension on the overall strength of the fitting is evaluated by combining the stress concentration area and fatigue crack extension simulation data. The interaction between the stress concentration area and crack extension is modeled as a key factor and analyzed by a step-by-step loading method. The loading stress is set to 50MPa and the maximum crack extension depth is 2mm. The influence of the crack on the overall bearing capacity of the fitting is analyzed by simulating the loading cycle. By monitoring the crack extension process, crack defect data is obtained, and the crack depth, extension speed, and the influence of the final defect formation are recorded.

[0136] Preferably, step S124 is specifically as follows:

[0137] Capturing images of crude industrial parts, thereby obtaining images of crude industrial parts;

[0138] In this embodiment, the image acquisition of the rough process industrial accessories is carried out by an industrial camera, and the accessories are photographed using a CCD camera with a resolution of 12MP. The camera light source uses a white LED lamp with a light source intensity of 500Lux to ensure that the surface of the accessories is clearly visible. During the acquisition, the camera and the accessories are kept at a fixed distance of 30cm, and the camera shooting angle is set to 45 degrees to avoid distortion due to surface reflection. During the acquisition process, the camera shutter is automatically triggered by the software to ensure the stability and accuracy of each image acquisition. After the image is captured, the data is stored in RAW format and directly transferred to the image processing platform for subsequent processing.

[0139] Performing image enhancement on the rough process industrial parts image, thereby obtaining a rough process industrial parts enhanced image;

[0140] In this embodiment, the image is gray-scale converted to ensure that the image processing is based on grayscale information. Next, the local contrast enhancement algorithm is applied to enhance the low contrast area in the image. In order to avoid excessive enhancement, the enhancement coefficient is set to 1.5 times, and the window size used in the enhancement process is 5x5 pixels to ensure the uniformity and stability of the image processing. In the image enhancement process, a convolution filter is used to smooth the image, and the window size of the filter is set to 3x3 to remove image noise. The enhanced image can more clearly show the detailed features of the industrial accessories, ready for subsequent image processing.

[0141] Binarization conversion is performed according to the rough process industrial parts enhanced image, so as to obtain a binary industrial parts enhanced image;

[0142] In this embodiment, the enhanced rough process industrial accessories image is binarized to facilitate subsequent feature extraction. The Otsu algorithm is used to automatically select the global threshold and set the binarization threshold of the image. The Otsu algorithm determines the optimal threshold by calculating the inter-class variance of the image to ensure that the separation of objects and background in the image is most obvious. After threshold calculation, the threshold is set to 150, which makes the objects and background in the image clearly distinguishable. After binarization, the pixel value of the image is set to 0 or 255, 0 represents the background, 255 represents the object area, and the image becomes black and white, making the identification of the burr area more intuitive.

[0143] Based on the preset burr threshold data, the burr area is identified on the binary industrial parts enhanced image, so as to obtain the burr area image;

[0144] In this embodiment, the minimum area threshold of the burr is set to 200 pixels, and any noise area smaller than this area will be filtered out. In the image, an edge detection algorithm (such as Canny edge detection) is used to detect the burr area in the image by calculating the gradient value of each pixel. For each detected edge area, morphological operations (dilation and erosion) are further applied to fill the edge gaps and eliminate the noise area. By performing connected domain analysis on the edge, all burr areas can be accurately identified, and the image data of each burr area can be obtained to generate a burr area image.

[0145] Performing burr contour detection on the burr area image to obtain burr contour data;

[0146] In this embodiment, the edge detection method is used to extract the contour of the burr. The Sobel operator is used to calculate the gradient of the image in the horizontal and vertical directions to obtain the gradient value of each pixel in the image. Then, the background noise is removed by threshold processing, and the significant contour part is retained. The threshold value of contour detection is set to 20, which ensures that the burr contour can be accurately displayed. For the obtained contour data, a polygonal approximation algorithm (such as the Ramer-Douglas-Peucker algorithm) is applied to simplify the contour, reduce redundant points, and finally extract the precise boundary of the burr. The obtained contour data is used to further analyze the location and size of the burr.

[0147] Determine the burr position of the burr area image according to the burr contour data, thereby obtaining the burr position data;

[0148] In this embodiment, the center point of each burr contour is determined by the geometric center calculation method, and the calculation formula is the coordinate mean of all contour points. For each burr contour, the minimum and maximum diameters are set to 2mm to 10mm, and small areas that do not meet the size requirements are excluded. For burrs that meet the conditions, their position coordinates and size information are recorded. The position data of each burr is stored through a data structure (such as a matrix) to ensure that subsequent assembly simulation can be accurately analyzed based on these positions.

[0149] According to the burr position data, the rough process industrial accessories are assembled and simulated to obtain assembly simulation data;

[0150] In this embodiment, a digital model of the accessories is established using three-dimensional CAD software, the burr position data is input into the system, and the position of each burr is marked. During the assembly simulation, the assembly order of the accessories and the mechanical constraints applied are set. During the assembly process, it is assumed that the assembly force of each accessory is 50N and the maximum displacement during assembly is 0.2mm. During the simulation, based on the geometric shape of the accessory and the position of the burr, it is calculated whether the burr will interfere with other accessories. After the assembly simulation is completed, the system outputs the assembly status of each accessory and the area where interference occurs, and generates assembly simulation data for subsequent burr defect evaluation.

[0151] The burr defect impact assessment is performed based on assembly simulation data to obtain burr defect data.

[0152] In this embodiment, the defect impact of burrs is evaluated by calculating the degree of interference of burrs on the assembly of accessories. First, the burr impact threshold is set to 0.5mm, that is, when the distance between the burr and the accessory is less than 0.5mm, the burr is considered to have a negative impact on the assembly. For each burr, based on the relative position relationship between its position data and other accessories, it is judged whether the burr will increase the difficulty of assembly or damage the accessory during the assembly process. For burrs with a greater impact, the system will generate defect data, including the specific location, size and type of impact of the burr. Finally, the system outputs burr defect data for quality control and the formulation of subsequent repair plans.

[0153] Preferably, step S25 is specifically as follows:

[0154] Step S131: extracting aperture process features from production process data to obtain aperture process data;

[0155] In this embodiment, the dimensional data related to the aperture is extracted from the production process data. The aperture size is measured using a precision measuring tool (such as a laser measuring instrument), and the data is in millimeters. For each process step, the actual size and hole position distribution of the aperture are recorded. By comparing the measurement results with the design drawings, the deviation value of the aperture size is extracted. The hole position of each process is digitally extracted using a dedicated software tool (such as CAD or CAM software) to obtain the aperture process data. These data include the diameter, depth, and relative position relationship of the aperture with other components, and the data are used for subsequent dimensional deviation analysis.

[0156] Step S132: obtaining standard aperture size data;

[0157] In this embodiment, standard aperture size data is obtained. Standard aperture size data is formulated according to design requirements and industry standards. Standard aperture size includes multiple key parameters, such as the minimum diameter, maximum diameter and tolerance range of the aperture. The standard size can be extracted from the drawing or determined according to industry specifications (such as ISO 2768 standard) to ensure that the accuracy of the aperture meets the assembly requirements. This data is stored in a database and connected to the production system through a data interface for real-time comparison and update during the production process.

[0158] Step S133: Calculating the size deviation of the aperture process data according to the standard aperture size data, thereby obtaining size deviation data;

[0159] In this embodiment, the standard aperture size data is compared with the actual aperture size in the process data, and the deviation value of each aperture is recorded. For each hole, the deviation between the actual aperture size and the standard size is calculated, and the deviation calculation formula is: deviation = actual size - standard size. Through this formula, the size deviation data of each aperture can be obtained. In order to improve the accuracy of the calculation, the tolerance range of the deviation is set to ± 0.1mm. If the calculation result exceeds this range, it is marked as unqualified. The calculation process uses data analysis software (such as MATLAB or Python) for automated calculation to generate a detailed size deviation report.

[0160] Step S134: identifying the size deviation location of the fine craft industrial accessories according to the size deviation data, thereby obtaining the size deviation location of the accessories;

[0161] In this embodiment, by analyzing the dimensional deviation data, data visualization tools (such as 3D CAD models or heat maps) are used to identify the specific locations where dimensional deviations exist on the accessories. The dimensional data of each accessory is compared with the standard dimensional data to generate a dimensional deviation area. In this process, the location where the deviation exceeds the preset threshold (such as ±0.2mm) is identified as the deviation location and marked in red, while other normal locations are represented in green. The coordinate information of these dimensional deviation locations is automatically identified and recorded through software tools to facilitate subsequent assembly and quality control.

[0162] Step S135: performing assembly simulation according to the size deviation part of the accessory, thereby obtaining assembly simulation data of the size deviation part of the accessory;

[0163] In this embodiment, the 3D digital model of the accessory is imported into the assembly simulation software (such as CATIA or Siemens NX). The software compares the dimensional data of the accessory with the actual process data to simulate the actual assembly of the accessory during the assembly process. Through simulation, the assembly problems caused by dimensional deviation of the accessory during the assembly process, such as insufficient matching clearance or hole offset, are determined. In this process, the assembly force is set to 50N to simulate the motion trajectory, sequence and matching of the assembly operation. The assembly simulation results provide areas where interference or poor assembly occurs, and these data will be used as evaluation criteria.

[0164] Step S136: evaluating the strength defect of the accessory according to the assembly simulation data of the accessory size deviation position, thereby obtaining the strength defect data of the accessory;

[0165] In this embodiment, finite element analysis (FEA) software (such as ABAQUS or ANSYS) is used to simulate the mechanical response of the accessory according to the size deviation position of the accessory. In this step, the loading condition is set to an assembly force of 50N, and the maximum stress threshold of the accessory is 200MPa. When the stress value of a certain part of the accessory exceeds this threshold, the part is marked as having a strength defect. For each part involved in the assembly simulation process, the material properties (such as yield strength, elastic modulus, etc.) are used to perform strength calculations, generate strength defect data, and record in detail the stress level of each deviation part and the type of damage caused.

[0166] Step S137: performing accessory durability prediction according to the accessory size deviation position assembly simulation data, thereby obtaining accessory durability data;

[0167] In this embodiment, the fatigue life of accessories during long-term use is simulated by a fatigue analysis method based on mechanics. First, the reference parameters for fatigue life calculation, such as cyclic stress amplitude, strain frequency, etc., are set according to the material and working environment of the accessories. The fatigue life prediction uses the SN curve method to calculate the life of accessories under specific stress. When the stress amplitude of a deviation part in the assembly simulation exceeds the set threshold (such as 500MPa), it is predicted that the durability of the part will be greatly reduced. Through the simulation results, the durability data of the accessories under different working conditions is generated, and the expected life cycle of each part is recorded.

[0168] Step S138: judging the impact of the size deviation defect on the accessory durability data based on the accessory strength defect data, thereby obtaining the accessory size deviation defect data.

[0169] In this embodiment, the impact of dimensional deviation defects on the long-term service life of accessories is determined through a comprehensive analysis of accessory strength defect data and durability data. In this process, a multivariate analysis method (such as regression analysis) is used to quantitatively analyze the relationship between strength defects and durability of each deviation location. The impact threshold is set to 25%, that is, when the strength defect of a certain deviation location exceeds 25%, it is judged that the dimensional deviation of this location has a significant impact on the durability of the accessory. Through this analysis method, the key dimensional deviation locations are determined, and dimensional deviation defect data is generated for subsequent optimization of process flow or formulation of repair plans.

[0170] Preferably, step S2 specifically comprises:

[0171] Step S21: extracting production equipment status features based on accessory production and processing data, thereby obtaining equipment status data;

[0172] In this embodiment, the operation data of the equipment during the production process is collected and sorted, including but not limited to parameters such as current, voltage, temperature, rotation speed, pressure, etc. These parameters are usually collected in real time by sensors in the production equipment. Then, the collected equipment operation data is processed using a data analysis tool to calculate the key characteristics of the equipment operation, such as equipment operation frequency, load changes, equipment temperature fluctuations, etc. These characteristic data are used to evaluate the working status of the equipment, thereby obtaining equipment status data and forming basic data for subsequent analysis.

[0173] Step S22: Identify the device abnormality code according to the device status data, thereby obtaining the device abnormality code data;

[0174] In this embodiment, the device status data includes the operating parameters and status characteristics of the device, and these data need to be extracted from the fault identification algorithm for abnormal code. First, based on the equipment operation standard or the preset parameter range, the device status data is compared with the standard value, and the data exceeding the set threshold will be marked as abnormal. For example, if the current of the device exceeds the set safety range, it is marked as current abnormality and recorded as an abnormal code. Then, by setting the tolerance range of the equipment working state, these abnormal data are matched with the specific equipment fault type, and finally the equipment abnormal code data is generated to facilitate subsequent fault diagnosis and processing.

[0175] Step S23: identifying mechanical abnormal equipment according to the equipment abnormality code data, thereby obtaining mechanical abnormal equipment;

[0176] In this embodiment, mechanical abnormal devices can be identified by screening and aggregating the device abnormality code data generated in step S22. For example, for devices that generate multiple abnormality codes such as abnormal current, excessive temperature, unstable speed, etc., centralized analysis is performed to identify faulty devices. In this process, threshold settings are used to define the conditions under which the device can be judged as mechanically abnormal. The threshold setting is determined based on historical fault records and device specifications. For example, if the device failure rate exceeds the normal range and multiple types of abnormality codes appear, it is judged as a mechanically abnormal device.

[0177] Step S24: performing abnormal state time statistics on mechanical abnormal equipment, thereby obtaining equipment abnormal state time data;

[0178] In this embodiment, the specific abnormality codes of mechanical abnormal equipment and the time points of their occurrence are collected to calculate the duration of the equipment in the abnormal state. The specific operation is to identify the start and end time points of each fault in the equipment abnormality code data, and count the duration of each abnormal state based on these time points. When counting, the time difference calculation method is used to obtain the duration of the fault based on the timestamp in the equipment fault record. The statistical equipment abnormal state time data can provide a basis for subsequent decision-making.

[0179] Step S25: performing defect detection time statistics on the accessory quality defect data, thereby obtaining defect detection time data;

[0180] In this embodiment, during the production process of accessories, dedicated inspection equipment (such as visual inspection system, X-ray inspection equipment, etc.) is used to perform quality inspection on each batch of accessories, and the inspection time of each accessory is recorded. By statistically analyzing the recorded data of quality inspection, the defect detection time of each accessory is calculated to obtain the inspection cycle of each accessory. The key to this process is to accurately record the start and end time of each defect detection, and to calculate based on these time data to obtain statistical results.

[0181] Step S26: Perform time-related intersection operations based on the equipment abnormal state time data and the defect detection time data to obtain equipment state-accessory defect correlation data.

[0182] In this embodiment, the data obtained in step S24 and step S25 are collected and sorted, and the equipment abnormal state time data and the defect detection time data are aligned according to the timestamp. Then, a time intersection operation is performed, that is, the overlapping area of ​​the equipment abnormal state time and the accessory quality defect detection time is identified. The specific operation is to compare the time periods in the two data sets, find the overlapping parts of the two, and obtain the associated data of the equipment state and quality defects. By setting the time window size, the recognition accuracy of the overlapping area can be optimized to ensure the accuracy of the time association operation.

[0183] Preferably, step S3 specifically comprises:

[0184] Step S31: constructing a quality defect prediction model according to the equipment status-accessory defect association data, thereby obtaining a quality defect prediction model;

[0185] In this embodiment, in combination with the associated data of equipment status and accessory defects, an exploratory analysis is performed on the relationship between equipment status and quality defects through data processing tools (such as the Pandas library in Python) to identify potential associated features. For example, the overlap between the abnormal state time of the equipment and the detection time of accessory defects is analyzed to find feature variables with high correlation. Then, a quality defect prediction model is established using the extracted feature data using statistical learning methods (such as regression analysis, support vector machines, etc.). In this process, it is necessary to set appropriate parameters, such as setting an abnormal threshold for the equipment status. When the operating state of the equipment exceeds the set threshold, it is predicted that the accessory will have a quality defect. The training process of the model requires a training data set, which contains data on equipment status, defect type, and other related variables, and the effect of the model is evaluated and optimized by a cross-validation method.

[0186] Step S32: Predicting quality defects of the parts production and processing data according to the quality defect prediction model, thereby obtaining quality defect prediction data;

[0187] In this embodiment, by applying the model established in step S31, the production and processing data of the accessories, including production environment parameters, equipment status parameters, processing parameters, etc., are input into the prediction model to predict quality defects. In the specific operation, data preprocessing is first required to ensure that the format of the input data meets the model requirements and remove outliers. Then, the model will predict the quality defects of each accessory based on the characteristics of the input data and the patterns learned during the training process. For different input data, the model can be used to perform risk assessment on each accessory and output quality defect prediction data, specifically including predicted defect probability values ​​or defect categories.

[0188] Step S33: performing confusion matrix analysis based on the quality defect prediction data, thereby obtaining quality defect prediction confusion matrix data;

[0189] In this embodiment, confusion matrix analysis is performed based on quality defect prediction data. According to the quality defect prediction data obtained in step S32 and the actual quality defect data, confusion matrix analysis is performed. First, it is necessary to compare the actual quality defect label with the defect result predicted by the model to generate a confusion matrix. Each element in the confusion matrix reflects the degree of match between the predicted result and the actual result. For example, the diagonal elements of the matrix represent the number of defect types predicted correctly, while the non-diagonal elements represent the situation of incorrect prediction. Through this analysis, the prediction ability of the model under different quality defect types and the prediction accuracy of different categories can be evaluated. This analysis is performed through tools (such as the confusion_matrix function in Scikit-learn) to calculate evaluation indicators such as precision and recall.

[0190] Step S34: performing prediction error analysis on the quality defect prediction data according to the quality defect prediction confusion matrix data, thereby obtaining prediction error data;

[0191] In this embodiment, the prediction types that produce errors in the prediction process are identified by analyzing the confusion matrix. For example, check which categories of defect predictions have the most frequent errors, especially those with higher false positives and false negatives. In order to conduct a more detailed error analysis, it is necessary to extract performance indicators such as prediction accuracy and recall rate for each defect type based on the data in the confusion matrix, and analyze the causes of erroneous predictions in combination with actual production data. Through in-depth analysis of different types of defect prediction errors, the shortcomings of the model and room for improvement are found. For example, a certain defect type has more prediction errors due to insufficient selection of equipment status features or unbalanced defect data.

[0192] Step S35: Correcting the quality defect prediction model according to the prediction error data, thereby obtaining a quality defect prediction correction model;

[0193] In this embodiment, it is analyzed which features have a greater impact on error prediction, and it is checked whether the equipment status data, defect detection time data, etc. are sufficient, and whether more feature variables need to be introduced. According to the results of the error analysis, it is necessary to adjust the threshold setting, for example, adjust the judgment criteria for abnormal equipment status, or add new parameters (such as production environment parameters) to improve the accuracy of the model. In addition, if there is a problem of data imbalance, the training data can be processed by oversampling or undersampling to further improve the model performance. Finally, the model is retrained with the adjusted data set, and the parameters in the original model are updated to make it more suitable for actual production conditions.

[0194] Step S36: Perform intelligent warning of accessory abnormality on accessory production and processing data according to the quality defect prediction and correction model, thereby obtaining intelligent warning data of accessory abnormality.

[0195] In this embodiment, the quality defect prediction model corrected in step S35 is used to input the production and processing data of accessories and perform abnormal warnings. In the specific implementation process, new production data, including equipment status, processing parameters, etc., are input, and after model processing, the prediction results are obtained. If the model predicts that a certain accessory has a high risk of quality defects, the system will trigger an early warning and inform the relevant personnel. The early warning data includes information such as the type of defect, the time of occurrence, and the affected production links. The parameter setting of the early warning is usually obtained through experiments at different production stages, and appropriate thresholds are set to ensure the timeliness and accuracy of the early warning.

[0196] Preferably, step S4 is specifically:

[0197] Step S41: performing abnormal warning classification according to the abnormal intelligent warning data of the accessories, thereby obtaining abnormal surface defect warning data and abnormal dimension deviation warning data;

[0198] In this embodiment, the intelligent warning data of accessory abnormality obtained in step S36 is used as input, and the data includes the defect types of various accessories and their predicted probability values. When these data are classified and processed, by setting specific defect standards, data segmentation methods (such as threshold-based judgment, clustering algorithm or classification model) are used to classify them into two categories: surface defects and dimensional deviations. Specifically, a warning value threshold for surface defects is set. For example, when the defect area exceeds a certain set value, it is defined as a surface defect; and the dimensional deviation is classified as a dimensional deviation abnormality by comparing the standard size and actual size data of the production process. If the dimensional deviation exceeds the set tolerance range. Through these specific parameter standards, tools such as Pandas or Scikit-learn in Python are used for data preprocessing and classification operations, and finally two types of warning data are obtained.

[0199] Step S42: Perform quality control processing based on the surface defect abnormality warning data to obtain quality control data;

[0200] In this embodiment, based on the set defect tolerance range, such as defect depth, area, shape and other parameters, determine which accessories have surface defects that exceed the tolerance standard. Then, use quality management tools (such as SPC control charts, Six Sigma methods, etc.) to perform quality control on these accessories. Specific operations include: comparing defect data with historical data, identifying defect patterns, and taking necessary remedial measures, such as adjusting the production line, checking the equipment status, or training operators. If an SPC control chart is used, control limits (such as UCL, LCL) can be set. When the number of surface defects exceeds these control limits, quality control processing is triggered, and each quality control data is recorded, including processing methods, results, etc.

[0201] Step S43: dynamically allocating production tasks based on the abnormal dimension deviation warning data to obtain dynamic production task allocation data;

[0202] In this embodiment, a rule-based algorithm or optimization method, such as a priority scheduling algorithm (for example, based on the shortest job first algorithm), is used to sort different production tasks. For those accessories whose dimensional deviation exceeds the tolerance range, they are preferentially allocated to production lines with higher equipment accuracy and that can meet the dimensional requirements. In this process, the parameter data of the production task, such as the type of accessory, the capacity of the processing equipment, and the process parameters, are used to calculate the task allocation. With these data, the scheduling algorithm is used to sort the task priorities, so that the production tasks are reasonably allocated to the equipment, equipment failures and production delays are avoided during the production process, and dynamic allocation data of production tasks is generated.

[0203] Step S44: updating the accessory production and processing data according to the quality control data and the production task dynamic allocation data to obtain the production process update data;

[0204] In this embodiment, first, the production process is checked based on the quality control data to see if there are any links that need to be adjusted, especially those involving adjustments to the equipment or changes in the operating method. For example, if it is found in the quality control data that a certain production equipment frequently has surface defects, the process parameters of the equipment are corrected. Secondly, according to the dynamic allocation data of production tasks, the task allocation and production line arrangement in the production process are adjusted. Through the scheduling algorithm, combined with the actual status and production capacity of the equipment, the execution order of the production tasks and the layout of the production line are replanned. All updated data, such as new production parameters, task allocation information and process steps, form production process update data, which are recorded and archived.

[0205] Step S45: Upload the production process update data to the digital production management platform to perform digital visualization management tasks.

[0206] In this embodiment, the production process update data obtained in step S44 is uploaded to the digital production management platform through the network interface. The data formats supported by the platform include JSON, XML, etc., and are connected to the existing production management system through the API interface. During the upload process, the integrity and accuracy of the data are ensured, and the data verification mechanism is used to verify the data to avoid erroneous data. After uploading the data, the platform will perform digital visualization management tasks based on the updated production process data, including real-time display of key parameters in the production process, task execution status, equipment operation status, etc. Through these updates, the platform can provide real-time production monitoring, scheduling optimization and decision support to ensure that the production line operates in accordance with the latest process standards.

[0207] Preferably, the present specification also provides a digital visualization management system for parts production and processing, which is used to execute the digital visualization management method for parts production and processing as described above, and the digital visualization management system for parts production and processing includes:

[0208] The accessory quality defect analysis module is used to obtain accessory production and processing data, and extract production process features based on the accessory production and processing data to obtain production process data; perform accessory quality defect analysis on the production process data to obtain accessory quality defect data;

[0209] The correlation analysis module is used to extract the production equipment status features based on the parts production and processing data, so as to obtain the equipment status data; the correlation analysis of the parts quality defect data is performed based on the equipment status data to obtain the equipment status-parts defect correlation data;

[0210] The accessory abnormality intelligent early warning module is used to construct a quality defect prediction model based on the equipment status-accessory defect correlation data, thereby obtaining a quality defect prediction model; perform accessory abnormality intelligent early warning on the accessory production and processing data based on the quality defect prediction model, thereby obtaining accessory abnormality intelligent early warning data;

[0211] The digital visualization management module is used to optimize the production scheduling of the parts production and processing data according to the intelligent warning data of the parts abnormality, obtain the production scheduling optimization data, and update the parts production and processing data according to the production scheduling optimization data to obtain the production process update data; upload the production process update data to the digital production management platform to perform digital visualization management tasks.

[0212] The digital visualization management system for parts production and processing of the present invention can implement any one of the digital visualization management methods for parts production and processing of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the digital visualization management method for parts production and processing. The internal modules of the system cooperate with each other, which significantly improves the production efficiency and the quality qualification rate of parts.

[0213] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0214] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A digital visualization management method for parts production and processing, characterized in that: The following steps are involved: Step S1: Acquire accessory production and processing data, and extract production process features based on the accessory production and processing data, thereby obtaining production process data; Analyze the quality defects of accessories on the production process data to obtain the quality defect data of accessories; Step S2: extracting production equipment status features based on accessory production and processing data to obtain equipment status data; performing correlation analysis on accessory quality defect data based on the equipment status data to obtain equipment status-accessory defect correlation data; Step S3: construct a quality defect prediction model based on the equipment status-accessory defect association data, thereby obtaining a quality defect prediction model; perform intelligent warning of accessory abnormalities on the accessory production and processing data based on the quality defect prediction model, thereby obtaining intelligent warning data of accessory abnormalities; Step S4: optimizing the production scheduling of the production and processing data of the accessories according to the intelligent early warning data of accessory abnormality to obtain the production scheduling optimization data, and updating the production and processing data of the accessories according to the production scheduling optimization data to obtain the production process update data; Upload production process update data to the digital production management platform to perform digital visualization management tasks.

2. The digital visualization management method for parts production and processing according to claim 1 is characterized in that: Step S1 is specifically as follows: Step S11: acquiring accessory production and processing data, and extracting production process features according to the accessory production and processing data, thereby obtaining production process data; Step S12: Analyze the surface defects of the parts on the production process data, so as to obtain the surface defect data of the parts; Step S13: performing accessory size deviation defect analysis on the production process data, thereby obtaining accessory size deviation defect data; Step S14: Integrate the accessory quality defects according to the accessory surface defect data and the accessory size deviation defect data, so as to obtain the accessory quality defect data.

3. The digital visualization management method for parts production and processing according to claim 2 is characterized in that: Step S12 is specifically as follows: Step S121: Obtain industrial accessories; Step S122: dividing the industrial accessories based on the production process data, thereby obtaining fine process industrial accessories and coarse process industrial accessories; Step S123: performing crack defect detection on the fine craft industrial accessories to obtain crack defect data; Step S124: performing burr defect detection on rough process industrial accessories to obtain burr defect data; Step S125: integrating the surface defects of the accessory according to the crack defect data and the burr defect data, thereby obtaining the surface defect data of the accessory.

4. The digital visualization management method for parts production and processing according to claim 3 is characterized in that: Step S123 is specifically as follows: Apply penetrant to fine craft industrial accessories to obtain penetrant application industrial accessories; Surface penetrant removal is performed according to the penetrant application industrial accessories to obtain surface penetrant removal industrial accessories; Evenly spray the developer on the surface penetrant removal industrial accessories and collect data to obtain the developer spraying data; Based on the developer spraying data, the crack appearance area of ​​the surface penetrant removal industrial accessories is identified to obtain the crack appearance area data; Perform crack length statistics on the crack appearance area data to obtain long crack data; Based on the long crack data, stress concentration analysis is performed on precision industrial parts to obtain long crack stress concentration data; Based on the long crack data, fatigue crack growth simulation is performed on the precision industrial parts to obtain fatigue crack growth data; The influence of long crack defects on fatigue crack growth data is analyzed based on long crack stress concentration data to obtain crack defect data.

5. The digital visualization management method for parts production and processing according to claim 3 is characterized in that: Step S124 is specifically as follows: Capturing images of crude industrial parts, thereby obtaining images of crude industrial parts; Performing image enhancement on the rough process industrial parts image, thereby obtaining a rough process industrial parts enhanced image; Binarization conversion is performed according to the rough process industrial parts enhanced image, so as to obtain a binary industrial parts enhanced image; Based on the preset burr threshold data, the burr area is identified on the binary industrial parts enhanced image, so as to obtain the burr area image; Performing burr contour detection on the burr area image to obtain burr contour data; Determine the burr position of the burr area image according to the burr contour data, thereby obtaining the burr position data; According to the burr position data, the rough process industrial accessories are assembled and simulated to obtain assembly simulation data; The burr defect impact assessment is performed based on assembly simulation data to obtain burr defect data.

6. The digital visualization management method for parts production and processing according to claim 5 is characterized in that: Step S25 is specifically as follows: Step S131: extracting aperture process features from production process data to obtain aperture process data; Step S132: obtaining standard aperture size data; Step S133: Calculating the size deviation of the aperture process data according to the standard aperture size data, thereby obtaining size deviation data; Step S134: identifying the size deviation location of the fine craft industrial accessories according to the size deviation data, thereby obtaining the size deviation location of the accessories; Step S135: performing assembly simulation according to the size deviation part of the accessory, thereby obtaining assembly simulation data of the size deviation part of the accessory; Step S136: evaluating the strength defect of the accessory according to the assembly simulation data of the accessory size deviation position, thereby obtaining the strength defect data of the accessory; Step S137: performing accessory durability prediction according to the accessory size deviation position assembly simulation data, thereby obtaining accessory durability data; Step S138: judging the impact of the size deviation defect on the accessory durability data based on the accessory strength defect data, thereby obtaining the accessory size deviation defect data.

7. The digital visualization management method for parts production and processing according to claim 1 is characterized in that: Step S2 is specifically as follows: Step S21: extracting production equipment status features based on accessory production and processing data, thereby obtaining equipment status data; Step S22: Identify the device abnormality code according to the device status data, thereby obtaining the device abnormality code data; Step S23: identifying mechanical abnormal equipment according to the equipment abnormality code data, thereby obtaining mechanical abnormal equipment; Step S24: performing abnormal state time statistics on mechanical abnormal equipment, thereby obtaining equipment abnormal state time data; Step S25: performing defect detection time statistics on the accessory quality defect data, thereby obtaining defect detection time data; Step S26: Perform time-related intersection operations based on the equipment abnormal state time data and the defect detection time data to obtain equipment state-accessory defect correlation data.

8. The digital visualization management method for parts production and processing according to claim 1 is characterized in that: Step S3 is specifically as follows: Step S31: constructing a quality defect prediction model according to the equipment status-accessory defect association data, thereby obtaining a quality defect prediction model; Step S32: Predicting quality defects of the parts production and processing data according to the quality defect prediction model, thereby obtaining quality defect prediction data; Step S33: performing confusion matrix analysis based on the quality defect prediction data, thereby obtaining quality defect prediction confusion matrix data; Step S34: performing prediction error analysis on the quality defect prediction data according to the quality defect prediction confusion matrix data, thereby obtaining prediction error data; Step S35: Correcting the quality defect prediction model according to the prediction error data, thereby obtaining a quality defect prediction correction model; Step S36: Perform intelligent warning of accessory abnormality on accessory production and processing data according to the quality defect prediction and correction model, thereby obtaining intelligent warning data of accessory abnormality.

9. The digital visualization management method for parts production and processing according to claim 1 is characterized in that: Step S4 is specifically as follows: Step S41: performing abnormal warning classification according to the abnormal intelligent warning data of the accessories, thereby obtaining abnormal surface defect warning data and abnormal dimension deviation warning data; Step S42: Perform quality control processing based on the surface defect abnormality warning data to obtain quality control data; Step S43: dynamically allocating production tasks based on the abnormal dimension deviation warning data to obtain dynamic production task allocation data; Step S44: updating the accessory production and processing data according to the quality control data and the production task dynamic allocation data to obtain the production process update data; Step S45: Upload the production process update data to the digital production management platform to perform digital visualization management tasks.

10. A digital visualization management system for parts production and processing, characterized in that: Used to execute a digital visualization management method for parts production and processing as claimed in claim 1, the digital visualization management system for parts production and processing comprises: The accessory quality defect analysis module is used to obtain accessory production and processing data, and extract production process features based on the accessory production and processing data to obtain production process data; perform accessory quality defect analysis on the production process data to obtain accessory quality defect data; The correlation analysis module is used to extract the production equipment status features based on the parts production and processing data, so as to obtain the equipment status data; the correlation analysis of the parts quality defect data is performed based on the equipment status data to obtain the equipment status-parts defect correlation data; The accessory abnormality intelligent early warning module is used to construct a quality defect prediction model based on the equipment status-accessory defect correlation data, thereby obtaining a quality defect prediction model; perform accessory abnormality intelligent early warning on the accessory production and processing data based on the quality defect prediction model, thereby obtaining accessory abnormality intelligent early warning data; The digital visualization management module is used to optimize the production scheduling of the parts production and processing data according to the intelligent warning data of the parts abnormality, obtain the production scheduling optimization data, and update the parts production and processing data according to the production scheduling optimization data to obtain the production process update data; upload the production process update data to the digital production management platform to perform digital visualization management tasks.

Images