Method and system for monitoring and evaluating machining state of mechanical part

Abstract

The invention discloses a mechanical part machining state monitoring and evaluating method and system, and relates to the technical field of machining state monitoring and evaluating, and the method comprises the steps: setting a machining parameter and a machining detection threshold value through the historical machining record of a mechanical part sample; analyzing the machining monitoring data of the mechanical part by using the machining detection threshold value, and outputting a round of monitoring result of the machining state of the mechanical part; the method comprises the following steps: performing sample extraction on a mechanical part, outputting a sampling inspection result, further judging whether a machining detection threshold adjustment mechanism is triggered or not, updating a machining detection threshold by utilizing a manually detected mechanical part machining record, and outputting a mechanical part machining state two-round monitoring result until the machining detection threshold adjustment mechanism is not triggered any more. And outputting a final mechanical part machining state evaluation report. Comprehensive real-time monitoring and dynamic threshold adjustment of the mechanical part machining process are achieved, the machining precision and efficiency are remarkably improved, and the stability and consistency of product quality are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of machining state monitoring and evaluation, and in particular to a method and system for monitoring and evaluating the machining state of a mechanical part. Background Art

[0002] As the manufacturing industry develops towards intelligence and automation, the precision and efficiency requirements of mechanical parts processing are constantly increasing. In order to improve the qualified rate of mechanical parts processing, modern manufacturing technology has gradually adopted closed-loop control systems to adjust the processing status through real-time monitoring and feedback mechanisms. However, the existing closed-loop system still has room for improvement in flexibility, real-time and detection accuracy.

[0003] Traditional processing status monitoring is often limited to certain key areas, ignoring the comprehensive monitoring of the edge and center areas of parts, resulting in the inability to detect potential defects in a timely manner. In addition, traditional processing status evaluation often relies on preset fixed detection thresholds, which are difficult to cope with changes in different part types, processing conditions or production environments, resulting in a lack of flexibility and accuracy in the detection results. Processing status evaluation mostly relies on manual inspection, which is not only inefficient but also easily affected by human subjective factors, resulting in unstable evaluation results. In addition, there is a lack of real-time feedback mechanism during the processing process, making it difficult to adjust processing parameters or process conditions in a timely manner to correct processing problems, resulting in the accumulation of processing defects. Many detection systems only focus on the early monitoring of the processing process, ignoring the subsequent closed-loop control of random inspections, comparisons and optimizations, resulting in inadequate quality management.

[0004] Therefore, in view of the above problems, a method and system for monitoring and evaluating the machining status of mechanical parts is urgently needed. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for monitoring and evaluating the machining status of mechanical parts, which solves the problems of incomplete traditional machining status monitoring, poor adaptability of fixed detection thresholds, and reliance on manual labor for quality assessment.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for monitoring and evaluating the processing status of mechanical parts, comprising the following steps: obtaining a processing picture of a mechanical part sample, and then identifying the historical processing record of the mechanical part sample according to the scanning result; using the historical processing record of the mechanical part sample to set processing parameters and processing detection thresholds; real-time monitoring of the mechanical part processing process and obtaining mechanical part processing monitoring data, specifically including edge area monitoring and center area monitoring; using the processing detection threshold to analyze the mechanical part processing monitoring data, and outputting a round of monitoring results of the mechanical part processing status, wherein the mechanical part processing status monitoring results include qualified and unqualified; and evaluating the mechanical part processing status. Samples are extracted and sampling results are output, including sampling results of qualified mechanical parts and sampling results of unqualified mechanical parts; the sampling results are counted, and the sampling results are compared with the first-round monitoring results, and it is determined whether the processing detection threshold adjustment mechanism is triggered according to the comparison results; in response to the triggering of the processing detection threshold adjustment condition, the processing detection threshold is updated using the manually inspected mechanical parts processing records; based on the output method of the first-round monitoring result of the mechanical parts processing status, the second-round monitoring results of the mechanical parts processing status are output using the updated processing detection threshold, and the sampling results of the second round of monitoring are compared until the processing detection threshold adjustment mechanism is no longer triggered, and a final mechanical parts processing status assessment report is output.

[0007] Furthermore, the specific analysis of setting processing parameters and processing detection thresholds using the historical processing records of the mechanical part sample is as follows: extracting a processing image of the mechanical part sample based on a comprehensive scan of the mechanical part sample; preprocessing the processing image of the mechanical part sample, the preprocessing including denoising, contrast enhancement, brightness adjustment and cropping; extracting mechanical part sample features based on the preprocessed processing image of the mechanical part sample; comparing the extracted mechanical part sample features with the mechanical part sample features stored in the mechanical part processing records to identify the type and model of the mechanical part sample; retrieving historical processing records of mechanical parts of the same type and model according to the identified mechanical part sample type and model, the historical processing records including processing parameters and processing status detection results; the processing status detection results Specifically, the qualified rate of mechanical parts processing is compared with the qualified rates of mechanical parts processing of each processing record, and the processing record with the highest qualified rate of mechanical parts processing is screened out; the screened out processing record is used as the reference record for the current mechanical parts processing, and then the processing parameters of the reference record are set as the processing parameters of the current mechanical parts processing, and the average value of qualified mechanical parts processing data in the reference record is set as the processing detection threshold of the current mechanical parts processing, the qualified average value of mechanical parts processing data in the reference record specifically includes the edge roughness average, the edge flatness average, the center residual stress average and the center area concentricity average, and the current processing detection threshold of mechanical parts processing specifically includes the edge roughness threshold, the edge flatness threshold, the center residual stress threshold and the center area concentricity threshold.

[0008] Furthermore, the processing detection threshold is used to analyze the mechanical parts processing monitoring data, and the specific analysis of the output of a round of monitoring results of the mechanical parts processing status is as follows: the mechanical parts processing monitoring data specifically includes edge area monitoring data and center area monitoring data, the edge area monitoring data includes edge roughness and edge flatness, and the center area monitoring data includes center residual stress and center area concentricity; the edge roughness, edge flatness, center residual stress and center area concentricity of each processed mechanical part are compared with the edge roughness threshold, edge flatness threshold, center residual stress threshold and center area concentricity threshold respectively, and the comparison results are output; the mechanical parts processing status one-round monitoring results whose edge roughness is less than the edge roughness threshold, edge flatness is less than the edge flatness threshold, center residual stress is less than the center residual stress threshold, and center area concentricity is less than the center area concentricity threshold in the comparison results are marked as qualified, and the remaining mechanical parts processing status one-round monitoring results are marked as unqualified.

[0009] Furthermore, the specific steps of sampling mechanical parts and outputting sampling results are: obtaining the sampling ratio, and dividing the mechanical parts into batches according to the processing time; randomly sampling each batch of mechanical parts based on the sampling ratio, and marking the mechanical parts sampling samples of each batch of mechanical parts; obtaining the sampling results of each batch of mechanical parts sampling samples based on manual inspection, and the sampling results specifically include whether the mechanical parts marked as qualified in the processing status monitoring results of the mechanical parts sampling samples are qualified in the manual inspection, and whether the mechanical parts marked as unqualified in the processing status monitoring results of the mechanical parts sampling samples are qualified in the manual inspection.

[0010] Furthermore, the sampling results are compared with the results of a round of monitoring, and the specific analysis of whether to trigger the processing detection threshold adjustment mechanism is determined based on the comparison results: for the mechanical parts in the mechanical parts sampling sample that are marked as qualified in the processing status monitoring results of a round, if the manual inspection shows that they are unqualified, they are marked as abnormal mechanical parts; for the mechanical parts in the mechanical parts sampling sample that are marked as unqualified in the processing status monitoring results of a round, if the manual inspection shows that they are qualified, they are marked as abnormal mechanical parts; the total amount of abnormal mechanical parts is counted, and then the total amount of abnormal mechanical parts is compared with the total amount threshold of abnormal mechanical parts. When the total amount of abnormal mechanical parts is less than the total amount threshold of abnormal mechanical parts, a final mechanical parts processing status assessment report is output, and the final mechanical parts processing status assessment report is specifically the sampling results of a round of monitoring results of the processing status of each mechanical part and each batch of mechanical parts sampling samples; when the total amount of abnormal mechanical parts is greater than or equal to the total amount threshold of abnormal mechanical parts, the processing detection threshold adjustment mechanism is triggered.

[0011] Furthermore, in response to the condition of triggering the adjustment of the processing detection threshold, the processing detection threshold is updated using the manually inspected mechanical parts processing records, specifically: obtaining the manually inspected mechanical parts processing records, the manually inspected mechanical parts processing records are the qualified conditions of the manually inspected mechanical parts; obtaining the average of the mechanical parts processing data marked as qualified by manual inspection, and then updating the average of the mechanical parts processing data marked as qualified by manual inspection to the processing detection threshold.

[0012] A mechanical part processing state monitoring and evaluation system, which applies the above-mentioned mechanical part processing state monitoring and evaluation method, includes: an image acquisition and recognition module, which is used to obtain the processing image of the mechanical part sample, and then identify the historical processing record of the mechanical part sample according to the scanning result; a parameter setting module, which is used to set the processing parameters and processing detection thresholds using the historical processing records of the mechanical part sample; a real-time monitoring module, which is used to monitor the mechanical part processing process in real time and obtain the mechanical part processing monitoring data, and the specific monitoring includes edge area monitoring and center area monitoring; a data analysis module, which is used to analyze the mechanical part processing monitoring data using the processing detection threshold, and output a round of monitoring results of the mechanical part processing state, and the mechanical part processing state monitoring results include qualified and unqualified; a sampling inspection module, It is used to extract samples of mechanical parts and output the sampling results, which include the sampling results of qualified mechanical parts and the sampling results of unqualified mechanical parts; the threshold adjustment judgment module is used to count the sampling results, compare the sampling results with the first-round monitoring results, and judge whether to trigger the processing detection threshold adjustment mechanism according to the comparison results; the threshold update module is used to update the processing detection threshold according to the condition of triggering the processing detection threshold adjustment by using the processing records of mechanical parts inspected manually; the re-inspection module is used to output the second-round monitoring results of the processing status of mechanical parts based on the output method of the first-round monitoring results of the processing status of mechanical parts by using the updated processing detection threshold, and compare the sampling results of the second round of monitoring until the processing detection threshold adjustment mechanism is no longer triggered, and the final mechanical parts processing status evaluation report is output.

[0013] The present invention has the following beneficial effects:

[0014] The present invention discloses a method and system for monitoring and evaluating the machining state of mechanical parts. Through real-time monitoring and machining parameters and detection thresholds set according to historical machining records, potential problems can be immediately discovered during the machining process, machining defects can be avoided, and machining accuracy and stability of mechanical parts can be improved. Through automated machining state monitoring, data in the machining process can be collected and analyzed in real time, the need for manual detection can be reduced, the intelligent level of the machining process can be improved, and the possibility of human error or omission can be reduced. By comparing the sampling results with the results of a round of monitoring, a machining detection threshold adjustment mechanism can be triggered to achieve dynamic optimization of the machining detection threshold, so that the machining process can be more flexibly adapted to different machining conditions, and the machining quality can be further improved. By comparing the results of the second round of monitoring with the updated machining detection threshold, the machining state can be more accurately evaluated, and the evaluation report finally output can provide a more reliable basis for judging the machining quality. Through monitoring of the edge area and the center area, potential problems of mechanical parts in different areas can be fully captured, missed detection can be reduced, and unqualified products can be avoided from flowing into the production link, thereby reducing the risk of unqualified products. When processing different mechanical part samples, machining parameters and detection standards can be flexibly adjusted according to historical machining records to adapt to different production environments and process requirements, thereby improving the versatility and flexibility of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The present invention is a flow chart of a method for monitoring and evaluating the machining status of mechanical parts.

[0016] Figure 2 This is a structural diagram of a mechanical parts processing status monitoring and evaluation system of the present invention. DETAILED DESCRIPTION

[0017] The embodiment of the present application realizes comprehensive real-time monitoring and dynamic threshold adjustment of the mechanical parts processing process through a mechanical parts processing status monitoring and evaluation method and system, significantly improves the processing accuracy and efficiency, and ensures the stability and consistency of product quality.

[0018] The overall idea of ​​the embodiments of this application is:

[0019] By obtaining the processing pictures of the mechanical parts samples, the scanning results are used to identify and extract the historical processing records of the parts, which provides basic data for the subsequent processing parameter setting and the determination of the detection threshold. Based on the historical processing records, the processing parameters and detection thresholds are set to guide the subsequent processing process; during the processing process, the mechanical parts are monitored in real time, including specific monitoring of the edge area and the center area to ensure that every part of the processing process is under control; using the set processing detection threshold, the monitoring data is analyzed and the monitoring results of the processing status are output; sampling inspection is carried out on the processed parts, and the sampling inspection results are compared with the monitoring results to verify the accuracy of the processing status assessment; if there is a difference between the sampling inspection results and the monitoring results, the threshold adjustment mechanism is triggered, and the detection threshold is updated through the records of manual inspection to further improve the accuracy of subsequent monitoring; based on the updated threshold, the second round of monitoring and sampling comparison is continued until the adjustment mechanism is no longer triggered, and finally a reliable processing status assessment report is output.

[0020] See also Figure 1 The embodiment of the present invention provides a technical solution: a method for monitoring and evaluating the machining status of mechanical parts, comprising the following steps: obtaining a machining picture of a mechanical part sample, and then identifying the historical machining record of the mechanical part sample according to the scanning result; using the historical machining record of the mechanical part sample to set machining parameters and machining detection thresholds; real-time monitoring of the machining process of the mechanical part and obtaining machining monitoring data of the mechanical part, specifically monitoring includes edge area monitoring and center area monitoring; using the machining detection threshold to analyze the machining monitoring data of the mechanical part, and outputting a round of monitoring results of the machining status of the mechanical part, wherein the machining status of the mechanical part includes qualified and unqualified; sampling the mechanical parts Obtain and output sampling results, including sampling results of qualified mechanical parts and sampling results of unqualified mechanical parts; compile statistics on sampling results, compare sampling results with the first-round monitoring results, and determine whether to trigger the processing detection threshold adjustment mechanism based on the comparison results; use the manually inspected mechanical parts processing records to update the processing detection threshold for the triggered processing detection threshold adjustment condition; based on the output method of the first-round monitoring results of the mechanical parts processing status, use the updated processing detection threshold to output the second-round monitoring results of the mechanical parts processing status, and compare the sampling results of the second-round monitoring until the processing detection threshold adjustment mechanism is no longer triggered, and output the final mechanical parts processing status assessment report.

[0021] Specifically, the specific analysis of setting processing parameters and processing detection thresholds using the historical processing records of the mechanical part sample is as follows: extracting processing images of the mechanical part sample based on a comprehensive scan of the mechanical part sample; preprocessing the processing images of the mechanical part sample, the preprocessing includes denoising, contrast enhancement, brightness adjustment and cropping; extracting mechanical part sample features based on the preprocessed processing images of the mechanical part sample; comparing the extracted mechanical part sample features with the mechanical part sample features stored in the mechanical part processing records to identify the type and model of the mechanical part sample; retrieving historical processing records of mechanical parts of the same type and model based on the identified mechanical part sample type and model, the historical processing records including processing parameters and processing status detection results; the processing status detection results are specific. The qualified rate of mechanical parts processing is compared with the qualified rates of mechanical parts processing of each processing record, and the processing record with the largest qualified rate of mechanical parts processing is screened out; the screened out processing record is used as a reference record for the current mechanical parts processing, and then the processing parameters of the reference record are set as the processing parameters of the current mechanical parts processing, and the average value of qualified mechanical parts processing data in the reference record is set as the processing detection threshold of the current mechanical parts processing, the average value of qualified mechanical parts processing data in the reference record specifically includes the edge roughness average, the edge flatness average, the center residual stress average and the center area concentricity average, and the current processing detection threshold of mechanical parts processing specifically includes the edge roughness threshold, the edge flatness threshold, the center residual stress threshold and the center area concentricity threshold.

[0022] In this embodiment, processing parameters refer to various parameters used to control processing conditions during mechanical processing, such as rotation speed, feed speed, cutting depth, tool selection, etc. Different processing parameters affect processing efficiency and processing quality. Processing detection threshold refers to the standard threshold for detecting certain key quality indicators during the processing. If the detection result exceeds the threshold, the processing process is considered abnormal or unqualified.

[0023] Edge roughness is used to measure the roughness of the surface quality of the edge of a part. It is usually assessed by slight fluctuations in the surface profile and is a key indicator of the edge processing accuracy of the part, affecting the assembly performance and aesthetics of the part. Edge flatness refers to the flatness of the edge of the part. It is used to assess whether the edge has obvious protrusions or depressions, affecting the assembly accuracy and sealing of the part. Center residual stress refers to the stress that cannot be completely eliminated due to external forces during the processing process, affecting its service life and stability. Center area concentricity refers to the concentricity of the axes in the center area of ​​the part. It is used to assess whether the processing has caused deviation. If the concentricity of the center area is too low, it will affect the functionality and fitting accuracy of the part.

[0024] The edge roughness threshold is used to limit the maximum allowable value of edge roughness. If this value is exceeded, the processing is considered unqualified; the edge flatness threshold is used to limit the maximum allowable unevenness of the edge. If this threshold is exceeded, it is judged as unqualified; the center residual stress threshold is used to limit the maximum allowable value of the residual stress in the center area. Exceeding this threshold indicates that there may be processing quality problems; the center area concentricity threshold is used to limit the maximum tolerance of the concentricity of the center area. Exceeding this value may affect the accuracy of the part.

[0025] The monitoring data of edge roughness, edge flatness, center residual stress and center area concentricity are collected in real time through various processing monitoring equipment (such as surface roughness meter, flatness meter, residual stress tester, concentricity meter, etc.).

[0026] The specific logical steps for extracting the features of mechanical parts samples based on the processing images of preprocessed mechanical parts samples are as follows: use denoising algorithms (such as median filtering, Gaussian filtering, etc.) to remove noise in the processing images to ensure image clarity; use contrast stretching or histogram equalization and other techniques to enhance the contrast of the image to make the processing features more prominent; adjust the brightness of the image according to actual conditions to avoid the impact of lighting changes; crop out the feature area of ​​the parts and remove irrelevant parts to improve the efficiency of subsequent processing; use edge detection algorithms (such as Canny edge detection) to extract the shape contour or processing edge features of the parts; analyze the surface texture of the parts through the grayscale co-occurrence matrix of the image to extract surface roughness, flatness and other features; extract the geometric shape of the parts and extract features such as roundness, symmetry, flatness, etc.; extract concentricity features through geometric shape deviation analysis to evaluate the concentricity of the parts; indirectly infer the residual stress distribution during the processing process through traces such as cracks and deformations in the image.

[0027] The type and model identification method is: compare the extracted features with the sample features in the historical records, and identify the type and model of the part through similarity metrics (such as Euclidean distance, cosine similarity, etc.).

[0028] By comparing and screening historical processing records, the optimal processing parameters can be automatically selected, thereby reducing manual intervention and reliance on experience; qualified data in historical processing records are used to set detection thresholds to ensure that the quality of each processing meets the requirements and reduce the production of unqualified products; processing parameters are automatically adjusted to reduce fluctuations during the processing process, thereby improving production efficiency and ensuring long-term processing stability; through automated feature extraction and parameter setting, the need for manual intervention is greatly reduced, while the accuracy of processing detection is improved; based on a large number of historical processing records, real-time decision optimization can be performed based on actual processing data, which not only ensures processing accuracy, but also makes the production process more intelligent.

[0029] Specifically, the processing detection threshold is used to analyze the mechanical parts processing monitoring data, and the specific analysis of the output of a round of monitoring results of the mechanical parts processing status is as follows: the mechanical parts processing monitoring data specifically includes edge area monitoring data and center area monitoring data, the edge area monitoring data includes edge roughness and edge flatness, and the center area monitoring data includes center residual stress and center area concentricity; the edge roughness, edge flatness, center residual stress and center area concentricity of each processed mechanical part are compared with the edge roughness threshold, edge flatness threshold, center residual stress threshold and center area concentricity threshold respectively, and the comparison results are output; the mechanical parts processing status one-round monitoring results whose edge roughness is less than the edge roughness threshold, edge flatness is less than the edge flatness threshold, center residual stress is less than the center residual stress threshold, and center area concentricity is less than the center area concentricity threshold in the comparison results are marked as qualified, and the remaining mechanical parts processing status one-round monitoring results are marked as unqualified.

[0030] By setting clear processing detection thresholds and comparing the actual monitoring data with these thresholds, it is possible to automatically determine whether the parts are qualified, reducing the need for manual intervention and the errors and inconsistencies that may be caused by manual judgment; real-time monitoring and analysis of the four key indicators in the processing process can accurately identify potential problems in the processing, so that corrective measures can be taken in time to prevent unqualified products from entering the production line or the market; by setting reasonable processing thresholds and performing qualification inspections on each batch of parts, the product failure rate caused by errors and fluctuations in the processing process can be effectively reduced, thereby improving production efficiency and product quality; automated detection and real-time feedback mechanisms enable the production line to quickly identify problems, reduce the subsequent processing time of unqualified parts, and determine whether the parts are qualified in real time, so that decisions can be made quickly, production scheduling and resource allocation can be optimized, and overall production efficiency can be improved; with the accumulation of monitoring data, it can be optimized based on historical data to provide more accurate processing parameter settings and quality standard adjustments for future processing processes; by reducing waste such as rework and repair caused by quality problems, it helps to reduce production costs and improve the unit production cost efficiency by increasing the qualification rate.

[0031] Specifically, the specific steps of sampling mechanical parts and outputting the inspection results are: obtaining the inspection ratio, and dividing the mechanical parts into batches according to the processing time; randomly sampling each batch of mechanical parts based on the inspection ratio, and marking the mechanical parts sampling samples of each batch of mechanical parts; obtaining the sampling results of each batch of mechanical parts sampling samples based on manual inspection, and the sampling results specifically include whether the mechanical parts marked as qualified in the processing status monitoring results of the mechanical parts sampling samples are qualified in the manual inspection, and whether the mechanical parts marked as unqualified in the processing status monitoring results of the mechanical parts sampling samples are qualified in the manual inspection.

[0032] In this implementation scheme, the sampling ratio is set as follows: if the batch is small (such as a few hundred parts), the sampling ratio can be relatively high because the sample is more representative. When the batch is large (such as tens of thousands of parts), the sampling ratio can be appropriately reduced, but it is still necessary to ensure that the sample is sufficiently representative. If the quality requirements of the parts are strict (such as high precision or high safety requirements), the sampling ratio should be increased to ensure that the quality of each batch meets the standards. If the quality requirements of the parts are relatively loose (such as ordinary consumer products), the sampling ratio can be appropriately reduced. If the qualified rate of the production line is low, the sampling ratio should be increased to ensure that potential quality problems are discovered in time. If historical data shows that the qualified rate is high and the processing process is stable, the sampling ratio can be appropriately reduced. The sample extraction ratio can also be determined by statistical methods, such as normal distribution, binomial distribution, etc., or a reasonable sampling ratio can be set based on historical quality data and experience, usually set to 5%, 10%, 20%, etc.

[0033] The specific steps for dividing mechanical parts into batches according to processing time are as follows: according to the production rhythm or time period of the production line, the parts in the production process are divided into different batches according to the processing time. The batch division can be based on days, hours or specific production cycles (such as one batch every 10 hours or every 24 hours) to ensure the consistency of the number of parts and production time in each batch. The start time and end time of each production batch are specifically recorded. Within each production cycle, a reasonable batch range is determined based on the processing progress of the production line and the working status of the equipment. The number of parts that each batch should include can be set according to the production plan and production capacity.

[0034] By conducting sampling inspections in each batch, the number of parts that require manual inspection can be effectively reduced, thereby improving the efficiency of quality control. The feedback of the sampling inspection results of each batch can reflect the overall processing quality of the batch without the need to conduct full inspections on every part. Compared with full inspections, sampling inspections greatly reduce the workload and time of manual inspections and reduce inspection costs. At the same time, the sampling ratio can be adjusted as needed to ensure that costs are controlled while ensuring quality. By sampling and inspecting each batch of mechanical parts, potential problems in the processing process can be discovered in a timely manner. If a batch of samples shows unqualified results, the production process can be quickly adjusted to reduce the generation of defective products. Combining the sampling results with batch processing records can help companies track quality problems in the production process and provide a basis for subsequent quality improvements. By comparing with manual inspection results, the differences between automatic inspection systems and manual inspections can be further evaluated to help improve the accuracy of the quality management system.

[0035] Specifically, the sampling results are compared with the results of a round of monitoring, and the specific analysis of whether to trigger the processing detection threshold adjustment mechanism is determined based on the comparison results: for mechanical parts in the mechanical parts sampling sample that are marked as qualified in the processing status monitoring results of a round, if the manual inspection shows that they are unqualified, they are marked as abnormal mechanical parts; for mechanical parts in the mechanical parts sampling sample that are marked as unqualified in the processing status monitoring results of a round, if the manual inspection shows that they are qualified, they are marked as abnormal mechanical parts; the total amount of abnormal mechanical parts is counted, and then the total amount of abnormal mechanical parts is compared with the total amount threshold of abnormal mechanical parts. When the total amount of abnormal mechanical parts is less than the total amount threshold of abnormal mechanical parts, the final mechanical parts processing status evaluation report is output. The final mechanical parts processing status evaluation report specifically includes the results of a round of monitoring of the processing status of each mechanical part and the sampling results of each batch of mechanical parts sampling samples; when the total amount of abnormal mechanical parts is greater than or equal to the total amount threshold of abnormal mechanical parts, the processing detection threshold adjustment mechanism is triggered.

[0036] In this implementation scheme, the method for setting the total amount threshold of abnormal mechanical parts is specifically as follows: in combination with the historical quality inspection data in the production process (such as the unqualified rate, the proportion of abnormal parts, etc.), a reasonable total amount threshold of abnormal parts is set by analyzing the historical production data through statistical methods. For example, in the past production, the number of unqualified parts in every 1,000 parts is 20 (i.e., a 2% unqualified rate), and the threshold can be set to 2% of the number of unqualified parts; if the production quantity of a batch is very large, the threshold of abnormal parts can be appropriately increased, and if the batch is small, the threshold can be appropriately lowered. For example, for a production batch of 1,000 parts, if there are more than 5 unqualified parts, the threshold adjustment mechanism is triggered, and for a batch of 10,000 parts, more unqualified parts may be tolerated; for parts with high precision requirements, the threshold should be set more strictly; and for parts with low precision or a larger tolerance, the threshold can be relatively loose.

[0037] By comparing the results of a round of monitoring with those of manual inspection, the differences between automatic monitoring and manual inspection can be identified, which helps to timely discover potential errors or omissions in the monitoring system and optimize the monitoring algorithm. If a round of monitoring fails to correctly mark certain problematic parts (such as parts marked as qualified are actually problematic), these potential misjudgments can be discovered through manual inspection and comparison, thus preventing unqualified products from entering subsequent production links or leaving the factory. By marking "abnormal mechanical parts", quality problems that may occur during the processing process can be quickly identified, and these abnormal parts can be tracked and processed, which can help managers target specific batches or parts. Conduct in-depth investigations to prevent problems from spreading. When the number of abnormal parts reaches a threshold, the threshold adjustment mechanism is triggered, prompting the production line or quality monitoring system to make necessary adjustments (for example, changing processing parameters, checking equipment accuracy, etc.) to solve quality problems at the source. By counting the total number of abnormal mechanical parts and automatically triggering the adjustment of the detection threshold when the total amount reaches a certain threshold, the quality control standards in the processing process can be dynamically optimized based on actual production data. The processing status and sampling test results of each batch of mechanical parts are recorded and marked as "abnormal mechanical parts" so that every link in the production process can be traced, reducing the risk of potential quality risks spreading.

[0038] Specifically, in order to trigger the adjustment of the processing detection threshold, the processing detection threshold is updated using the manually inspected mechanical parts processing records. Specifically, the manually inspected mechanical parts processing records are obtained, and the manually inspected mechanical parts processing records are the qualified conditions of the manually inspected mechanical parts; the mean of the mechanical parts processing data marked as qualified by manual inspection is obtained, and then the mean of the mechanical parts processing data marked as qualified by manual inspection is updated to the processing detection threshold.

[0039] In this implementation scheme, the processing detection threshold is continuously updated through manual inspection results, and can be dynamically adjusted according to the actual inspection results, rather than sticking to the initially set standards. The adaptive design enables the detection threshold to more accurately reflect the actual processing quality in different production stages, different batches or different working conditions; manual inspection usually has high accuracy, especially when errors or deficiencies occur in the detection system, using the qualified status of manual inspection and the data mean to adjust the threshold can effectively reduce detection errors, optimize error tolerance, reduce the risk of missed detection or misjudgment, and help maintain long-term stability of quality in the production process; updating the processing detection threshold can timely adjust the threshold standard to avoid Avoid quality problems caused by fluctuations; by updating the processing detection threshold through manual inspection processing records, the production line can make flexible adjustments according to the actual processing conditions, rather than relying on fixed processing standards. This helps the production line to quickly adapt and ensure quality when facing different types of parts, different production batches or new processes, avoiding rework, repairs or waste disposal due to quality problems, thereby saving production costs; the mean of manual inspection data can eliminate the accidental errors of single data points and provide a more stable reference value. By using these means as the basis for updating, it can effectively avoid detection errors caused by extreme cases or occasional problems, and improve the stability of threshold updates.

[0040] See also Figure 2A mechanical part processing state monitoring and evaluation system, using the above-mentioned mechanical part processing state monitoring and evaluation method, includes: an image acquisition and recognition module, used to obtain the processing image of the mechanical part sample, and then identify the historical processing record of the mechanical part sample according to the scanning result; a parameter setting module, used to use the historical processing record of the mechanical part sample to set the processing parameters and processing detection threshold; a real-time monitoring module, used to monitor the mechanical part processing process in real time and obtain the mechanical part processing monitoring data, and the specific monitoring includes edge area monitoring and center area monitoring; a data analysis module, used to analyze the mechanical part processing monitoring data using the processing detection threshold, and output a round of monitoring results of the mechanical part processing state, and the mechanical part processing state monitoring results include qualified and unqualified; a sampling inspection module, used to The module is used to extract samples of mechanical parts and output sampling results, which include sampling results of qualified mechanical parts and sampling results of unqualified mechanical parts; the threshold adjustment judgment module is used to count the sampling results, compare the sampling results with the first-round monitoring results, and judge whether to trigger the processing detection threshold adjustment mechanism according to the comparison results; the threshold update module is used to update the processing detection threshold according to the condition of triggering the processing detection threshold adjustment by using the processing records of mechanical parts inspected manually; the re-inspection module is used to output the second-round monitoring results of the processing status of mechanical parts based on the output method of the first-round monitoring results of the processing status of mechanical parts by using the updated processing detection threshold, and compare the sampling results of the second round of monitoring until the processing detection threshold adjustment mechanism is no longer triggered, and the final mechanical parts processing status evaluation report is output.

[0041] In summary, this application has at least the following effects:

[0042] By monitoring the real-time processing status of mechanical parts and combining historical processing records with set processing detection thresholds, potential quality problems can be discovered in time during the processing process, and precise control of processing accuracy can be achieved; by analyzing various types of data during the processing process and continuously adjusting the processing detection thresholds through machine learning and other means, intelligent evaluation of the processing status can be achieved, reducing the need for manual intervention and improving the automation level of the processing process; based on the comparison of monitoring data and sampling results, the processing detection threshold can be dynamically adjusted to effectively adapt to different processing states and workpiece characteristics, avoiding misjudgments or missed judgments caused by fixed threshold settings, thereby improving detection accuracy; by relying on data-driven automatic evaluation to reduce manual inspection intervention, and by comparing monitoring data and sampling results, human errors and deviations can be effectively reduced to ensure consistency and stability of processing quality;

[0043] Through real-time monitoring and automatic adjustment of detection thresholds, potential problems in the processing process can be identified and resolved in advance, avoiding rework or waste, thereby reducing production costs and improving production efficiency. Continuous optimization based on evaluation results can further improve the production line's capacity.

[0044] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods and systems. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0045] The present invention is described with reference to the flowcharts and structure diagrams of the methods and systems according to the embodiments of the present invention. It should be understood that each process and combination of modules in the flowcharts and structure diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 process or processes and structures Figure 1 A device that specifies functionality in a module or modules.

[0046] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 process or processes and structures Figure 1 A function specified in a module or multiple modules.

[0047] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 process or processes and structures Figure 1 Steps to specify functionality in a module or multiple modules.

[0048] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0049] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for monitoring and evaluating the machining status of mechanical parts, characterized in that: The following steps are involved: Obtain a processing picture of a mechanical part sample, and then identify the historical processing record of the mechanical part sample based on the scanning result; Using the historical processing records of the mechanical parts sample to set processing parameters and processing detection thresholds; Real-time monitoring of the machining process of mechanical parts and obtaining the monitoring data of mechanical parts machining, including edge area monitoring and center area monitoring; Analyze the mechanical parts processing monitoring data using the processing detection threshold, and output a round of monitoring results of the mechanical parts processing status, wherein the mechanical parts processing status monitoring results include qualified and unqualified; Sampling mechanical parts and outputting sampling results, the sampling results including the sampling results of qualified mechanical parts and the sampling results of unqualified mechanical parts; Collect the sampling inspection results, compare them with the results of a round of monitoring, and determine whether to trigger the processing detection threshold adjustment mechanism based on the comparison results; In response to the adjustment of the trigger processing detection threshold, the processing detection threshold is updated using the processing records of mechanical parts inspected manually; Based on the output method of the first-round monitoring results of the mechanical parts processing status, the updated processing detection threshold is used to output the second-round monitoring results of the mechanical parts processing status, and the sampling results of the second round of monitoring are compared until the processing detection threshold adjustment mechanism is no longer triggered, and the final mechanical parts processing status assessment report is output.

2. A method for monitoring and evaluating the machining status of mechanical parts according to claim 1, characterized in that: The specific analysis of using the historical processing records of the mechanical parts sample to set the processing parameters and processing detection thresholds is as follows: Extract processing pictures of mechanical parts samples based on comprehensive scanning of mechanical parts samples; Preprocess the processed images of mechanical parts samples, including denoising, contrast enhancement, brightness adjustment and cropping; Extracting features of mechanical parts samples based on preprocessed processing images of mechanical parts samples; Compare the extracted mechanical part sample features with the mechanical part sample features stored in the mechanical part processing record to identify the type and model of the mechanical part sample; Retrieve historical processing records of mechanical parts of the same type and model according to the identified mechanical parts sample type and model, the historical processing records including processing parameters and processing status detection results; The processing state detection result is specifically the processing qualification rate of mechanical parts. The processing qualification rates of mechanical parts of various processing records are compared to select the processing record with the highest processing qualification rate of mechanical parts; The screened processing record is used as a reference record for the current mechanical part processing, and then the processing parameters of the reference record are set as the processing parameters of the current mechanical part processing, and the average value of qualified mechanical part processing data in the reference record is set as the processing detection threshold of the current mechanical part processing. The average value of qualified mechanical part processing data in the reference record specifically includes the edge roughness average, the edge flatness average, the center residual stress average and the center area concentricity average. The processing detection threshold of the current mechanical part processing specifically includes the edge roughness threshold, the edge flatness threshold, the center residual stress threshold and the center area concentricity threshold.

3. A method for monitoring and evaluating the machining status of mechanical parts according to claim 2, characterized in that: The processing detection threshold is used to analyze the mechanical parts processing monitoring data, and the specific analysis of the output of a round of monitoring results of the mechanical parts processing status is as follows: The mechanical parts processing monitoring data specifically includes edge area monitoring data and center area monitoring data, the edge area monitoring data includes edge roughness and edge flatness, and the center area monitoring data includes center residual stress and center area concentricity; The edge roughness, edge flatness, center residual stress and center area concentricity of each machined part are respectively compared with the edge roughness threshold, edge flatness threshold, center residual stress threshold and center area concentricity threshold, and the comparison results are output; The one-round monitoring results of the machining status of mechanical parts whose edge roughness in the comparison results is less than the edge roughness threshold, the edge flatness is less than the edge flatness threshold, the center residual stress is less than the center residual stress threshold, and the center area concentricity is less than the center area concentricity threshold are marked as qualified, and the remaining mechanical parts processing status monitoring results are marked as unqualified.

4. A method for monitoring and evaluating the machining status of mechanical parts according to claim 2, characterized in that: The specific steps of sampling mechanical parts and outputting sampling results are as follows: Obtain sampling ratio and divide mechanical parts into batches according to processing time; Randomly sample each batch of mechanical parts based on the sampling ratio, and mark the sampling samples of mechanical parts in each batch of mechanical parts; The sampling results of each batch of mechanical parts sampling samples are obtained based on manual inspection. The sampling results specifically include whether the mechanical parts marked as qualified in a round of monitoring results of the processing status in the mechanical parts sampling samples are qualified in manual inspection, and whether the mechanical parts marked as unqualified in a round of monitoring results of the processing status in the mechanical parts sampling samples are qualified in manual inspection.

5. A method for monitoring and evaluating the machining status of mechanical parts according to claim 4, characterized in that: The sampling results are compared with the results of a round of monitoring, and the specific analysis of whether to trigger the processing detection threshold adjustment mechanism is based on the comparison results: For the mechanical parts sampled from the mechanical parts, if the mechanical parts are marked as qualified in a round of monitoring results of the processing status, they are marked as abnormal mechanical parts if they are found to be unqualified in manual inspection; For the mechanical parts sampled from the mechanical parts, if the mechanical parts are marked as unqualified in a round of monitoring results of the processing status, they are marked as abnormal mechanical parts if they are qualified in manual inspection; Counting the total amount of abnormal mechanical parts, and then comparing the total amount of abnormal mechanical parts with the total amount threshold of abnormal mechanical parts, when the total amount of abnormal mechanical parts is less than the total amount threshold of abnormal mechanical parts, outputting the final mechanical parts processing status assessment report, wherein the final mechanical parts processing status assessment report specifically includes a round of monitoring results of the processing status of each mechanical part and the sampling results of each batch of mechanical parts sampling samples; When the total amount of abnormal mechanical parts is greater than or equal to the total amount threshold of abnormal mechanical parts, the processing detection threshold adjustment mechanism is triggered.

6. A method for monitoring and evaluating the machining status of mechanical parts according to claim 1, characterized in that: In view of the adjustment of the trigger processing detection threshold, the processing detection threshold is updated using the processing records of mechanical parts detected manually as follows: Obtaining a manually inspected mechanical part processing record, wherein the manually inspected mechanical part processing record is a qualified condition of the manually inspected mechanical part; The mean value of machining data of mechanical parts marked as qualified by manual inspection is obtained, and then the mean value of machining data of mechanical parts marked as qualified by manual inspection is updated as the machining inspection threshold.

7. A mechanical parts processing state monitoring and evaluation system, using a mechanical parts processing state monitoring and evaluation method according to any one of claims 1 to 6, characterized in that: include: The image acquisition and recognition module is used to obtain the processing image of the mechanical part sample, and then identify the historical processing record of the mechanical part sample according to the scanning result; A parameter setting module, used to set processing parameters and processing detection thresholds using the historical processing records of the mechanical part sample; Real-time monitoring module, used to monitor the machining process of mechanical parts in real time and obtain mechanical parts machining monitoring data. Specific monitoring includes edge area monitoring and center area monitoring; A data analysis module, used to analyze the mechanical parts processing monitoring data using the processing detection threshold, and output a round of monitoring results of the mechanical parts processing status, wherein the mechanical parts processing status monitoring results include qualified and unqualified; The sampling inspection module is used to sample mechanical parts and output the sampling inspection results, which include the sampling inspection results of qualified mechanical parts and the sampling inspection results of unqualified mechanical parts; The threshold adjustment judgment module is used to collect statistics on the sampling results, compare the sampling results with the results of a round of monitoring, and determine whether to trigger the processing detection threshold adjustment mechanism based on the comparison results; A threshold update module, for updating the processing detection threshold by using the processing records of the mechanical parts detected manually according to the triggering processing detection threshold adjustment condition; The re-inspection module is used to output the second-round monitoring results of the machining status of mechanical parts based on the output method of the first-round monitoring results of the machining status of mechanical parts using the updated machining detection threshold, and compare the sampling results of the second round of monitoring until the machining detection threshold adjustment mechanism is no longer triggered, and output the final mechanical parts machining status evaluation report.

Images