Automobile part production quality management and control method and system based on classification management
Through image data acquisition and BP neural network classification, combined with quality detection and clustering algorithm, the problem of processing that the quality of parts does not meet the requirements in the existing technology is solved, the fine classification and efficient quality control of parts are realized, and the production process is improved accuracy and accuracy.
Patent Information
- Application Number
- CN202411361300.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-27
- Publication Date
- 2024-12-31
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the production process of existing automotive parts, parts that do not meet the requirements cannot be reprocessed, and parts that are qualified and unqualified in detail cannot be classified, which affects the accuracy of production quality control.
The image data of parts is collected through the image data acquisition module, the BP neural network is used for classification, the corresponding quality detection device is selected for detection, the components that do not meet the requirements are reprocessed, and the quality level is determined through the clustering algorithm to realize the components classification management.
It improves the accuracy of component production quality control, avoids waste of parts, ensures the accuracy of quality inspection, and realizes fine classification management of parts, improving the accuracy of production quality control.
Smart Images
Figure CN119227957A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of parts quality control, and in particular to a method and system for controlling the production quality of automobile parts based on classification management. Background Art
[0002] As a common means of transportation in modern times, cars have gradually replaced livestock over the years and become an indispensable part of modern life. Over the past few hundred years, the achievements of the times have been fully demonstrated in the development of cars, from the initial steam engine to the internal combustion engine to the pure electric smart car promoted today.
[0003] As the foundation of the automobile industry, auto parts are an essential factor to support the sustainable and healthy development of the automobile industry. In particular, the current independent development and innovation in the automobile industry, which is in full swing, requires a strong parts system to support it. The independent brand and technological innovation of the whole vehicle need parts as the basis, and the independent innovation of parts has a strong driving force for the development of the whole vehicle industry. They influence and interact with each other. Without the independent brand of the whole vehicle, the R&D and innovation capabilities of the strong parts system will be difficult to burst out. Without the support of a strong parts system, it will be difficult for the independent brand to grow bigger and stronger.
[0004] During the production process of existing automobile parts, corresponding quality inspection methods are adopted according to the categories of automobile parts to determine whether the automobile parts are qualified for quality control. However, in the above process, automobile parts that do not meet the quality requirements cannot be reprocessed, and it is also impossible to classify automobile parts that do not meet the quality requirements and automobile parts that meet the quality requirements more finely, which is not conducive to the accuracy of production quality control of automobile parts and affects the effectiveness of production quality control of automobile parts. Summary of the invention
[0005] 1. Technical issues to be resolved In order to address the deficiencies in the background technology, the present invention provides a method and system for automobile parts production quality control based on classification management, which determines whether the produced automobile parts are qualified through a quality inspection device, selects automobile parts that meet the rework requirements from automobile parts that do not meet the quality requirements, reprocesses them, and re-inspects their quality, and finally determines the quality grade of automobile parts through a clustering algorithm, implements classification management of automobile parts according to their quality grade, and improves the accuracy of the automobile parts production quality control process.
[0006] (II) Technical solution A method for controlling the production quality of automobile parts based on classification management, the method comprising the following steps: S1. Collecting image data of produced automobile parts through an image data acquisition module, inputting the image data of the produced automobile parts into a trained BP neural network model for image recognition, and classifying the produced automobile parts according to the categories of automobile parts according to the recognition results; S2. Selecting corresponding quality inspection devices according to the categories of automobile parts to perform quality inspection on automobile parts, and obtaining automobile parts that do not meet quality requirements and automobile parts that meet quality requirements; S3. Analyze the data of the automobile parts that do not meet the quality requirements, select the automobile parts that meet the rework conditions, and reprocess the automobile parts; S4. Conduct quality inspection on the reprocessed auto parts to obtain auto parts that do not meet quality requirements and auto parts that meet quality requirements; S5. Cluster all automobile parts data in each category separately. According to the clustering results, obtain the quality grade of automobile parts in each category, match all automobile parts that have passed quality inspection with corresponding automobile parts quality grades, and according to the quality grades of automobile parts, transport all automobile parts that do not meet the quality requirements and all automobile parts that meet the quality requirements into the warehouse separately to realize further classification management.
[0007] By inputting the image data of the produced automobile parts into the trained BP neural network model for image recognition, the produced automobile parts are classified according to the categories of automobile parts according to the recognition results, thereby improving the accuracy of classification according to the categories of automobile parts; using quality inspection devices to conduct quality inspection on automobile parts, and selecting automobile parts that meet the rework conditions among the automobile parts that do not meet the quality requirements for reprocessing, and re-quality inspection, while avoiding waste of automobile parts, the accuracy of quality inspection is guaranteed; all automobile parts data in each category are clustered separately, and according to the clustering results, all automobile parts that have passed quality inspection are matched with corresponding automobile parts quality grades, and automobile parts to be stored are transported into the warehouse according to the quality grades of automobile parts, so as to realize classification management, ensure the accuracy of the automobile parts production quality control process, and pass the accuracy of automobile parts production quality control results.
[0008] Preferably, the specific operation steps of collecting the image data of the produced automobile parts through the image data collection module and inputting the image data of the produced automobile parts into the trained BP neural network model for image recognition are as follows: S11, collecting image data of all categories of automobile parts in the warehouse through the image data collection module, and performing model training using the BP neural network algorithm; S111, set the training data ratio to , the test data ratio is , according to the training data ratio And test data ratio Performing data division on the image data of all categories of automobile parts in the warehouse to obtain training image data of all categories of automobile parts in the warehouse and test image data of all categories of automobile parts in the warehouse; S112, constructing an initial BP neural network model, setting the number of input nodes of the initial BP neural network model to The number of hidden nodes is The number of output nodes is The initial weight is And the initial threshold is ; S113, setting a training error threshold, inputting the training image data of all categories of automobile parts in the warehouse into the initial BP neural network model to perform a training operation on the initial BP neural network model, and when the training error during the training process is less than or equal to the training error threshold, stopping the training to obtain the trained BP neural network model; S114, setting an accuracy threshold, inputting the test image data of all categories of automobile parts in the warehouse into the initial BP neural network model for testing, and obtaining the test accuracy after the test is completed. When the test accuracy is greater than or equal to the accuracy threshold, the trained BP neural network model is used as the final BP neural network model; otherwise, the initial weights of the trained BP neural network model are and the initial threshold Adjust and retest until the calculated accuracy is greater than or equal to the accuracy threshold, and use the current BP neural network model as the final BP neural network model; S12, collecting image data of the produced automobile parts through the image data collection module, inputting the image data of the produced automobile parts into the final BP neural network model for image recognition, and outputting the recognition results, and classifying the produced automobile parts according to the categories of automobile parts according to the recognition results.
[0009] The BP neural network algorithm was used to train the model of image data of all categories of auto parts in the warehouse, and the final BP neural network model was obtained, which provided a tool for further image recognition and improved the accuracy of classification according to the category of auto parts. Preferably, the corresponding quality inspection device is selected according to the category of automobile parts to perform quality inspection on automobile parts, and the specific operation steps of obtaining automobile parts that do not meet the quality requirements and automobile parts that meet the quality requirements are as follows: S21. Select corresponding quality inspection devices according to the categories of automobile parts, perform quality inspection on the automobile parts to be inspected according to the set quality standards, obtain automobile parts that do not meet the quality requirements and automobile parts that meet the quality requirements, and record the quality inspection data of automobile parts.
[0010] The quality inspection device is used to determine whether the automotive parts to be inspected are qualified and record the quality inspection data, providing material and data basis for subsequent operations.
[0011] Preferably, the specific operation steps of analyzing the data of the automobile parts that do not meet the quality requirements, selecting the automobile parts that meet the rework conditions, and reprocessing the automobile parts are as follows: S31. Analyze the data of automobile parts whose quality does not meet the requirements, match the processing methods of the corresponding automobile parts, select the automobile parts that meet the rework conditions, and reprocess the automobile parts.
[0012] Select the auto parts that meet the rework conditions among the auto parts that do not meet the quality requirements and reprocess them to avoid unnecessary waste of auto parts.
[0013] Preferably, the specific operation steps of performing quality inspection on the reprocessed automobile parts to obtain automobile parts that do not meet the quality requirements and automobile parts that meet the quality requirements are as follows: S41. Select the corresponding quality inspection device according to the category of automobile parts, re-inspect the quality of the reprocessed automobile parts according to the set quality standards, select the automobile parts with qualified quality, and directly include the automobile parts that fail the quality inspection into the automobile parts that do not meet the quality requirements, and record the data of the automobile parts quality inspection.
[0014] The reprocessed auto parts are re-inspected for quality and the quality inspection data is recorded to ensure the accuracy of quality inspection and provide a data basis for subsequent operations.
[0015] Preferably, all automobile parts data in each category are clustered respectively, and the quality level of automobile parts in each category is obtained according to the clustering result, and all automobile parts that have undergone quality inspection are matched with the corresponding automobile parts quality level. According to the quality level of automobile parts, all automobile parts that do not meet the quality requirements and all automobile parts that meet the quality requirements are transported into the warehouse respectively, and the specific operation steps for realizing further classification management are as follows: S51. Obtain all automobile parts data in each category, arbitrarily select all automobile parts data in one category as data samples, and cluster the data samples using the k-means clustering algorithm, wherein the k value is set to 6, so that the cluster categories are 6, that is, the quality grades of automobile parts are 1-6, the lower the grade, the better the quality, and automobile parts with quality grades of 1-3 are set as automobile parts with qualified quality, and automobile parts with quality grades of 4-6 are set as automobile parts with unsatisfactory quality. The specific steps of clustering are as follows: Initialize k different center points, assign each data point to the center point closest to it according to the principle of closest distance, obtain k clusters, recalculate the center points of k clusters, and repeat the above operation until the center points of the calculated k clusters are the same as the current k center points or the number of iterations reaches the maximum number of iterations , k center points correspond to a quality grade of automobile parts respectively; According to the above steps, the quality grade of each category of automotive parts is calculated; the calculation formula of Euclidean distance is as follows:
[0016] in, express and The Euclidean distance between is the n-dimensional feature vector of two data points, For vector In the The value of the dimension, For vector In the The value of the dimension; S52, taking the data of all automobile parts that have undergone quality inspection as data samples, calculating the Euclidean distance between the data samples and all center points, and dividing the data samples into clusters corresponding to the center points closest to them, that is, the quality grades of all automobile parts that have undergone quality inspection are the quality grades of automobile parts corresponding to the center points; S53. According to the quality grade of auto parts, all auto parts that do not meet the quality requirements and all auto parts that meet the quality requirements are transported to the warehouse separately to achieve further classification management.
[0017] The k-means clustering algorithm is used to cluster all automobile parts data in each category, and the quality grade of automobile parts in each category is calculated. The Euclidean distance between the data of automobile parts to be stored and all center points is calculated to match the corresponding quality grade of automobile parts. According to the quality grade of automobile parts, automobile parts are transported into the warehouse for further classification management, ensuring the accuracy of automobile parts production quality control process and improving the accuracy of automobile parts production quality control results.
[0018] An automobile parts production quality control system based on classification management, including an automobile parts image data acquisition module, an automobile parts quality inspection module, an automobile parts processing module that does not meet quality requirements, an automobile parts data clustering module and an automobile parts warehousing module; The automobile parts image data acquisition module collects image data of produced automobile parts through the image data acquisition module, inputs the image data of produced automobile parts into the trained BP neural network model for image recognition, and classifies the produced automobile parts according to the categories of automobile parts according to the recognition results; The automobile parts quality inspection module selects corresponding quality inspection devices according to the categories of automobile parts, and performs quality inspection on the automobile parts to be inspected according to the corresponding quality standards; The automobile parts processing module that does not meet the quality requirements analyzes the data of the automobile parts that do not meet the quality requirements, matches the processing methods of the corresponding automobile parts, selects automobile parts that meet the rework conditions for reprocessing, and re-inspects the quality of the reprocessed automobile parts; The automobile parts data clustering module clusters all automobile parts data in each category respectively, and matches all automobile parts that have undergone quality inspection with corresponding automobile parts quality grades according to the clustering results; The automobile parts warehousing module transports all automobile parts that do not meet the quality requirements and all automobile parts that meet the quality requirements into the warehouse respectively according to the quality grades of the automobile parts, so as to realize further classification management.
[0019] (III) Beneficial effects The present invention provides a method and system for controlling the production quality of automobile parts based on classification management, which has the following beneficial effects: 1. The automobile parts image data acquisition module collects the image data of the produced automobile parts through the image data acquisition module, inputs the image data of the produced automobile parts into the trained BP neural network model for image recognition, and provides a data basis for subsequent operations; the automobile parts quality inspection module determines whether the automobile parts to be inspected are qualified through the quality inspection device and records the quality inspection data, providing materials and data basis for subsequent operations; 2. The automobile parts processing module that do not meet the quality requirements selects automobile parts that meet the rework conditions from the automobile parts that do not meet the quality requirements for reprocessing, thereby avoiding unnecessary waste of automobile parts, and re-tests the reprocessed automobile parts and records the quality test data, thereby ensuring the accuracy of the quality test and providing a data basis for subsequent operations; 3. The automobile parts data clustering module adopts the k-means clustering algorithm to cluster all automobile parts data in each category respectively, calculates the quality grade of automobile parts in each category, and matches the quality grade of corresponding automobile parts by calculating the Euclidean distance between the data of automobile parts to be stored and all center points. The k-means clustering algorithm reduces the complexity of the calculation process and ensures the accuracy of the automobile parts production quality control process; the automobile parts warehousing module transports automobile parts into the warehouse according to their quality grades, realizes further classification management, and improves the accuracy of automobile parts production quality control results. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 A flow chart of a method for controlling the production quality of automobile parts based on classification management provided by the present invention; Figure 2 A module schematic diagram of an automobile parts production quality control system based on classification management provided by the present invention; DETAILED DESCRIPTION
[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0022] The embodiment of the automobile parts production quality control method and system based on classification management is as follows: See also Figure 1 — Figure 2, a method for controlling the production quality of automobile parts based on classification management, the method comprising the following steps: S1. Collecting image data of produced automobile parts through an image data acquisition module, inputting the image data of the produced automobile parts into a trained BP neural network model for image recognition, and classifying the produced automobile parts according to the categories of automobile parts according to the recognition results; S11, collecting image data of all categories of automobile parts in the warehouse through the image data collection module, and performing model training using the BP neural network algorithm; S111, set the training data ratio to , the test data ratio is , according to the training data ratio And test data ratio Performing data division on the image data of all categories of automobile parts in the warehouse to obtain training image data of all categories of automobile parts in the warehouse and test image data of all categories of automobile parts in the warehouse; S112, constructing an initial BP neural network model, setting the number of input nodes of the initial BP neural network model to The number of hidden nodes is The number of output nodes is The initial weight is And the initial threshold is ; S113, setting a training error threshold, inputting the training image data of all categories of automobile parts in the warehouse into the initial BP neural network model to perform a training operation on the initial BP neural network model, and when the training error during the training process is less than or equal to the training error threshold, stopping the training to obtain the trained BP neural network model; S114, setting an accuracy threshold, inputting the test image data of all categories of automobile parts in the warehouse into the initial BP neural network model for testing, and obtaining the test accuracy after the test is completed. When the test accuracy is greater than or equal to the accuracy threshold, the trained BP neural network model is used as the final BP neural network model; otherwise, the initial weights of the trained BP neural network model are and the initial threshold Adjust and retest until the calculated accuracy is greater than or equal to the accuracy threshold, and use the current BP neural network model as the final BP neural network model; S12, collecting image data of the produced automobile parts through the image data collection module, inputting the image data of the produced automobile parts into the final BP neural network model for image recognition, and outputting the recognition results, and classifying the produced automobile parts according to the categories of automobile parts according to the recognition results.
[0023] S2. Selecting corresponding quality inspection devices according to the categories of automobile parts to perform quality inspection on automobile parts, and obtaining automobile parts that do not meet quality requirements and automobile parts that meet quality requirements; S21. Select corresponding quality inspection devices according to the categories of automobile parts, perform quality inspection on the automobile parts to be inspected according to the set quality standards, obtain automobile parts that do not meet the quality requirements and automobile parts that meet the quality requirements, and record the quality inspection data of automobile parts.
[0024] S3. Analyze the data of the automobile parts that do not meet the quality requirements, select the automobile parts that meet the rework conditions, and reprocess the automobile parts; S31. Analyze the data of automobile parts whose quality does not meet the requirements, match the processing methods of the corresponding automobile parts, select the automobile parts that meet the rework conditions, and reprocess the automobile parts.
[0025] S4. Conduct quality inspection on the reprocessed auto parts to obtain auto parts that do not meet quality requirements and auto parts that meet quality requirements; S41. Select the corresponding quality inspection device according to the category of automobile parts, re-inspect the quality of the reprocessed automobile parts according to the set quality standards, select the automobile parts with qualified quality, and directly include the automobile parts that fail the quality inspection into the automobile parts that do not meet the quality requirements, and record the data of the automobile parts quality inspection.
[0026] S5. Cluster all the auto parts data in each category respectively, obtain the quality grade of each category of auto parts according to the clustering results, match all the auto parts that have passed the quality inspection with the corresponding auto parts quality grade, and transport all the auto parts that do not meet the quality requirements and all the auto parts that meet the quality requirements into the warehouse respectively according to the quality grade of the auto parts, so as to realize the classification management; S51. Obtain all automobile parts data in each category, arbitrarily select all automobile parts data in one category as data samples, and cluster the data samples using the k-means clustering algorithm, wherein the k value is set to 6, so that the cluster categories are 6, that is, the quality grades of automobile parts are 1-6, the lower the grade, the better the quality, and automobile parts with quality grades of 1-3 are set as automobile parts with qualified quality, and automobile parts with quality grades of 4-6 are set as automobile parts with unsatisfactory quality. The specific steps of clustering are as follows: Initialize k different center points, assign each data point to the center point closest to it according to the principle of closest distance, obtain k clusters, recalculate the center points of k clusters, and repeat the above operation until the center points of the calculated k clusters are the same as the current k center points or the number of iterations reaches the maximum number of iterations , k center points correspond to a quality grade of automobile parts respectively; According to the above steps, the quality grade of each category of automotive parts is calculated; the calculation formula of Euclidean distance is as follows:
[0027] in, express and The Euclidean distance between is the n-dimensional feature vector of two data points, For vector In the The value of the dimension, For vector In the The value of the dimension; S52, taking the data of all automobile parts that have undergone quality inspection as data samples, calculating the Euclidean distance between the data samples and all center points, and dividing the data samples into clusters corresponding to the center points closest to them, that is, the quality grades of all automobile parts that have undergone quality inspection are the quality grades of automobile parts corresponding to the center points; S53. According to the quality grade of automobile parts, all automobile parts that do not meet the quality requirements and all automobile parts that meet the quality requirements are transported to the warehouse separately to realize classified management.
[0028] The k-means clustering algorithm is used to cluster all automobile parts data in each category, and the quality grade of automobile parts in each category is calculated. The Euclidean distance between the data of automobile parts to be stored and all center points is calculated to match the corresponding quality grade of automobile parts. According to the quality grade of automobile parts, automobile parts are transported into the warehouse for classified management, ensuring the accuracy of automobile parts production quality control and improving the accuracy of automobile parts production quality control.
[0029] An automobile parts production quality control system based on classification management, including an automobile parts image data acquisition module, an automobile parts quality inspection module, an automobile parts processing module that does not meet quality requirements, an automobile parts data clustering module, and an automobile parts warehousing module; The automobile parts image data acquisition module collects image data of produced automobile parts through the image data acquisition module, inputs the image data of produced automobile parts into the trained BP neural network model for image recognition, and classifies the produced automobile parts according to the categories of automobile parts according to the recognition results; The automobile parts quality inspection module selects corresponding quality inspection devices according to the categories of automobile parts, and performs quality inspection on the automobile parts to be inspected according to the corresponding quality standards; The automobile parts processing module that does not meet the quality requirements analyzes the data of the automobile parts that do not meet the quality requirements, matches the processing methods of the corresponding automobile parts, selects automobile parts that meet the rework conditions for reprocessing, and re-inspects the quality of the reprocessed automobile parts; The automobile parts data clustering module clusters all automobile parts data in each category respectively, and matches all automobile parts that have passed quality inspection to corresponding automobile parts quality grades according to the clustering results; The automobile parts warehousing module transports all automobile parts that do not meet the quality requirements and all automobile parts that meet the quality requirements into the warehouse respectively according to the quality grades of the automobile parts, so as to realize classified management.
[0030] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for controlling the production quality of automobile parts based on classification management, characterized in that: The following steps are involved: S1. Collecting image data of produced automobile parts through an image data acquisition module, inputting the image data of the produced automobile parts into a trained BP neural network model for image recognition, and classifying the produced automobile parts according to the categories of automobile parts according to the recognition results; S2. Selecting corresponding quality inspection devices according to the categories of automobile parts to perform quality inspection on automobile parts, and obtaining automobile parts that do not meet quality requirements and automobile parts that meet quality requirements; S3. Analyze the data of the automobile parts that do not meet the quality requirements, select the automobile parts that meet the rework conditions, and reprocess the automobile parts; S4. Conduct quality inspection on the reprocessed auto parts to obtain auto parts that do not meet quality requirements and auto parts that meet quality requirements; S5. Cluster all automobile parts data in each category separately, and obtain the quality grade of automobile parts in each category based on the clustering results. Match all automobile parts that have passed quality inspection with corresponding automobile parts quality grades. According to the quality grades of automobile parts, all automobile parts that do not meet the quality requirements and all automobile parts that meet the quality requirements are transported into the warehouse separately.
2. The method for controlling the production quality of automobile parts based on classification management according to claim 1 is characterized in that: The S1 comprises the following steps: S11, collecting image data of all categories of automobile parts in the warehouse through the image data collection module, and performing model training using the BP neural network algorithm; S12, collecting image data of the produced automobile parts through the image data collection module, inputting the image data of the produced automobile parts into the final BP neural network model for image recognition, and outputting the recognition results, and classifying the produced automobile parts according to the categories of automobile parts according to the recognition results.
3. The method for controlling the production quality of automobile parts based on classification management according to claim 2 is characterized in that: The model training includes the following steps: S111, set the training data ratio to , the test data ratio is , according to the training data ratio And test data ratio Performing data division on the image data of all categories of automobile parts in the warehouse to obtain training image data of all categories of automobile parts in the warehouse and test image data of all categories of automobile parts in the warehouse; S112, constructing an initial BP neural network model, setting the number of input nodes of the initial BP neural network model to The number of hidden nodes is The number of output nodes is The initial weight is And the initial threshold is ; S113, setting a training error threshold, inputting the training image data of all categories of automobile parts in the warehouse into the initial BP neural network model to perform a training operation on the initial BP neural network model, and when the training error during the training process is less than or equal to the training error threshold, stopping the training to obtain the trained BP neural network model; S114, setting an accuracy threshold, inputting the test image data of all categories of automobile parts in the warehouse into the initial BP neural network model for testing, and obtaining the test accuracy after the test is completed. When the test accuracy is greater than or equal to the accuracy threshold, the trained BP neural network model is used as the final BP neural network model; otherwise, the initial weights of the trained BP neural network model are and the initial threshold Adjustments are made and the test is repeated until the calculated accuracy is greater than or equal to the accuracy threshold, and the current BP neural network model is used as the final BP neural network model.
4. The method for controlling the production quality of automobile parts based on classification management according to claim 1 is characterized in that: The S2 comprises the following steps: S21. Select corresponding quality inspection devices according to the categories of automobile parts, perform quality inspection on the automobile parts to be inspected according to the set quality standards, obtain automobile parts that do not meet the quality requirements and automobile parts that meet the quality requirements, and record the quality inspection data of automobile parts.
5. The method for controlling the production quality of automobile parts based on classification management according to claim 1 is characterized in that: The S3 comprises the following steps: S31. Analyze the data of automobile parts whose quality does not meet the requirements, match the processing methods of the corresponding automobile parts, select the automobile parts that meet the rework conditions, and reprocess the automobile parts.
6. The method for controlling the production quality of automobile parts based on classification management according to claim 1 is characterized in that: The S4 comprises the following steps: S41. Select the corresponding quality inspection device according to the category of automobile parts, re-inspect the quality of the reprocessed automobile parts according to the set quality standards, select the automobile parts with qualified quality, and directly include the automobile parts that fail the quality inspection into the automobile parts that do not meet the quality requirements, and record the data of the automobile parts quality inspection.
7. The method for controlling the production quality of automobile parts based on classification management according to claim 1 is characterized in that: The S5 comprises the following steps: S51, obtaining all automobile parts data in each category, arbitrarily selecting all automobile parts data in one category as data samples, and clustering the data samples using a k-means clustering algorithm, wherein the k value is set to 6, so that the number of cluster categories is 6, that is, the quality grades of automobile parts are 1-6, the lower the grade, the better the quality, and automobile parts with quality grades of 1-3 are set as automobile parts with qualified quality, and automobile parts with quality grades of 4-6 are set as automobile parts that do not meet the quality requirements; S52, taking the data of all automobile parts that have undergone quality inspection as data samples, calculating the Euclidean distance between the data samples and all center points, and dividing the data samples into clusters corresponding to the center points closest to them, that is, the quality grades of all automobile parts that have undergone quality inspection are the quality grades of automobile parts corresponding to the center points; S53. According to the quality grade of automobile parts, all automobile parts that do not meet the quality requirements and all automobile parts that meet the quality requirements are transported to the warehouse separately to realize classified management.
8. The method for controlling the production quality of automobile parts based on classification management according to claim 7 is characterized in that: The clustering process includes the following steps: Initialize k different center points, assign each data point to the center point closest to it according to the principle of closest distance, obtain k clusters, recalculate the center points of k clusters, and repeat the above operation until the center points of the calculated k clusters are the same as the current k center points or the number of iterations reaches the maximum number of iterations , k center points correspond to a quality grade of automobile parts respectively; According to the above steps, the quality grade of each category of automotive parts is calculated; the calculation formula of Euclidean distance is as follows: in, express and The Euclidean distance between is the n-dimensional feature vector of two data points, For vector In the The value of the dimension, For vector In the The value of the dimension.
9. A system for implementing the method for controlling the production quality of automobile parts based on classification management as described in any one of claims 1 to 8.