Automobile safety part welding quality management system

Abstract

The invention discloses an automobile safety part welding quality management system, and relates to the field of quality detection. Comprising a data acquisition layer for acquiring multi-dimensional data of a weldment by using weldment quality detection equipment; the data processing layer is used for processing the collected data, a defect evaluation model is arranged in the data processing layer, processed output is input into the defect evaluation model, and a weldment quality evaluation result is output; and the defect classification module is used for classifying the weldments with defects according to the output quality evaluation result. By establishing the standardized welding data management system, on one hand, welding quality evaluation can be carried out after welding multi-dimensional data are collected and fused, the welding quality evaluation precision is improved, on the other hand, welding defective weldments are classified, analysis and welding process optimization are facilitated, and the welding quality is improved. And meanwhile, distributed storage is adopted for data storage, and dynamic space optimization can be performed on each storage unit, so that data over-limit loss is prevented, and the data read-write operation fluency of the system is also improved.

Description

Technical Field

[0001] The present invention relates to the field of quality inspection, and in particular to a welding quality management system for automobile safety parts. Background Art

[0002] In modern industrial manufacturing, welding, as a crucial joining process, is widely used in a wide range of industries, including aerospace, automotive, petrochemical, and shipbuilding. The quality of welds is directly related to the safety, reliability, and service life of the entire product, making accurate weld quality testing crucial.

[0003] Traditional weld quality inspection relies primarily on manual visual inspection and simple random destructive testing. Limited by factors such as operator experience, fatigue, and ambient lighting, manual visual inspection can be difficult to detect even minor defects within welds and can leave blind spots for welds with complex structures. While destructive testing can obtain accurate data, it can cause irreversible damage to welds, increasing production costs, and its results cannot directly reflect the overall quality of welds in actual use.

[0004] To this end, the present invention proposes an automobile safety parts welding quality management system. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a welding quality management system for automobile safety parts.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] An automobile safety parts welding quality management system, comprising:

[0008] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0009] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0010] A defect classification module classifies weldments with defects based on the output quality evaluation results;

[0011] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0012] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0013] Preferably, the data acquisition layer includes:

[0014] Ultrasonic flaw detectors use the propagation characteristics of ultrasonic waves in materials to detect defects and identify internal cracks and pores through reflected waves;

[0015] X-ray flaw detectors, which penetrate welds with X-rays and display internal defects using film or digital imaging systems;

[0016] Magnetic particle detectors, which apply a magnetic field to the surface of ferromagnetic materials and reveal surface or near-surface cracks through the accumulation of magnetic particles;

[0017] Penetrant testing equipment uses penetrant fluid to penetrate surface open defects and displays the defect outline through a developer;

[0018] Tensile testing machine, which applies tensile force to weldment specimens to measure their yield strength and tensile strength;

[0019] Impact testing machine, which tests the toughness of weldments through impact loads and evaluates their performance under low temperatures or dynamic loads;

[0020] Hardness tester, which measures the hardness value by pressing the indenter into the weldment surface;

[0021] Metallographic microscopy, which observes the microstructure of weld metal using an optical or electron microscope;

[0022] Scanning electron microscopes, which use an electron beam to scan the surface of a sample, producing high-resolution images;

[0023] Any combination of .

[0024] Preferably, the method for evaluating weldment quality by the data processing layer comprises the following steps:

[0025] A1: Data normalization: normalize the collected data so that the collected multi-dimensional data have the same dimension;

[0026] A2: Threshold setting: set the qualified data interval [a, b] for each collected data dimension, and simultaneously set the error proportional coefficient k;

[0027] A3: Compare each dimension with the set data interval and output the evaluation results for that dimension. Then, use the random forest voting algorithm to vote and select the result with the most votes to output.

[0028] Preferably: in step A2, k∈(0, 1).

[0029] Preferably: in the step A2, .

[0030] Preferably, the step A3 comprises the following steps:

[0031] A31: If the collected and processed data falls into If the dimension is within the interval, it is determined to be qualified and the output result is "1", otherwise there is a defect and the output result is "0";

[0032] A32: Repeat step A31 until multiple dimensions are compared, count all the results, and output the result with the highest number of occurrences as the final result.

[0033] Preferably, the classification method of the defect classification module includes the following steps:

[0034] B1: Establish multiple defect type sets, each defect type set corresponds to defects in one or more dimensions of the data collection dimension;

[0035] B2: Retrieve the dimension data whose output result is "0" from the data processing layer;

[0036] B3: Match all dimensions with output results of "0" with the sets in the defect type set one by one, and finally classify the defective weldments into the defect type set with the largest number of matching dimensions.

[0037] Preferably, the data storage module adopts distributed storage, and its storage method includes the following steps:

[0038] C1: Divide the entire storage module into multiple storage units according to data types;

[0039] C2: All data are classified and data of the same category are stored in one storage unit;

[0040] C3: Dynamic storage space optimization, dynamically adjusting the storage space size of each storage unit based on the growth rate of data space occupied by different storage units.

[0041] Preferably, the C3 step includes the following steps:

[0042] C31: When initially allocated, all storage units have the same storage space size. , where Q is the total storage space size, p is the number of storage units, is the storage space size of the i-th storage unit;

[0043] C32: Set the update time interval t and obtain the data growth rate f of each storage unit within the time interval t;

[0044] C33: Then follow the formula Dynamically optimize the storage space of each storage unit.

[0045] Preferably, the identity authentication module includes any one or more combinations of account-password verification, voiceprint verification, iris verification, and fingerprint verification.

[0046] The beneficial effects of the present invention are:

[0047] 1. The present invention establishes a standardized welding data management system, which can, on the one hand, collect and integrate multi-dimensional welding data to evaluate welding quality, thereby increasing the accuracy of welding quality evaluation; on the other hand, it can classify welded parts with welding defects, facilitating analysis and welding process optimization. At the same time, distributed storage is used for data storage, and dynamic space optimization can be performed on each storage unit to prevent data from being lost due to excessive limits, while also increasing the fluency of system data reading and writing operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a schematic diagram of the architecture of an automotive safety parts welding quality management system proposed by the present invention. DETAILED DESCRIPTION

[0049] The technical solution of the present invention will be further described in detail below in conjunction with specific implementation methods.

[0050] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "installed," "connected," "connected," and "disposed" should be understood in a broad sense. For example, they may refer to fixed connection or disposition, detachable connection or disposition, or integral connection or disposition. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0051] Example 1:

[0052] An automobile safety parts welding quality management system, comprising:

[0053] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0054] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0055] A defect classification module, which classifies weldments with defects based on the output quality evaluation results;

[0056] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0057] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0058] The data acquisition layer includes:

[0059] Ultrasonic flaw detectors use the propagation characteristics of ultrasonic waves in materials to detect defects and identify internal cracks and pores through reflected waves;

[0060] X-ray flaw detectors, which penetrate welds with X-rays and display internal defects using film or digital imaging systems;

[0061] Magnetic particle detectors, which apply a magnetic field to the surface of ferromagnetic materials and reveal surface or near-surface cracks through the accumulation of magnetic particles;

[0062] Penetrant testing equipment uses penetrant liquid to penetrate open surface defects and displays the defect outline through a developer.

[0063] Example 2:

[0064] An automobile safety parts welding quality management system, comprising:

[0065] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0066] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0067] A defect classification module, which classifies weldments with defects based on the output quality evaluation results;

[0068] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0069] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0070] The data acquisition layer includes:

[0071] Tensile testing machine, which applies tensile force to weldment specimens to measure their yield strength and tensile strength;

[0072] Impact testing machine, which tests the toughness of weldments through impact loads and evaluates their performance under low temperatures or dynamic loads;

[0073] The hardness tester measures the hardness value by pressing the indenter into the surface of the weldment.

[0074] Example 3:

[0075] An automobile safety parts welding quality management system, comprising:

[0076] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0077] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0078] A defect classification module classifies weldments with defects based on the output quality evaluation results;

[0079] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0080] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0081] The data acquisition layer includes:

[0082] Metallographic microscopy, which observes the microstructure of weld metal using an optical or electron microscope;

[0083] Scanning electron microscopes use a beam of electrons to scan the surface of a sample, producing high-resolution images.

[0084] Example 4:

[0085] An automobile safety parts welding quality management system, comprising:

[0086] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0087] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0088] A defect classification module classifies weldments with defects based on the output quality evaluation results;

[0089] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0090] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0091] The data acquisition layer includes:

[0092] Ultrasonic flaw detectors use the propagation characteristics of ultrasonic waves in materials to detect defects and identify internal cracks and pores through reflected waves;

[0093] X-ray flaw detectors, which penetrate welds with X-rays and display internal defects using film or digital imaging systems;

[0094] Magnetic particle detectors, which apply a magnetic field to the surface of ferromagnetic materials and reveal surface or near-surface cracks through the accumulation of magnetic particles;

[0095] Penetrant testing equipment uses penetrant fluid to penetrate surface open defects and displays the defect outline through a developer;

[0096] Tensile testing machine, which applies tensile force to weldment specimens to measure their yield strength and tensile strength;

[0097] Impact testing machine, which tests the toughness of weldments through impact loads and evaluates their performance under low temperatures or dynamic loads;

[0098] Hardness tester, which measures the hardness value by pressing the indenter into the weldment surface;

[0099] Metallographic microscopy, which observes the microstructure of weld metal using an optical or electron microscope;

[0100] Scanning electron microscopes, which use an electron beam to scan the surface of a sample, producing high-resolution images;

[0101] Any combination of .

[0102] The method for evaluating weldment quality by the data processing layer comprises the following steps:

[0103] A1: Data normalization: normalize the collected data so that the collected multi-dimensional data have the same dimension;

[0104] A2: Threshold setting: set the qualified data interval [a, b] for each collected data dimension, and simultaneously set the error proportional coefficient k;

[0105] A3: Compare each dimension with the set data interval and output the evaluation results for that dimension. Then, use the random forest voting algorithm to vote and select the result with the most votes to output.

[0106] In step A2, k∈(0, 1).

[0107] Example 5:

[0108] An automobile safety parts welding quality management system, comprising:

[0109] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0110] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0111] A defect classification module classifies weldments with defects based on the output quality evaluation results;

[0112] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0113] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0114] The data acquisition layer includes:

[0115] Ultrasonic flaw detectors use the propagation characteristics of ultrasonic waves in materials to detect defects and identify internal cracks and pores through reflected waves;

[0116] X-ray flaw detectors, which penetrate welds with X-rays and display internal defects using film or digital imaging systems;

[0117] Magnetic particle detectors, which apply a magnetic field to the surface of ferromagnetic materials and reveal surface or near-surface cracks through the accumulation of magnetic particles;

[0118] Penetrant testing equipment uses penetrant fluid to penetrate surface open defects and displays the defect outline through a developer;

[0119] Tensile testing machine, which applies tensile force to weldment specimens to measure their yield strength and tensile strength;

[0120] Impact testing machine, which tests the toughness of weldments through impact loads and evaluates their performance under low temperatures or dynamic loads;

[0121] Hardness tester, which measures the hardness value by pressing the indenter into the weldment surface;

[0122] Metallographic microscopy, which observes the microstructure of weld metal using an optical or electron microscope;

[0123] Scanning electron microscopes, which use an electron beam to scan the surface of a sample, producing high-resolution images;

[0124] Any combination of .

[0125] The method for evaluating weldment quality by the data processing layer comprises the following steps:

[0126] A1: Data normalization: normalize the collected data so that the collected multi-dimensional data have the same dimension;

[0127] A2: Threshold setting: set the qualified data interval [a, b] for each collected data dimension, and simultaneously set the error proportional coefficient k;

[0128] A3: Compare each dimension with the set data interval and output the evaluation results for that dimension. Then, use the random forest voting algorithm to vote and select the result with the most votes to output.

[0129] In the step A2, .

[0130] Example 6:

[0131] An automobile safety parts welding quality management system, comprising:

[0132] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0133] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0134] A defect classification module, which classifies weldments with defects based on the output quality evaluation results;

[0135] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0136] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0137] The data acquisition layer includes:

[0138] Ultrasonic flaw detectors use the propagation characteristics of ultrasonic waves in materials to detect defects and identify internal cracks and pores through reflected waves;

[0139] X-ray flaw detectors, which penetrate welds with X-rays and display internal defects using film or digital imaging systems;

[0140] Magnetic particle detectors, which apply a magnetic field to the surface of ferromagnetic materials and reveal surface or near-surface cracks through the accumulation of magnetic particles;

[0141] Penetrant testing equipment uses penetrant fluid to penetrate surface open defects and displays the defect outline through a developer;

[0142] Tensile testing machine, which applies tensile force to weldment specimens to measure their yield strength and tensile strength;

[0143] Impact testing machine, which tests the toughness of weldments through impact loads and evaluates their performance under low temperatures or dynamic loads;

[0144] Hardness tester, which measures the hardness value by pressing the indenter into the weldment surface;

[0145] Metallographic microscopy, which observes the microstructure of weld metal using an optical or electron microscope;

[0146] Scanning electron microscopes, which use an electron beam to scan the surface of a sample, producing high-resolution images;

[0147] Any combination of .

[0148] The method for evaluating weldment quality by the data processing layer comprises the following steps:

[0149] A1: Data normalization: normalize the collected data so that the collected multi-dimensional data have the same dimension;

[0150] A2: Threshold setting: set the qualified data interval [a, b] for each collected data dimension, and simultaneously set the error proportional coefficient k;

[0151] A3: Compare each dimension with the set data interval and output the evaluation results for that dimension. Then, use the random forest voting algorithm to vote and select the result with the most votes to output.

[0152] In the step A2, .

[0153] The A3 step comprises the following steps:

[0154] A31: If the collected and processed data falls into If the dimension is within the interval, it is determined to be qualified and the output result is "1", otherwise there is a defect and the output result is "0";

[0155] A32: Repeat step A31 until multiple dimensions are compared, count all the results, and output the result with the highest number of occurrences as the final result.

[0156] The classification method of the defect classification module includes the following steps:

[0157] B1: Establish multiple defect type sets, each defect type set corresponds to defects in one or more dimensions of the data collection dimension;

[0158] B2: Retrieve the dimension data whose output result is "0" from the data processing layer;

[0159] B3: Match all dimensions with output results of "0" with the sets in the defect type set one by one, and finally classify the defective weldments into the defect type set with the largest number of matching dimensions.

[0160] The data storage module adopts distributed storage, and its storage method includes the following steps:

[0161] C1: Divide the entire storage module into multiple storage units according to data types;

[0162] C2: All data are classified and data of the same category are stored in one storage unit;

[0163] C3: Dynamic storage space optimization, dynamically adjusting the storage space size of each storage unit based on the growth rate of data space occupied by different storage units.

[0164] The C3 step includes the following steps:

[0165] C31: When initially allocated, all storage units have the same storage space size. , where Q is the total storage space size, p is the number of storage units, is the storage space size of the i-th storage unit;

[0166] C32: Set the update time interval t and obtain the data growth rate f of each storage unit within the time interval t;

[0167] C33: Then follow the formula Dynamically optimize the storage space of each storage unit.

[0168] Example 7:

[0169] An automobile safety parts welding quality management system, comprising:

[0170] The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments;

[0171] The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results.

[0172] A defect classification module classifies weldments with defects based on the output quality evaluation results;

[0173] A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data;

[0174] Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

[0175] The data acquisition layer includes:

[0176] Ultrasonic flaw detectors use the propagation characteristics of ultrasonic waves in materials to detect defects and identify internal cracks and pores through reflected waves;

[0177] X-ray flaw detectors, which penetrate welds with X-rays and display internal defects using film or digital imaging systems;

[0178] Magnetic particle detectors, which apply a magnetic field to the surface of ferromagnetic materials and reveal surface or near-surface cracks through the accumulation of magnetic particles;

[0179] Penetrant testing equipment uses penetrant fluid to penetrate surface open defects and displays the defect outline through a developer;

[0180] Tensile testing machine, which applies tensile force to weldment specimens to measure their yield strength and tensile strength;

[0181] Impact testing machine, which tests the toughness of weldments through impact loads and evaluates their performance under low temperatures or dynamic loads;

[0182] Hardness tester, which measures the hardness value by pressing the indenter into the weldment surface;

[0183] Metallographic microscopy, which observes the microstructure of weld metal using an optical or electron microscope;

[0184] Scanning electron microscopes, which use an electron beam to scan the surface of a sample, producing high-resolution images;

[0185] Any combination of .

[0186] The method for evaluating weldment quality by the data processing layer comprises the following steps:

[0187] A1: Data normalization: normalize the collected data so that the collected multi-dimensional data have the same dimension;

[0188] A2: Threshold setting: set the qualified data interval [a, b] for each collected data dimension, and simultaneously set the error proportional coefficient k;

[0189] A3: Compare each dimension with the set data interval and output the evaluation results for that dimension. Then, use the random forest voting algorithm to vote and select the result with the most votes to output.

[0190] In step A2, k∈(0, 1).

[0191] In the step A2, .

[0192] The A3 step comprises the following steps:

[0193] A31: If the collected and processed data falls into If the dimension is within the interval, it is determined to be qualified and the output result is "1", otherwise there is a defect and the output result is "0";

[0194] A32: Repeat step A31 until multiple dimensions are compared, count all the results, and output the result with the highest number of occurrences as the final result.

[0195] The classification method of the defect classification module includes the following steps:

[0196] B1: Establish multiple defect type sets, each defect type set corresponds to defects in one or more dimensions of the data collection dimension;

[0197] B2: Retrieve the dimension data whose output result is "0" from the data processing layer;

[0198] B3: Match all dimensions with output results of "0" with the sets in the defect type set one by one, and finally classify the defective weldments into the defect type set with the largest number of matching dimensions.

[0199] The data storage module adopts distributed storage, and its storage method includes the following steps:

[0200] C1: Divide the entire storage module into multiple storage units according to data types;

[0201] C2: All data are classified and data of the same category are stored in one storage unit;

[0202] C3: Dynamic storage space optimization, dynamically adjusting the storage space size of each storage unit based on the growth rate of data space occupied by different storage units.

[0203] The C3 step includes the following steps:

[0204] C31: When initially allocated, all storage units have the same storage space size. , where Q is the total storage space size, p is the number of storage units, is the storage space size of the i-th storage unit;

[0205] C32: Set the update time interval t and obtain the data growth rate f of each storage unit within the time interval t;

[0206] C33: Then follow the formula Dynamically optimize the storage space of each storage unit.

[0207] The identity verification module includes any one or more combinations of account-password verification, voiceprint verification, iris verification, and fingerprint verification.

[0208] This device, by establishing a standardized welding data management system, can, on the one hand, collect and integrate multi-dimensional welding data to evaluate welding quality, thereby increasing the accuracy of welding quality evaluation; on the other hand, it can classify welded parts with welding defects, facilitating analysis and welding process optimization. At the same time, distributed storage is used for data storage, and dynamic space optimization can be performed on each storage unit to prevent data from being lost due to excessive limits, while also increasing the fluency of system data reading and writing operations.

[0209] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An automobile safety parts welding quality management system, characterized in that: include: The data acquisition layer uses weldment quality inspection equipment to collect multi-dimensional data of weldments; The data processing layer processes the collected data and has a built-in defect evaluation model. The processed output is input into the defect evaluation model to output the weldment quality evaluation results. A defect classification module classifies weldments with defects based on the output quality evaluation results; A data storage module stores the collected raw data, weldment quality evaluation results, defective weldment classification results, and other system data; Identity verification module, users log into the system through the identity verification module and view and operate data in the system.

2. The automobile safety parts welding quality management system according to claim 1, characterized in that: The data acquisition layer includes: Ultrasonic flaw detectors use the propagation characteristics of ultrasonic waves in materials to detect defects and identify internal cracks and pores through reflected waves; X-ray flaw detectors, which penetrate welds with X-rays and display internal defects using film or digital imaging systems; Magnetic particle detectors, which apply a magnetic field to the surface of ferromagnetic materials and reveal surface or near-surface cracks through the accumulation of magnetic particles; Penetrant testing equipment uses penetrant fluid to penetrate surface open defects and displays the defect outline through a developer; Tensile testing machine, which applies tensile force to weldment specimens to measure their yield strength and tensile strength; Impact testing machine, which tests the toughness of weldments through impact loads and evaluates their performance under low temperatures or dynamic loads; Hardness tester, which measures the hardness value by pressing the indenter into the weldment surface; Metallographic microscopy, which observes the microstructure of weld metal using an optical or electron microscope; Scanning electron microscopes, which use an electron beam to scan the surface of a sample, producing high-resolution images; Any combination of .

3. The automobile safety parts welding quality management system according to claim 1, characterized in that: The method for evaluating weldment quality by the data processing layer comprises the following steps: A1: Data normalization: normalize the collected data so that the collected multi-dimensional data have the same dimension; A2: Threshold setting: set the qualified data interval [a, b] for each collected data dimension, and simultaneously set the error proportional coefficient k; A3: Compare each dimension with the set data interval and output the evaluation results for that dimension. Then, use the random forest voting algorithm to vote and select the result with the most votes to output.

4. The automobile safety parts welding quality management system according to claim 3, characterized in that: In step A2, k∈(0, 1).

5. The automobile safety parts welding quality management system according to claim 3, characterized in that: In the step A2, .

6. The automobile safety parts welding quality management system according to claim 3, characterized in that: The A3 step comprises the following steps: A31: If the collected and processed data falls into If the dimension is within the interval, the dimension is considered qualified and the output result is "1", otherwise there is a defect and the output result is "0"; A32: Repeat step A31 until multiple dimensions are compared, count all the results, and output the result with the highest number of occurrences as the final result.

7. The automobile safety parts welding quality management system according to claim 1, characterized in that: The classification method of the defect classification module includes the following steps: B1: Establish multiple defect type sets, each defect type set corresponds to defects in one or more dimensions of the data collection dimension; B2: Retrieve the dimension data whose output result is "0" from the data processing layer; B3: Match all dimensions with output results of "0" with the sets in the defect type set one by one, and finally classify the defective weldments into the defect type set with the largest number of matching dimensions.

8. The automobile safety parts welding quality management system according to claim 1, characterized in that: The data storage module adopts distributed storage, and its storage method includes the following steps: C1: Divide the entire storage module into multiple storage units according to data types; C2: All data are classified and data of the same category are stored in one storage unit; C3: Dynamic storage space optimization, dynamically adjusting the storage space size of each storage unit based on the growth rate of data space occupied by different storage units.

9. The automobile safety parts welding quality management system according to claim 8, characterized in that: The C3 step includes the following steps: C31: When initially allocated, all storage units have the same storage space size. , where Q is the total storage space size, p is the number of storage units, is the storage space size of the i-th storage unit; C32: Set the update time interval t and obtain the data growth rate f of each storage unit within the time interval t; C33: Then follow the formula Dynamically optimize the storage space of each storage unit.

10. The automobile safety parts welding quality management system according to claim 1, characterized in that: The identity verification module includes any one or more combinations of account-password verification, voiceprint verification, iris verification, and fingerprint verification.

Images