Sampling inspection method and system for large-batch spiral spring samples

Abstract

The invention relates to the technical field of sampling inspection, and particularly discloses a sampling inspection method and system for a large batch of spiral spring samples, and the method comprises the following steps: S1, determining a detection point, and obtaining all detection items and historical detection data; s2, dividing the spiral springs produced by the same equipment into the same batch, and randomly selecting a preset number of spiral springs from the batch as detection samples; comprehensively detecting the samples by using detection points, and recording passing conditions; s3, based on a detection result, calculating an abnormal rate of the detection sample on each detection item; calculating the priority of each detection item in combination with the historical detection data and the abnormal rate of the current sample; and S4, determining different detection sequences of different batches according to the abnormal rate and the detection time of different batches of detection samples on each detection item, thereby improving the detection efficiency.

Description

Technical Field

[0001] The present invention relates to the field of random inspection technology, and in particular to a random inspection method and system for large batches of coil spring samples. Background Art

[0002] A coil spring is a mechanical element that uses the elasticity and torsional deformation of metal materials to store and release energy. In the field of rail transit, coil springs are widely used in a variety of equipment and systems, mainly playing the role of shock absorption, buffering, support and connection.

[0003] Spring sampling inspection is an indispensable part of the spring production process. It not only helps to ensure product quality, but also improves production efficiency and reduces costs. Through scientific and reasonable sampling methods and standards, the production quality of springs can be effectively controlled to ensure its reliability and safety in various applications.

[0004] In the existing spring manufacturing industry, the sampling inspection method mostly relies on traditional manual operation. This method has several significant limitations. First, the accuracy of manual sampling is relatively low, because people's subjective judgment is easily affected by factors such as fatigue and experience differences, which may lead to deviations in the test results. Secondly, since the production process of springs usually has a large output, the manual sampling method is inefficient and cannot meet the needs of modern large-scale production. In addition, manual sampling also has problems such as high labor intensity and high cost, which is a considerable burden for enterprises in the long run. Therefore, with the development and progress of science and technology, finding a more efficient and accurate spring sampling method has become an urgent problem to be solved in the industry. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for sampling large quantities of coil spring samples to solve the above technical problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for sampling a large number of coil spring samples, comprising the following steps:

[0008] Step S1: setting a detection point, which is used to perform quality inspection on the coil spring; obtaining all detection items of the detection point, and obtaining historical detection data, which includes the detection time of each detection item;

[0009] It should be noted that the coil springs are an important component of the suspension system of rail vehicles (such as trains, subway trains, etc.); they are usually used in conjunction with hydraulic shock absorbers or air springs to absorb the impact and vibration caused by track unevenness and improve passenger comfort and driving stability; they can also be used as elastic connecting elements between bogies and wheelsets to help disperse the impact force between wheels and rails and protect wheels and tracks from excessive wear;

[0010] Step S2: the coil springs produced by the same equipment are recorded as the same batch, and several batches of coil springs are obtained; a preset number threshold of coil springs are selected from the same batch of coil springs as test samples; all test items are tested on the test samples through the test points to obtain test results, which are the passing conditions of each coil spring on each test item, and the passing conditions include passing the test and abnormal test;

[0011] Step S3: according to the test results, obtain the abnormality rate Pr=P′ / P of the test sample in each test item, wherein P is the total number of coil springs in the test sample, and P′ is the number of coil springs in the test item in the test sample with abnormal detection;

[0012] According to the historical detection data and the abnormality rate of the detection sample, the priority of each detection item in the detection sample is obtained. Among them, Pr ave represents the average abnormality rate of all test items in the test results of the test sample, T is the test time of the test item, T ave represents the mean detection time of all detection items in the historical detection data, λ is a preset correction coefficient and 0<λ<1;

[0013] Step S4: Performing quality inspection on the batch of the test sample according to the priority of each test item of the test sample.

[0014] As a further solution of the present invention: the setting of the detection points is based on computer vision detection technology.

[0015] As a further solution of the present invention: the detection items include size detection and surface defect detection.

[0016] As a further solution of the present invention: the process of quality inspection of the coil spring at the detection point includes:

[0017] Selecting a number of coil spring samples, wherein the coil spring samples are coil springs that meet the standards for all test items; setting a test area at the test point, placing the coil spring samples in the test area, and acquiring sample image data of the coil spring samples, wherein the sample image data includes photographed images of each surface of the coil spring samples; obtaining size data and surface image features of the coil spring samples according to the sample image data, which are recorded as sample size data and sample surface image features, respectively, wherein the surface image features include the grayscale value of each pixel;

[0018] When the detection point performs quality inspection on the coil spring, the dimension data and surface image features of the coil spring are obtained through the detection point; the dimension data are compared with the sample dimension data, and the surface image features are compared with the sample surface image features, and whether the coil spring is qualified is judged based on the comparison results.

[0019] As a further solution of the present invention: the process of judging whether the coil spring is qualified according to the comparison result includes:

[0020] Setting a dimension error threshold and a surface error threshold; obtaining a difference between the dimension data and the sample dimension data, and recording it as a dimension difference; if the dimension difference falls within the dimension error threshold, the dimension detection of the coil spring is qualified; otherwise, the dimension detection of the coil spring is unqualified;

[0021] Obtain the surface image features of the sample to obtain the grayscale value range of the coil spring sample; and obtain the surface image features of the coil spring to obtain the number of pixels in the surface image features whose grayscale values ​​do not belong to the grayscale value range; if the number is less than or equal to the surface error threshold, the surface inspection of the coil spring is qualified; otherwise, the surface inspection of the coil spring is unqualified; if and only if the size inspection and surface inspection of the coil spring are both qualified, the quality inspection of the coil spring is qualified.

[0022] As a further solution of the present invention: the process of setting the quantity threshold includes:

[0023] Get the total number of coil springs in the batch, denoted as N, then the quantity threshold Where N 0 is the preset minimum quantity threshold, ω is the preset proportion threshold and ω≥1.

[0024] As a further solution of the present invention: according to the priority of each test item of the test sample, the process of performing quality inspection on the batch of the test sample includes:

[0025] According to the priority, the test items of the test samples are sorted from large to small, and the quality inspection of each test item is carried out on each spiral spring in the batch according to the sorting.

[0026] As a further solution of the present invention: a sampling inspection system for large batches of coil spring samples, comprising:

[0027] Inspection module: setting inspection points, which are used to perform quality inspection on the coil springs; obtaining all inspection items of the inspection points, and obtaining historical inspection data, which includes the inspection time of each inspection item;

[0028] Sampling inspection module: the spiral springs produced by the same equipment are recorded as the same batch, and several batches of spiral springs are obtained; spiral springs with a preset number threshold are selected from the same batch of spiral springs as test samples; all test items are tested on the test samples through the test points to obtain test results, which are the pass status of each spiral spring on each test item, and the pass status includes test pass and test abnormality;

[0029] Testing sequence determination module: according to the test results, obtain the abnormality rate Pr=P′ / P of the test sample on each test item, where P is the total number of coil springs in the test sample, and P′ is the number of coil springs on which the test items in the test sample are abnormal;

[0030] According to the historical detection data and the abnormality rate of the detection sample, the priority of each detection item in the detection sample is obtained. Among them, Pr ave represents the average abnormality rate of all test items in the test results of the test sample, T is the test time of the test item, T ave represents the mean detection time of all detection items in the historical detection data, λ is a preset correction coefficient and 0<λ<1;

[0031] According to the priority of each test item of the test sample, the batch of the test sample is quality inspected.

[0032] Beneficial effects of the present invention:

[0033] The present invention can selectively select batches and test items that need to be tested in a targeted manner by setting test points and analyzing historical test data, avoiding the inefficient method of conducting comprehensive tests on all coil springs, thereby improving the overall test efficiency. By calculating the abnormality rate and priority, it is possible to accurately identify which test items or batches have a higher risk of quality problems, making the quality inspection work more targeted and helping to quickly discover and solve problems. By sorting the priority of the test samples, the test resources can be reasonably allocated, and those batches or test items that are more likely to have problems can be given priority, ensuring that key quality issues receive timely attention while avoiding unnecessary waste of resources. The comprehensive analysis of historical test data and current test results helps to establish a more scientific quality assessment system, which can effectively improve the overall quality level of coil spring products in the long run. The entire sampling method is based on data analysis, which makes the decision-making process more objective and quantitative, reduces the deviation caused by human subjective judgment, and improves the accuracy and reliability of management decisions. According to the calculated priority, the test items of all batches are sorted in descending order, and those test items with higher priority are tested first within the batch, that is, those test items with higher abnormality rate or relatively short detection time, which effectively improves the detection efficiency, ensures the rational use of resources, and promptly identifies and handles quality problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be further described below in conjunction with the accompanying drawings.

[0035] Figure 1 The present invention is a schematic flow chart of a method for sampling large quantities of coil spring samples. DETAILED DESCRIPTION

[0036] 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.

[0037] See also Figure 1 As shown, the present invention is a method for sampling a large number of coil spring samples, comprising the following steps:

[0038] Step S1: setting a detection point, which is used to perform quality inspection on the coil spring; obtaining all detection items of the detection point, and obtaining historical detection data, which includes the detection time of each detection item;

[0039] It is understandable that one or more inspection points need to be determined, which are specifically used for quality inspection of coil springs; the setting of the inspection points may be based on production lines, equipment or specific inspection requirements; after the inspection points are determined, all inspection items involved in the inspection points need to be clarified; these inspection items may include dimensional measurement, material testing, elasticity testing, etc.; at the same time, historical inspection data needs to be collected, which may include information such as the inspection results and inspection time of each inspection item;

[0040] As a preferred embodiment of the present invention, the setting of the detection points is based on computer vision detection technology;

[0041] It should be noted that computer vision inspection technology is a technology that uses computer algorithms and image processing technology to simulate the visual function of the human eye. It acquires images through a camera or other image acquisition device, and then uses image processing algorithms to analyze the features in the image, thereby achieving automatic detection and recognition. In the quality inspection of coil springs, computer vision inspection technology can be used to automatically identify and measure the size of coil springs, detect surface defects and other quality problems.

[0042] As a preferred embodiment of the present invention, the detection items include size detection and surface defect detection;

[0043] It should be noted that the computer vision system can measure the key dimensional parameters of the coil spring, such as diameter, pitch, length, etc. through image processing technology to ensure that they meet the design specifications; it can also capture the surface details of the coil spring through high-resolution images and use image recognition algorithms to detect and classify various surface defects;

[0044] It is understandable that this computer vision-based inspection method can improve the speed and accuracy of inspection, reduce human errors, and reduce labor intensity and costs; realize automated online inspection, monitor product quality on the production line in real time, detect problems in time and make adjustments, thereby improving production efficiency and product quality;

[0045] As a preferred embodiment of the present invention, the process of quality inspection of the coil spring by the detection point includes:

[0046] Selecting a number of coil spring samples, wherein the coil spring samples are coil springs that meet the standards for all test items; setting a test area at the test point, placing the coil spring samples in the test area, and acquiring sample image data of the coil spring samples, wherein the sample image data includes photographed images of each surface of the coil spring samples; obtaining size data and surface image features of the coil spring samples according to the sample image data, which are recorded as sample size data and sample surface image features, respectively, wherein the surface image features include the grayscale value of each pixel;

[0047] When the detection point performs quality inspection on the coil spring, the dimension data and surface image features of the coil spring are obtained through the detection point; the dimension data is compared with the sample dimension data, and the surface image features are compared with the sample surface image features, and whether the coil spring is qualified is determined according to the comparison results;

[0048] The process of judging whether the coil spring is qualified according to the comparison result includes:

[0049] Setting a dimension error threshold and a surface error threshold; obtaining a difference between the dimension data and the sample dimension data, and recording it as a dimension difference; if the dimension difference falls within the dimension error threshold, the dimension detection of the coil spring is qualified; otherwise, the dimension detection of the coil spring is unqualified;

[0050] Obtain the grayscale value range of the coil spring sample by acquiring the surface image feature of the sample; and obtain the surface image feature of the coil spring, and obtain the number of pixels in the surface image feature whose grayscale values ​​do not belong to the grayscale value range; if the number is less than or equal to the surface error threshold, the surface inspection of the coil spring is qualified; otherwise, the surface inspection of the coil spring is unqualified; if and only if the size inspection and surface inspection of the coil spring are both qualified, the quality inspection of the coil spring is qualified;

[0051] It can be understood that a number of known qualified coil springs are selected as samples, a detection area is set at the detection point, the coil spring sample is placed in the area, and an image acquisition device (such as a camera) is used to obtain the image data of the sample; based on the collected sample image data, the size data and surface image features of the coil spring are extracted through image processing technology; the size data may include diameter, pitch, etc., and the surface image features include information such as the gray value of each pixel; in the actual detection process, the size data and surface image features of the coil spring to be detected are also obtained through the detection point; the real-time acquired size data is compared with the previously extracted sample size data to calculate the size difference; at the same time, the real-time acquired surface image features are compared with the sample surface image features, especially the difference in gray value; computer vision technology is used to realize the automated quality inspection of coil springs, improve the inspection efficiency and accuracy, reduce human errors, and is suitable for large-scale production line quality control;

[0052] Step S2: the coil springs produced by the same equipment are recorded as the same batch, and several batches of coil springs are obtained; a preset number threshold of coil springs are selected from the same batch of coil springs as test samples; all test items are tested on the test samples through the test points to obtain test results, which are the passing conditions of each coil spring on each test item, and the passing conditions include passing the test and abnormal test;

[0053] It should be noted that the coil springs produced by the same equipment are divided into the same batch; in this way, several batches of coil springs can be obtained; the basis for batch division is the consistency of the production equipment to ensure that the coil springs in the same batch have similar production conditions and quality characteristics;

[0054] As a preferred embodiment of the present invention, the process of setting the quantity threshold includes:

[0055] Get the total number of coil springs in the batch, denoted as N, then the quantity threshold Where N 0 is the preset minimum number threshold, ω is the preset proportion threshold and ω≥1;

[0056] It is worth noting that the process of setting the quantity threshold ensures a certain basic sample size while flexibly adjusting the number of random inspections according to the size of different batches to ensure the representativeness of the samples and the accuracy of the detection;

[0057] Step S3: according to the test results, obtain the abnormality rate Pr=P′ / P of the test sample in each test item, wherein P is the total number of coil springs in the test sample, and P′ is the number of coil springs in the test item in the test sample with abnormal detection;

[0058] According to the historical detection data and the abnormality rate of the detection sample, the priority of each detection item in the detection sample is obtained. Among them, Pr ave represents the average abnormality rate of all test items in the test results of the test sample, T is the test time of the test item, T ave represents the mean detection time of all detection items in the historical detection data, λ is a preset correction coefficient and 0<λ<1;

[0059] It is understandable that the calculation of the priority takes into account the abnormality rate and detection time of the detection item, and the correction coefficient is adjusted to balance the impact of these two factors on the priority; detection items with high abnormality rate and short detection time will have higher priority, so that resources can be allocated more effectively and the detection process can be optimized;

[0060] Step S4: performing quality inspection on the batch of the test sample according to the priority of each test item of the test sample;

[0061] As a preferred embodiment of the present invention, according to the priority of each test item of the test sample, the process of performing quality inspection on the batch of the test sample includes:

[0062] According to the priority, the test items of the test samples are sorted from large to small, and the quality inspection of each test item is carried out on each spiral spring in the batch according to the sorting;

[0063] It is understandable that for the quality inspection of large quantities of coil springs, optimizing the inspection process and prioritizing inspection items with higher abnormality rates and shorter inspection times improves the efficiency of the entire production process and product quality.

[0064] A sampling inspection system for large batches of coil spring samples, comprising:

[0065] Inspection module: setting inspection points, which are used to perform quality inspection on the coil springs; obtaining all inspection items of the inspection points, and obtaining historical inspection data, which includes the inspection time of each inspection item;

[0066] Sampling inspection module: the spiral springs produced by the same equipment are recorded as the same batch, and several batches of spiral springs are obtained; spiral springs with a preset number threshold are selected from the same batch of spiral springs as test samples; all test items are tested on the test samples through the test points to obtain test results, which are the pass status of each spiral spring on each test item, and the pass status includes test pass and test abnormality;

[0067] Testing sequence determination module: according to the test results, obtain the abnormality rate Pr=P′ / P of the test sample on each test item, where P is the total number of coil springs in the test sample, and P′ is the number of coil springs on which the test items in the test sample are abnormal;

[0068] According to the historical detection data and the abnormality rate of the detection sample, the priority of each detection item in the detection sample is obtained. Among them, Pr ave represents the average abnormality rate of all test items in the test results of the test sample, T is the test time of the test item, T ave represents the mean detection time of all detection items in the historical detection data, λ is a preset correction coefficient and 0<λ<1;

[0069] According to the priority of each test item of the test sample, the batch of the test sample is quality inspected.

[0070] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for sampling large quantities of coil spring samples, characterized in that: The following steps are involved: Step S1: setting a detection point, which is used to perform quality inspection on the coil spring; obtaining all detection items of the detection point, and obtaining historical detection data, which includes the detection time of each detection item; Step S2: the coil springs produced by the same equipment are recorded as the same batch, and several batches of coil springs are obtained; a preset number threshold of coil springs are selected from the same batch of coil springs as test samples; all test items are tested on the test samples through the test points to obtain test results, which are the passing conditions of each coil spring on each test item, and the passing conditions include passing the test and abnormal test; Step S3: according to the test results, obtain the abnormality rate Pr=P′ / P of the test sample in each test item, wherein P is the total number of coil springs in the test sample, and P′ is the number of coil springs in the test item in the test sample with abnormal detection; According to the historical detection data and the abnormality rate of the detection sample, the priority of each detection item in the detection sample is obtained. Among them, Pr ave represents the average abnormality rate of all test items in the test results of the test sample, T is the test time of the test item, T ave represents the mean detection time of all detection items in the historical detection data, λ is a preset correction coefficient and 0<λ<1; Step S4: According to the priority of each test item of the test sample, the batch of the test sample is quality inspected.

2. A method for sampling large quantities of coil spring samples according to claim 1, characterized in that: In step S1, the detection points are set based on computer vision detection technology.

3. A method for sampling large quantities of coil spring samples according to claim 1, characterized in that: In step S1, the inspection items include size inspection and surface defect inspection.

4. A method for sampling large quantities of coil spring samples according to claim 1, characterized in that: In step S1, the process of quality inspection of the coil spring at the inspection point includes: Selecting a number of coil spring samples, wherein the coil spring samples are coil springs that meet the standards for all test items; setting a test area at the test point, placing the coil spring samples in the test area, and acquiring sample image data of the coil spring samples, wherein the sample image data includes photographed images of each surface of the coil spring samples; obtaining size data and surface image features of the coil spring samples according to the sample image data, which are recorded as sample size data and sample surface image features, respectively, wherein the surface image features include the grayscale value of each pixel; When the detection point performs quality inspection on the coil spring, the dimension data and surface image features of the coil spring are obtained through the detection point; the dimension data are compared with the sample dimension data, and the surface image features are compared with the sample surface image features, and whether the coil spring is qualified is judged based on the comparison results.

5. A method for sampling large quantities of coil spring samples according to claim 4, characterized in that: In step S1, the process of judging whether the coil spring is qualified according to the comparison result includes: Setting a dimension error threshold and a surface error threshold; obtaining a difference between the dimension data and the sample dimension data, and recording it as a dimension difference; if the dimension difference falls within the dimension error threshold, the dimension detection of the coil spring is qualified; otherwise, the dimension detection of the coil spring is unqualified; Obtain the surface image features of the sample to obtain the grayscale value range of the coil spring sample; and obtain the surface image features of the coil spring to obtain the number of pixels in the surface image features whose grayscale values ​​do not belong to the grayscale value range; if the number is less than or equal to the surface error threshold, the surface inspection of the coil spring is qualified; otherwise, the surface inspection of the coil spring is unqualified; if and only if the size inspection and surface inspection of the coil spring are both qualified, the quality inspection of the coil spring is qualified.

6. A method for sampling large quantities of coil spring samples according to claim 1, characterized in that: In step S2, the process of setting the quantity threshold includes: Get the total number of coil springs in the batch, denoted as N, then the quantity threshold Wherein N0 is the preset minimum number threshold, ω is the preset proportion threshold and ω≥1.

7. A method for sampling large quantities of coil spring samples according to claim 1, characterized in that: In step S4, according to the priority of each test item of the test sample, the process of performing quality inspection on the batch of the test sample includes: According to the priority, the test items of the test samples are sorted from large to small, and the quality inspection of each test item is carried out on each spiral spring in the batch according to the sorting.

8. A sampling inspection system for large quantities of coil spring samples, characterized in that: include: Inspection module: setting inspection points, which are used to perform quality inspection on the coil springs; obtaining all inspection items of the inspection points, and obtaining historical inspection data, which includes the inspection time of each inspection item; Sampling inspection module: the spiral springs produced by the same equipment are recorded as the same batch, and several batches of spiral springs are obtained; spiral springs with a preset number threshold are selected from the same batch of spiral springs as test samples; all test items are tested on the test samples through the test points to obtain test results, which are the pass status of each spiral spring on each test item, and the pass status includes test pass and test abnormality; Testing sequence determination module: according to the test results, obtain the abnormality rate Pr=P′ / P of the test sample on each test item, where P is the total number of coil springs in the test sample, and P′ is the number of coil springs on which the test items in the test sample are abnormal; According to the historical detection data and the abnormality rate of the detection sample, the priority of each detection item in the detection sample is obtained. Among them, Pr ave represents the average abnormality rate of all test items in the test results of the test sample, T is the test time of the test item, T ave represents the mean detection time of all detection items in the historical detection data, λ is a preset correction coefficient and 0<λ<1; According to the priority of each test item of the test sample, the batch of the test sample is quality inspected.

Images