Spring intelligent digital quenching process optimization method and system

Abstract

The application relates to the technical field of spring quenching process optimization, in particular to an intelligent digital quenching process optimization method and system for springs. First, a plurality of springs of the same type are acquired to construct a first spring detection set; second, the springs in the first spring detection set are screened to acquire a second spring detection set; third, the second spring detection set is divided into a plurality of spring detection subsets according to the number of quenching process parameter sets; then, the springs in each spring detection subset are quenched according to corresponding quenching process parameters, and the quenched springs are subjected to spring quenching performance detection to acquire spring quenching performance detection values corresponding to each spring detection subset; finally, the optimal quenching process parameter set is acquired according to the spring quenching performance detection values. The method can accurately acquire the optimal quenching process parameter set by combining the performances of the quenched springs, so that the quenching process optimization of the springs is realized.

Description

Technical Field

[0001] This invention relates to the field of spring quenching process optimization technology, specifically to an intelligent digital quenching process optimization method and system for springs. Background Technology

[0002] With the continuous development of the manufacturing industry, especially the widespread application of precision machinery and automated equipment, springs, as a key mechanical component, play a vital role in the industrial field. The performance of springs directly affects the stability and reliability of mechanical systems; therefore, it is necessary to optimize the spring manufacturing process to ensure spring performance.

[0003] The manufacturing process of springs includes cold forming, quenching, and tempering. The quenching step involves heating the spring to a suitable temperature and then rapidly cooling it to increase its hardness and strength. Key parameters such as temperature and cooling rate need to be controlled during quenching to ensure quality; therefore, the quenching process for springs needs to be optimized.

[0004] The optimization of the quenching process can be based on the quality of the spring after quenching, allowing for adjustments and improvements to the quenching process parameters. While existing methods for detecting the quality of quenched springs can achieve quality inspection, the detection is not comprehensive enough, leading to inaccurate quenching process optimization and an inability to obtain springs with high overall quality.

[0005] To address this, a method and system for optimizing the intelligent digital quenching process of springs are proposed. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent digital quenching process optimization method and system for springs. First, multiple springs of the same type are acquired to construct a first spring detection set. Second, the springs in the first spring detection set are screened to obtain a second spring detection set. Third, based on the number of quenching process parameter sets, the second spring detection set is divided into multiple spring detection subsets. Then, the springs in each spring detection subset are quenched according to the corresponding quenching process parameters, and the quenched springs are subjected to spring quenching performance testing to obtain the spring quenching performance test value corresponding to each spring detection subset. Finally, based on the spring quenching performance test values, the optimal quenching process parameter set is obtained.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for optimizing the intelligent digital quenching process of springs, comprising:

[0009] Collect multiple springs of the same type and construct the first spring detection set;

[0010] The springs in the first spring detection set are filtered to obtain the second spring detection set;

[0011] Obtain the M sets of quenching process parameters, and divide the second spring detection set into M sets of spring detection subsets;

[0012] The springs in each spring detection subset are quenched according to the quenching process parameters of the corresponding quenching process parameter set.

[0013] The quenching performance of the springs after quenching is tested, and the quenching performance test value corresponding to each spring test subset is obtained. The specific steps include: cleaning the surface of the quenched springs; performing spring appearance defect detection and spring hardness detection on the cleaned springs respectively, and obtaining the comprehensive value of spring appearance defect detection and the comprehensive value of spring hardness detection for the spring test subset.

[0014] Each spring detection subset is further divided into a spring tension detection subset and a spring compression detection subset;

[0015] Perform spring tension testing on the springs in each spring tension testing subset to obtain the comprehensive spring tension testing value for the corresponding spring testing subset; perform spring compression testing on the springs in each spring compression testing subset to obtain the comprehensive spring compression testing value for the corresponding spring testing subset.

[0016] The spring quenching performance test value of each spring test subset is obtained based on the four comprehensive test values ​​corresponding to each spring test subset;

[0017] Based on the spring quenching performance test values, the optimal set of quenching process parameters is obtained.

[0018] Furthermore, none of the springs in the first spring detection set have undergone quenching treatment; each spring undergoes the same manufacturing steps before quenching.

[0019] Further, the step of screening the springs in the first spring detection set to obtain the second spring detection set includes: performing spring appearance defect detection on the springs in the first spring detection set, and screening out the springs that pass the spring appearance defect detection to construct the second spring detection set; the spring appearance defect detection process includes:

[0020] Obtain the spring appearance image and input the spring appearance image into the spring appearance defect detection model;

[0021] The spring appearance defect detection model includes a spring appearance image input layer, a spring appearance feature map acquisition layer, a spring appearance feature map fusion layer, and a spring external defect detection value acquisition layer.

[0022] The spring appearance feature map acquisition layer is used to perform a convolution operation on the spring appearance image to obtain multiple appearance feature maps; the spring appearance feature map fusion layer is used to fuse the multiple appearance feature maps to obtain spring appearance features.

[0023] The appearance features of the spring include spring shape features and spring defect features; the spring shape features include the maximum diameter of the spring wire, the minimum diameter of the spring wire, the maximum outer diameter of the spring, the minimum inner diameter of the spring, and the non-support coil pitch;

[0024] The spring defect features include spring surface crack features, spring surface foreign matter features, and spring surface dent features;

[0025] A first spring defect detection value is obtained based on the spring shape characteristics; a second spring defect detection value is obtained based on the spring defect characteristics; and a first spring appearance defect detection value is obtained based on the first spring defect detection value and the second spring defect detection value.

[0026] When the first spring appearance defect detection value of the spring is greater than the set defect detection threshold, the spring appearance defect detection is determined to be unqualified; otherwise, the spring appearance defect detection is determined to be qualified.

[0027] Furthermore, the quenching process parameter set includes: each quenching process parameter set includes N quenching process parameters; the quenching process parameter values ​​of different quenching process parameter sets are not completely the same.

[0028] Furthermore, the spring tension detection process includes:

[0029] Before performing spring tension testing, obtain the first spring shape characteristics of each spring;

[0030] A set tension is applied to the spring, and the force is applied. After a certain time, the tension is released; the extension length of the spring is obtained when the tension is released; after the tension is released... After a certain time, the second spring shape feature and spring defect feature of each spring are obtained; the spring tension detection value of the spring is obtained based on the first spring shape feature, the second spring shape feature, the stretch length and the spring defect feature; the comprehensive spring tension detection value of the spring detection subset corresponding to the spring tension detection subset is obtained based on the spring tension detection value of each spring in the spring tension detection subset.

[0031] Furthermore, the spring compression detection process includes:

[0032] Obtain the third spring shape characteristics of each spring before performing spring compression testing;

[0033] Apply a set pressure to the spring, and apply force. After a certain time, the pressure is released; the compression length of the spring is obtained when the pressure is released; and the pressure is released at the following time... After a certain time, the fourth spring shape feature and spring defect feature of each spring are obtained; the spring compression detection value of the spring is obtained based on the third spring shape feature, the fourth spring shape feature, the compression length and the spring defect feature; the comprehensive spring compression detection value of the spring detection subset corresponding to the spring compression detection subset is obtained based on the spring compression detection value of each spring in the spring compression detection subset.

[0034] Furthermore, obtaining the optimal set of quenching process parameters based on the spring quenching performance test values ​​includes:

[0035] The set of quenching process parameters corresponding to the spring detection subset with the largest spring quenching performance test value is selected as the optimal quenching process parameter set.

[0036] A smart digital quenching process optimization system for springs, comprising:

[0037] First Spring Detection Set Acquisition Unit: Used to acquire multiple springs of the same type and construct a first spring detection set;

[0038] Second spring detection set acquisition unit: used to filter the springs in the first spring detection set to acquire the second spring detection set;

[0039] Quenching Spring Detection Subset Division Unit: Used to obtain M sets of quenching process parameters and divide the second spring detection set into M sets of spring detection subsets;

[0040] Quenching unit: used to quench the springs of each spring detection subset according to the quenching process parameters of the corresponding quenching process parameter set;

[0041] Quenching performance testing unit: used to test the quenching performance of the springs after quenching, and to obtain the spring quenching performance test value corresponding to each spring test subset; the spring quenching performance test includes spring appearance defect detection, spring hardness detection, spring tensile testing and spring compression testing;

[0042] Optimal quenching process parameter set acquisition unit: used to acquire the optimal quenching process parameter set based on the spring quenching performance test value.

[0043] Furthermore, the step of obtaining the spring quenching performance test value includes: cleaning the surface of the quenched spring; performing spring appearance defect detection and spring hardness detection on the cleaned spring respectively, and obtaining the comprehensive value of spring appearance defect detection and the comprehensive value of spring hardness detection of the spring detection subset.

[0044] Each spring detection subset is further divided into a spring tension detection subset and a spring compression detection subset;

[0045] Perform spring tension testing on the springs in each spring tension testing subset and obtain the comprehensive spring tension testing value for the corresponding spring testing subset;

[0046] Perform spring compression testing on the springs in each spring compression testing subset to obtain the comprehensive spring compression test value for the corresponding spring testing subset;

[0047] The spring quenching performance test value of each spring test subset is obtained based on the four comprehensive test values ​​corresponding to each spring test subset.

[0048] Furthermore, the spring tension detection process includes:

[0049] Before performing spring tension testing, obtain the first spring shape characteristics of each spring;

[0050] A set tension is applied to the spring, and the force is applied. After a certain time, the tension is released; the extension length of the spring is obtained when the tension is released; after the tension is released... After a certain time, the second spring shape feature and spring defect feature of each spring are obtained; the spring tension detection value of the spring is obtained based on the first spring shape feature, the second spring shape feature, the stretch length and the spring defect feature; the comprehensive spring tension detection value of the spring detection subset corresponding to the spring tension detection subset is obtained based on the spring tension detection value of each spring in the spring tension detection subset.

[0051] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0052] 1. Before quenching, this invention performs appearance defect detection on each spring using a spring appearance defect detection model. The model extracts the shape and defect features of the spring and calculates a first spring appearance defect detection value. Springs are then screened based on this value. This method can detect various appearance defects in springs and eliminate those that fail the appearance defect detection, avoiding the impact of inherent spring defects on the quality of the quenched springs and thus improving the accuracy of quenching process optimization.

[0053] 2. This invention sets up different quenching environments and quenches the springs in these environments. By performing appearance defect detection, hardness testing, tensile testing, and compression testing on the quenched springs, the quenching performance values ​​are obtained. This method can combine various characteristics of the quenched spring to accurately obtain its comprehensive performance, thereby improving the accuracy of quenching process optimization.

[0054] 3. In performing tensile testing on springs, this invention calculates a comprehensive tensile test value by considering the tensile length after applying tension, spring defect characteristics, and changes in the spring's shape before and after tension. This comprehensive tensile test value not only measures the spring's tensile performance but also its recovery ability after tension. This method accurately obtains various properties of the spring under tension, accurately obtains the comprehensive tensile test value, and thus accurately obtains the comprehensive performance of the spring after quenching. Attached Figure Description

[0055] Figure 1 A flowchart of an intelligent digital quenching process optimization method for springs provided in an embodiment of the present invention;

[0056] Figure 2 A structural diagram of an intelligent digital quenching process optimization system for springs provided in an embodiment of the present invention;

[0057] Figure 3 This is a flowchart of the process for obtaining spring quenching performance test values ​​provided in an embodiment of the present invention;

[0058] Figure 4 A flowchart for obtaining the comprehensive value of spring tension testing provided in an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] With the continuous development of the manufacturing industry, especially the widespread application of precision machinery and automated equipment, springs, as a key mechanical component, play a vital role in the industrial field. The performance of springs directly affects the stability and reliability of mechanical systems; therefore, it is necessary to optimize the spring manufacturing process to ensure spring performance.

[0061] The manufacturing process of springs includes cold forming, quenching, and tempering. The quenching step involves heating the spring to a suitable temperature and then rapidly cooling it to increase its hardness and strength. Key parameters such as temperature and cooling rate need to be controlled during quenching to ensure quality; therefore, the quenching process for springs needs to be optimized.

[0062] This invention provides an intelligent digital quenching process optimization method for springs. The method is applied to an intelligent digital quenching process optimization system for springs. Specific method flowcharts and system structure diagrams are provided below. Figure 1 and Figure 2 .

[0063] Example 1

[0064] Reference Figure 1 S10 is applied to the first spring detection set acquisition unit of an intelligent digital quenching process optimization system for springs.

[0065] Furthermore, in this embodiment, 350 springs of the same type from factory A were collected to construct a first spring detection set; the same type means that the springs have the same material, shape, size and target quality after quenching; none of the springs in the first spring detection set have undergone quenching treatment; each spring has undergone the same manufacturing steps before quenching.

[0066] In this embodiment, multiple springs produced in the same batch were obtained and subjected to the same treatment before quenching. This method can avoid the influence of spring manufacturing differences on the quenching results, thereby improving the accuracy of quenching process optimization.

[0067] Reference Figure 1 S20 is applied to the second spring detection set acquisition unit of an intelligent digital quenching process optimization system for springs.

[0068] Further, the springs in the first spring inspection set are subjected to spring appearance defect inspection, and springs that pass the spring appearance defect inspection are selected to construct a second spring inspection set; the spring appearance defect inspection process includes:

[0069] Obtain the spring appearance image and input the spring appearance image into the spring appearance defect detection model;

[0070] The spring appearance defect detection model includes a spring appearance image input layer, a spring appearance feature map acquisition layer, a spring appearance feature map fusion layer, and a spring external defect detection value acquisition layer.

[0071] The spring appearance feature map acquisition layer is used to perform a convolution operation on the spring appearance image to obtain multiple appearance feature maps; the spring appearance feature map fusion layer is used to fuse the multiple appearance feature maps to obtain spring appearance features.

[0072] The spring's appearance features include spring shape features and spring defect features; the spring shape features include the maximum diameter of the spring wire, the minimum diameter of the spring wire, the maximum outer diameter of the spring, the minimum inner diameter of the spring, and the non-support coil pitch; the spring defect features include spring surface crack features, spring surface foreign object features, and spring surface depression features; the surface crack features include the number of cracks and the crack area of ​​each crack; the spring surface foreign object features include the number of foreign objects; the foreign objects include foreign objects such as sand particles generated in the production steps before the spring is quenched; the spring surface depression features include the number of depressions and the area of ​​each depression;

[0073] Obtain the standard shape features of the spring corresponding to the spring; obtain a first spring defect detection value based on the spring shape features and the standard shape features of the spring; obtain a second spring defect detection value based on the spring defect features; obtain a first spring appearance defect detection value based on the first spring defect detection value and the second spring defect detection value.

[0074] The formula for calculating the first spring defect detection value is:

[0075] ;

[0076] in, This is represented as the first spring defect detection value; This is represented by the spring shape features obtained through the spring appearance defect detection model; Represented as the standard shape characteristics of a spring; It is represented as a nonlinear function; the larger the first spring defect detection value, the greater the difference between the spring shape characteristics and the standard spring shape characteristics;

[0077] The formula for calculating the second spring defect detection value is:

[0078] ;

[0079] in, This is represented as the second spring defect detection value; This is expressed as the number of cracks; Let represent the crack area of ​​the i-th crack. This is expressed as the number of foreign objects on the surface; This is expressed as the number of depressions; Let be the area of ​​the j-th depression; , and These are respectively represented as crack coefficient, foreign matter coefficient, and indentation coefficient; It is represented as an exponential function with the natural constant as the base; , and The default values ​​are all The specific details can be modified according to the actual application.

[0080] The formula for calculating the first spring appearance defect detection value is:

[0081] ;

[0082] in, This is represented as the detected value of the first spring's appearance defect; and These are respectively represented as the defect coefficient of the first spring and the defect coefficient of the second spring; and The default value is 0.5, but it can be changed according to the actual application.

[0083] When the first spring appearance defect detection value of the spring is greater than the set defect detection threshold, the spring appearance defect detection is determined to be unqualified; otherwise, the spring appearance defect detection is determined to be qualified; the defect detection threshold is obtained based on experience.

[0084] In this embodiment, before quenching, each spring undergoes appearance defect detection using a spring appearance defect detection model. This model extracts the shape and defect features of the spring and calculates a first spring appearance defect detection value. Springs are then screened based on this value. This method can detect various appearance defects in springs and eliminate those that fail the appearance defect detection, avoiding the impact of inherent spring defects on the quality of the quenched springs and thus improving the accuracy of quenching process optimization.

[0085] Reference Figure 1 S30 in the text refers to the quenching spring detection subset partitioning unit in an intelligent digital quenching process optimization system for springs.

[0086] Further, M sets of quenching process parameters are obtained, and the second spring detection set is divided into M spring detection subsets. The quenching process parameter sets include: each set includes N quenching process parameters; the quenching process parameter values ​​of different sets are not completely the same; each set includes heating temperature, holding time, cooling medium type, cooling rate, and cooling time, etc. In this embodiment, five sets of quenching process parameters are collected, and the second spring detection set is divided into five spring detection subsets. Some quenching process parameters of each set are shown in Table 1.

[0087] Table 1. Quenching process parameter values ​​for some examples.

[0088]

[0089] In this embodiment, this step involves dividing the second spring detection set into multiple spring detection subsets based on the number of quenching process parameter sets, with each spring detection subset corresponding one-to-one with a quenching process parameter set. This method can verify the quenching effect of each quenching process parameter set on the spring, thereby selecting the optimal quenching process parameter set.

[0090] Reference Figure 1 The S40 is used in the quenching unit of an intelligent digital quenching process optimization system for springs.

[0091] Furthermore, in this embodiment, the springs of each spring detection subset are quenched according to the quenching process parameters of the corresponding quenching process parameter set.

[0092] Reference Figure 1 The S50 is used in the quenching performance testing unit of an intelligent digital quenching process optimization system for springs.

[0093] Furthermore, the quenched springs are subjected to spring quenching performance testing, and the spring quenching performance test values ​​corresponding to each spring test subset are obtained, with reference to... Figure 3 The specific steps include:

[0094] The surface of the quenched spring is cleaned; the cleaned spring is subjected to spring appearance defect detection and spring hardness detection respectively, and the comprehensive value of spring appearance defect detection and comprehensive value of spring hardness detection of the spring detection subset are obtained.

[0095] Each spring detection subset is further divided into a spring tension detection subset and a spring compression detection subset;

[0096] Perform spring tension testing on the springs in each spring tension testing subset to obtain the comprehensive spring tension testing value for the corresponding spring testing subset; perform spring compression testing on the springs in each spring compression testing subset to obtain the comprehensive spring compression testing value for the corresponding spring testing subset.

[0097] The spring quenching performance test value of each spring test subset is obtained based on the four comprehensive test values ​​corresponding to each spring test subset. The calculation formula is as follows:

[0098] ;

[0099] in, This represents the spring quenching performance test value of the kth spring test subset; This represents the comprehensive value of spring appearance defect detection for the kth spring detection subset. This represents the comprehensive value of the spring stiffness test for the kth spring test subset. This represents the comprehensive value of the spring tension test for the kth spring test subset. This is represented as the comprehensive value of spring compression detection for the k-th subset of spring detection. , , and These are respectively represented as the spring appearance defect coefficient, spring hardness coefficient, spring tensile coefficient, and spring compression coefficient; , , and The default value is 1, but it can be changed according to the actual application.

[0100] In this embodiment, different quenching environments are set up, and the spring is quenched in different environments. By performing appearance defect detection, hardness testing, tensile testing, and compression testing on the quenched spring, the quenching performance test values ​​are obtained. This method can combine various characteristics of the spring after quenching to accurately obtain the comprehensive performance of the quenched spring, thereby improving the accuracy of quenching process optimization.

[0101] Furthermore, the process of obtaining the comprehensive value of the spring appearance defect detection includes:

[0102] Each quenched spring is subjected to spring appearance defect detection to obtain a first spring appearance defect detection value for each spring; the first spring appearance defect detection values ​​of the springs in each spring detection subset are averaged to obtain a comprehensive spring appearance defect detection value corresponding to the spring detection subset.

[0103] Furthermore, the process of obtaining the comprehensive value of the spring stiffness test includes:

[0104] H hardness testing points are selected on each quenched spring; the hardness value of each hardness testing point is obtained; the hardness values ​​of the hardness testing points on each spring are averaged to obtain the average hardness of each spring.

[0105] The spring hardness test is judged to be qualified based on the average hardness value, and the hardness test qualification rate of each spring test subset is obtained; the hardness test qualification rate is the comprehensive value of the spring hardness test corresponding to the spring test subset.

[0106] Furthermore, referring to Figure 4 The spring tension detection process includes:

[0107] Before performing spring tension testing, obtain the first spring shape characteristics of each spring;

[0108] A set tension is applied to the spring, and the force is applied. After a certain time, the tension is released; The time can be 60 seconds or 90 seconds, etc.; the extension length of the spring is obtained when the tension is released; when the tension is released... After a certain time, the second spring shape characteristics and spring defect characteristics of each spring are obtained; The time can be 30 seconds or 40 seconds, etc.; the spring tension detection value of the spring is obtained based on the shape characteristics of the first spring, the shape characteristics of the second spring, the stretch length, and the spring defect characteristics;

[0109] The formula for calculating the spring tension test value is as follows:

[0110] ;

[0111] in, This represents the spring tension detection value of the p-th spring in the spring tension detection subset corresponding to the k-th spring tension detection subset; This represents the extension length of the p-th spring; This is represented as the set standard stretch length; This is represented as the spring stretching shape defect value of the spring obtained based on the first spring shape feature and the second spring shape feature; This is represented as the second spring defect detection value of the p-th spring obtained based on the spring defect characteristics; , and These are respectively represented as the tensile length factor, tensile shape factor, and tensile defect factor;

[0112] The formula for calculating the spring tension shape defect value is as follows:

[0113] ;

[0114] in, This represents the first spring shape feature of the p-th spring in the spring tension detection subset corresponding to the k-th spring tension detection subset; The shape feature of the second spring, represented by the p-th spring; It is represented as the same nonlinear function as the calculation process for the first spring defect detection value;

[0115] The average spring tension test values ​​of each spring in the spring tension test subset are used to obtain the comprehensive spring tension test value of the corresponding spring tension test subset.

[0116] In this embodiment, during the spring tensile testing, the comprehensive value of the spring tensile test is calculated by considering the tensile length after applying tension, the characteristics of spring defects, and the changes in the shape characteristics of the spring before and after tension. The comprehensive value of the spring tensile test not only measures the tensile performance of the spring but also its recovery ability after tension. This method can accurately obtain various properties of the spring under tension, accurately obtain the comprehensive value of the spring tensile test, and thus accurately obtain the comprehensive performance of the spring after quenching.

[0117] Furthermore, the spring compression detection process includes:

[0118] Obtain the third spring shape characteristics of each spring before performing spring compression testing;

[0119] Apply a set pressure to the spring, and apply force. After a certain time, the pressure is released; The duration can be 60 seconds or 90 seconds, etc.; the compression length of the spring is obtained when the pressure is released; when the pressure is released... After a certain time, the fourth spring shape characteristics and spring defect characteristics of each spring are obtained; The time can be 30 seconds or 40 seconds, etc.; the spring compression detection value of the spring is obtained based on the shape characteristics of the third spring, the shape characteristics of the fourth spring, the compression length, and the spring defect characteristics;

[0120] The formula for calculating the spring tension test value is as follows:

[0121] ;

[0122] in, This represents the spring tension detection value of the qth spring in the kth spring compression detection subset; This is expressed as the compression length of the q-th spring; This is represented as the set standard compression length; This is represented as the spring compression shape defect value of the spring obtained based on the third spring shape feature and the fourth spring shape feature; This is represented as the second spring defect detection value of the qth spring obtained based on the spring defect characteristics; , and These are respectively represented as the compression length factor, compression shape factor, and compression defect factor;

[0123] The formula for calculating the spring compression shape defect value is as follows:

[0124] ;

[0125] in, The third spring shape feature is represented by the spring q in the k-th spring compression detection subset; The fourth spring shape feature is represented by the q-th spring in the k-th spring compression detection subset; It is represented as the same nonlinear function as the calculation process for the first spring defect detection value;

[0126] The average spring compression detection values ​​of each spring in the spring compression detection subset are used to obtain the comprehensive spring compression detection value of the spring detection subset corresponding to the spring compression detection subset.

[0127] In this embodiment, during the spring compression test, a comprehensive spring compression test value is calculated by considering the compressed length after pressure is applied, spring defect characteristics, and changes in the spring's shape before and after compression. This comprehensive spring compression test value not only measures the spring's compression performance but also its recovery ability after compression. This method accurately obtains various properties of the spring under pressure, accurately obtains the comprehensive spring compression test value, and thus accurately obtains the comprehensive performance of the spring after quenching.

[0128] Reference Figure 1 The S60 in the text is applied to the optimal quenching process parameter set acquisition unit of an intelligent digital quenching process optimization system for springs.

[0129] Furthermore, the set of quenching process parameters corresponding to the spring detection subset with the largest spring quenching performance test value is selected as the optimal quenching process parameter set; in this embodiment, the quenching process parameter set with the sequence number 4 is calculated to have the largest spring quenching performance test value, that is, the quenching process parameter set with the sequence number 4 is selected as the optimal quenching process parameter set.

[0130] In this embodiment, this step obtains the optimal set of quenching process parameters based on the spring quenching performance test values ​​to optimize the spring quenching process. This method can obtain the optimal quenching process parameters in advance based on samples, thereby optimizing the quenching process for the same type of spring and ensuring the quality of the spring after quenching.

[0131] To verify the effectiveness of the spring quenching process optimization method used in this embodiment, 1200 springs of the same type that passed the appearance defect inspection were collected and randomly grouped into five spring subsets, each containing 240 springs. The 240 springs were then quenched according to the quenching process parameters of the five quenching process parameter sets. The quenched springs were then subjected to quality inspection, and the spring quenching performance test values ​​and corresponding spring quality pass rates were obtained for each spring subset. The spring quality pass rate refers to the proportion of springs in the spring subset that passed the appearance defect inspection, spring hardness inspection, spring tensile inspection, and spring compression inspection out of the total number of springs in the spring subset. Specific results are shown in Table 2.

[0132] Table 2. Validation Table of Quenching Process Optimization in Example 1

[0133]

[0134] In Table 2, the spring subset numbers and quenching process parameter set numbers correspond one-to-one. As can be seen from Table 2, the quenching process parameter set with number 4 corresponds to the spring quenching performance test value with the largest value, and the spring quality test pass rate is not less than the spring quality test pass rate corresponding to the other four quenching process parameter sets. Therefore, the quenching process parameter set with number 4 is the optimal quenching process parameter set among the five quenching process parameter sets. That is, the spring quenching process optimization method proposed in this embodiment can effectively obtain the optimal quenching process parameter set.

[0135] Example 2

[0136] In Embodiment 1, the method and system of the present invention are combined to optimize the spring quenching process; in Embodiment 2 of this application, the method of the present invention will be described again, and the specific implementation process includes:

[0137] Collect multiple springs of the same type and construct the first spring detection set.

[0138] Furthermore, in this embodiment, 300 springs of the same type from factory B were collected to construct a first spring detection set; the same type means that the springs have the same material, shape, size and target quality after quenching; none of the springs in the first spring detection set have undergone quenching treatment; each spring has undergone the same manufacturing steps before quenching.

[0139] The springs in the first spring detection set are filtered to obtain the second spring detection set.

[0140] Further, the springs in the first spring inspection set are subjected to spring appearance defect inspection, and springs that pass the spring appearance defect inspection are selected to construct a second spring inspection set; the spring appearance defect inspection process includes:

[0141] Obtain the spring appearance image and input the spring appearance image into the spring appearance defect detection model;

[0142] The spring appearance defect detection model includes a spring appearance image input layer, a spring appearance feature map acquisition layer, a spring appearance feature map fusion layer, and a spring external defect detection value acquisition layer.

[0143] The spring appearance feature map acquisition layer is used to perform a convolution operation on the spring appearance image to obtain multiple appearance feature maps; the spring appearance feature map fusion layer is used to fuse the multiple appearance feature maps to obtain spring appearance features.

[0144] The appearance features of the spring include spring shape features and spring defect features; the spring shape features include the maximum diameter of the spring wire, the minimum diameter of the spring wire, the maximum outer diameter of the spring, the minimum inner diameter of the spring, and the non-support coil pitch;

[0145] The spring defect features include spring surface crack features, spring surface foreign object features, and spring surface depression features; the surface crack features include the number of cracks and the crack area of ​​each crack; the spring surface foreign object features include the number of foreign objects; the spring surface depression features include the number of depressions and the area of ​​each depression;

[0146] Obtain the standard shape features of the spring corresponding to the spring; obtain a first spring defect detection value based on the spring shape features and the standard shape features of the spring; obtain a second spring defect detection value based on the spring defect features; obtain a first spring appearance defect detection value based on the first spring defect detection value and the second spring defect detection value.

[0147] The formula for calculating the first spring defect detection value is:

[0148] ;

[0149] in, This is represented as the first spring defect detection value; This is represented by the spring shape features obtained through the spring appearance defect detection model; Represented as the standard shape characteristics of a spring; It is represented as a nonlinear function; the larger the first spring defect detection value, the greater the difference between the spring shape characteristics and the standard spring shape characteristics;

[0150] The formula for calculating the second spring defect detection value is:

[0151] ;

[0152] in, This is represented as the second spring defect detection value; This is expressed as the number of cracks; Let represent the crack area of ​​the i-th crack. This is expressed as the number of foreign objects on the surface; This is expressed as the number of depressions; Let be the area of ​​the j-th depression; , and These are respectively represented as crack coefficient, foreign matter coefficient, and indentation coefficient; It is represented as an exponential function with the natural constant as the base; , and The default values ​​are all The specific details can be modified according to the actual application.

[0153] The formula for calculating the first spring's appearance defect detection value is:

[0154] ;

[0155] in, This is represented as the detection value of the first spring's appearance defect; and These are respectively represented as the defect coefficient of the first spring and the defect coefficient of the second spring; and The default value is 0.5, but it can be changed according to the actual application.

[0156] When the first spring appearance defect detection value of the spring is greater than the set defect detection threshold, the spring appearance defect detection is determined to be unqualified; otherwise, the spring appearance defect detection is determined to be qualified; the defect detection threshold is obtained based on experience.

[0157] Obtain the set of M quenching process parameters, and divide the second spring detection set into M spring detection subsets.

[0158] Furthermore, the quenching process parameter set includes: each quenching process parameter set includes N quenching process parameters; the quenching process parameter values ​​of different quenching process parameter sets are not completely the same; each quenching process parameter set includes heating temperature, holding time, cooling medium type, cooling rate, and cooling time, etc.; in this embodiment, six quenching process parameter sets were collected, and the second spring detection set was divided into six spring detection subsets. Some quenching process parameters of each quenching process parameter set are shown in Table 3.

[0159] Table 3. Partial Quenching Process Parameter Values ​​for Example 2

[0160]

[0161] The springs in each spring detection subset are quenched according to the quenching process parameters of the corresponding quenching process parameter set; the quenched springs are then tested for spring quenching performance, and the spring quenching performance test value corresponding to each spring detection subset is obtained.

[0162] Furthermore, the quenched springs are subjected to spring quenching performance testing, and the spring quenching performance test values ​​corresponding to each spring test subset are obtained. Specific steps include:

[0163] The surface of the quenched spring is cleaned; the cleaned spring is subjected to spring appearance defect detection and spring hardness detection respectively, and the comprehensive value of spring appearance defect detection and comprehensive value of spring hardness detection of the spring detection subset are obtained.

[0164] Each spring detection subset is further divided into a spring tension detection subset and a spring compression detection subset;

[0165] Perform spring tension testing on the springs in each spring tension testing subset to obtain the comprehensive spring tension testing value for the corresponding spring testing subset; perform spring compression testing on the springs in each spring compression testing subset to obtain the comprehensive spring compression testing value for the corresponding spring testing subset.

[0166] The spring quenching performance test value of each spring test subset is obtained based on the four comprehensive test values ​​corresponding to each spring test subset. The calculation formula is as follows:

[0167] ;

[0168] in, This represents the spring quenching performance test value of the kth spring test subset; This represents the comprehensive value of spring appearance defect detection for the kth spring detection subset. This represents the comprehensive value of the spring stiffness test for the kth spring test subset. This represents the comprehensive value of the spring tension test for the kth spring test subset. This is represented as the comprehensive value of spring compression detection for the kth subset of spring detection. , , and These are respectively represented as the spring appearance defect coefficient, spring hardness coefficient, spring tensile coefficient, and spring compression coefficient; , , and The default value is 0.25, but it can be changed according to the actual application.

[0169] Furthermore, the process of obtaining the comprehensive value of the spring appearance defect detection includes:

[0170] Each quenched spring is subjected to spring appearance defect detection to obtain a first spring appearance defect detection value for each spring; the first spring appearance defect detection values ​​of the springs in each spring detection subset are averaged to obtain a comprehensive spring appearance defect detection value corresponding to the spring detection subset.

[0171] Furthermore, the process of obtaining the comprehensive value of the spring stiffness test includes:

[0172] H hardness testing points are selected on each quenched spring; the hardness value of each hardness testing point is obtained; the hardness values ​​of the hardness testing points on each spring are averaged to obtain the average hardness of each spring.

[0173] The spring hardness test is judged to be qualified based on the average hardness value, and the hardness test qualification rate of each spring test subset is obtained; the hardness test qualification rate is the comprehensive value of the spring hardness test corresponding to the spring test subset.

[0174] Furthermore, the spring tension detection process includes:

[0175] Before performing spring tension testing, obtain the first spring shape characteristics of each spring;

[0176] A set tension is applied to the spring, and the force is applied. After a certain time, the tension is released; The time can be 60 seconds or 90 seconds, etc.; the extension length of the spring is obtained when the tension is released; when the tension is released... After a certain time, the second spring shape characteristics and spring defect characteristics of each spring are obtained; The time can be 30 seconds or 40 seconds, etc.; the spring tension detection value of the spring is obtained based on the shape characteristics of the first spring, the shape characteristics of the second spring, the stretch length, and the spring defect characteristics;

[0177] The formula for calculating the spring tension test value is as follows:

[0178] ;

[0179] in, This represents the spring tension detection value of the p-th spring in the spring tension detection subset corresponding to the k-th spring tension detection subset; This represents the extension length of the p-th spring; This is represented as the set standard stretch length; This is represented as the spring stretching shape defect value of the spring obtained based on the first spring shape feature and the second spring shape feature; This is represented as the second spring defect detection value of the p-th spring obtained based on the spring defect characteristics; , and These are respectively represented as the tensile length factor, tensile shape factor, and tensile defect factor;

[0180] The formula for calculating the spring tension shape defect value is as follows:

[0181] ;

[0182] in, This represents the first spring shape feature of the p-th spring in the spring tension detection subset corresponding to the k-th spring tension detection subset; The shape feature of the second spring, represented by the p-th spring; It is represented as the same nonlinear function as the calculation process for the first spring defect detection value;

[0183] The average spring tension test values ​​of each spring in the spring tension test subset are used to obtain the comprehensive spring tension test value of the corresponding spring tension test subset.

[0184] Furthermore, the spring compression detection process includes:

[0185] Obtain the third spring shape characteristics of each spring before performing spring compression testing;

[0186] Apply a set pressure to the spring, and apply force. After a certain time, the pressure is removed; The time can be 60 seconds or 90 seconds, etc.; the compression length of the spring is obtained when the pressure is released;

[0187] In removing pressure After a certain time, the fourth spring shape characteristics and spring defect characteristics of each spring are obtained; The time can be 30 seconds or 40 seconds, etc.; the spring compression detection value of the spring is obtained based on the shape characteristics of the third spring, the shape characteristics of the fourth spring, the compression length, and the spring defect characteristics;

[0188] The formula for calculating the spring tension test value is as follows:

[0189] ;

[0190] in, This represents the spring tension detection value of the qth spring in the kth spring compression detection subset; This is expressed as the compression length of the q-th spring; This is represented as the set standard compression length; This is represented as the spring compression shape defect value of the spring obtained based on the third spring shape feature and the fourth spring shape feature; This is represented as the second spring defect detection value of the qth spring obtained based on the spring defect characteristics; , and These are respectively represented as the compression length factor, compression shape factor, and compression defect factor;

[0191] The formula for calculating the spring compression shape defect value is as follows:

[0192] ;

[0193] in, The third spring shape feature is represented by the spring q in the k-th spring compression detection subset; The fourth spring shape feature is represented by the q-th spring in the k-th spring compression detection subset; It is represented as the same nonlinear function as the calculation process for the first spring defect detection value;

[0194] The average spring compression detection values ​​of each spring in the spring compression detection subset are used to obtain the comprehensive spring compression detection value of the spring detection subset corresponding to the spring compression detection subset.

[0195] Based on the spring quenching performance test values, the optimal set of quenching process parameters is obtained.

[0196] Furthermore, the set of quenching process parameters corresponding to the spring detection subset with the largest spring quenching performance test value is selected as the optimal quenching process parameter set; in this embodiment, the quenching process parameter set with the sequence number 5 is calculated to have the largest spring quenching performance test value, that is, the quenching process parameter set with the sequence number 5 is selected as the optimal quenching process parameter set.

[0197] To verify the effectiveness of the spring quenching process optimization method used in this embodiment, 1086 springs of the same type that passed the appearance defect inspection were collected and randomly grouped to generate six spring subsets, each containing 181 springs. The 181 springs were quenched according to the quenching process parameters of the six quenching process parameter sets. The quenched springs were then subjected to quality inspection, and the spring quenching performance test values ​​and the corresponding spring quality pass rates for each spring subset were obtained. The specific results are shown in Table 4.

[0198] Table 4. Validation Table of Quenching Process Optimization in Example 2

[0199]

[0200] In Table 4, the spring subset numbers and quenching process parameter set numbers correspond one-to-one. As can be seen from Table 4, the quenching process parameter set with serial number 5 corresponds to the spring quenching performance test value with the largest value, and the spring quality test pass rate is not less than the spring quality test pass rate corresponding to the other five quenching process parameter sets. Therefore, the quenching process parameter set with serial number 5 is the optimal quenching process parameter set among the six quenching process parameter sets. That is, the spring quenching process optimization method proposed in this embodiment can effectively obtain the optimal quenching process parameter set.

[0201] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing the intelligent digital quenching process of springs, characterized in that, include: Collect multiple springs of the same type and construct the first spring detection set; The springs in the first spring detection set are filtered to obtain the second spring detection set; The springs in the first spring detection set are subjected to spring appearance defect detection, and the springs that pass the spring appearance defect detection are selected to construct the second spring detection set; the spring defect detection value is obtained based on the spring shape characteristics. The defect detection value of the second spring is obtained based on the characteristics of the spring defect. Based on the first spring defect detection value and the second spring defect detection value, the first spring appearance defect detection value is obtained; when the first spring appearance defect detection value of the spring is greater than the set defect detection threshold, the spring appearance defect detection is determined to be unqualified; otherwise, the spring appearance defect detection is determined to be qualified. Obtain the M sets of quenching process parameters, and divide the second spring detection set into M sets of spring detection subsets; The springs in each spring detection subset are quenched according to the quenching process parameters of the corresponding quenching process parameter set. The quenching performance of the springs after quenching is tested, and the quenching performance test value corresponding to each spring test subset is obtained. The specific steps include: cleaning the surface of the quenched springs; performing spring appearance defect detection and spring hardness detection on the cleaned springs respectively, and obtaining the comprehensive value of spring appearance defect detection and the comprehensive value of spring hardness detection for the spring test subset. Each spring detection subset is further divided into a spring tension detection subset and a spring compression detection subset; Perform spring tension testing on the springs in each spring tension testing subset to obtain the comprehensive spring tension testing value for the corresponding spring testing subset; perform spring compression testing on the springs in each spring compression testing subset to obtain the comprehensive spring compression testing value for the corresponding spring testing subset; obtain the first spring shape characteristics of each spring before performing the spring tension testing; apply a set tension force to the spring, and apply the force... After a certain time, the tension is released; For 60 seconds and 90 seconds; obtain the extension length of the spring when the tension is released; when the tension is released... After a certain time, the second spring shape characteristics and second spring defect characteristics of each spring are obtained; For 30 seconds and 40 seconds; The spring quenching performance test value of each spring test subset is obtained based on the four comprehensive test values ​​corresponding to each spring test subset; Based on the spring quenching performance test values, the optimal set of quenching process parameters is obtained.

2. The intelligent digital quenching process optimization method for springs according to claim 1, characterized in that, The step of acquiring multiple springs of the same type and constructing a first spring detection set includes: none of the springs in the first spring detection set have undergone quenching treatment; and each spring has undergone the same manufacturing steps before quenching.

3. The intelligent digital quenching process optimization method for springs according to claim 1, characterized in that, The quenching process parameter set includes: each quenching process parameter set includes N quenching process parameters; the quenching process parameter values ​​of different quenching process parameter sets are not completely the same.

4. The intelligent digital quenching process optimization method for springs according to claim 1, characterized in that, The spring tension detection process includes: Before performing spring tension testing, obtain the first spring shape characteristics of each spring; A set tension is applied to the spring, and the force is applied. After a certain time, the tension is released; the extension length of the spring is obtained when the tension is released; after the tension is released... After a certain time, the second spring shape feature and the second spring defect feature of each spring are obtained; the spring tension detection value of the spring is obtained based on the first spring shape feature, the second spring shape feature, the stretch length and the second spring defect feature; the comprehensive spring tension detection value of the spring detection subset corresponding to the spring tension detection subset is obtained based on the spring tension detection value of each spring in the spring tension detection subset.

5. The intelligent digital quenching process optimization method for springs according to claim 1, characterized in that, The spring compression detection process includes: Obtain the third spring shape characteristics of each spring before performing spring compression testing; Apply a set pressure to the spring, and apply force. After a certain time, the pressure is released; the compression length of the spring is obtained when the pressure is released; and the pressure is released at the following time... After a certain time, the fourth spring shape feature and the fourth spring defect feature of each spring are obtained; the spring compression detection value of the spring is obtained based on the third spring shape feature, the fourth spring shape feature, the compression length and the fourth spring defect feature; the comprehensive spring compression detection value of the spring detection subset corresponding to the spring compression detection subset is obtained based on the spring compression detection value of each spring in the spring compression detection subset.

6. The intelligent digital quenching process optimization method for springs according to claim 1, characterized in that, The process of obtaining the optimal set of quenching process parameters based on the spring quenching performance test value includes: The set of quenching process parameters corresponding to the spring detection subset with the largest spring quenching performance test value is selected as the optimal quenching process parameter set.

7. A smart digital quenching process optimization system for springs, the system being used to execute the smart digital quenching process optimization method for springs according to claim 1, characterized in that, include: First Spring Detection Set Acquisition Unit: Used to acquire multiple springs of the same type and construct a first spring detection set; Second spring detection set acquisition unit: used to filter the springs in the first spring detection set to acquire the second spring detection set; Quenching Spring Detection Subset Division Unit: Used to obtain M sets of quenching process parameters and divide the second spring detection set into M sets of spring detection subsets; Quenching unit: used to quench the springs of each spring detection subset according to the quenching process parameters of the corresponding quenching process parameter set; Quenching performance testing unit: used to test the quenching performance of the springs after quenching, and to obtain the spring quenching performance test value corresponding to each spring test subset; the spring quenching performance test includes spring appearance defect detection, spring hardness detection, spring tensile testing and spring compression testing; Optimal quenching process parameter set acquisition unit: used to acquire the optimal quenching process parameter set based on the spring quenching performance test value.

Images