Preparation method of high-precision high-temperature alloy spring

Abstract

The invention relates to the technical field of spring preparation, and particularly discloses a preparation method of a high-precision high-temperature alloy spring, which comprises the following steps: selecting a specific material, processing the spring by three times of heat treatment in combination with corresponding heat preservation and cooling modes, drawing an alloy blank subjected to heat treatment to a wire diameter of 0.1-3mm, carrying out normal-temperature primary drawing, carrying out low-temperature precise drawing, and carrying out high-precision high-temperature alloy spring preparation. Then a numerical control spring forming machine is used for winding forming, the drawing force and the curling speed are controlled to guarantee the size standard, surface polishing and chromium plating protection are conducted, the spring length difference and the related ratio are calculated through spring length detection, the spring precision Jd is obtained, the spring precision Jd is compared with a threshold value Jdy, and an abnormal analysis signal is generated; a temperature-spring precision Jd curve is drawn based on an abnormal analysis signal, an abnormal degree coefficient Cx is calculated, finally, a model is established by using a clustering algorithm, the optimal temperature is calculated, and the precision and quality of the spring are improved through monitoring and optimization of the spring.

Description

Technical Field

[0001] The invention relates to the technical field of spring preparation, and in particular to a method for preparing a high-precision high-temperature alloy spring. Background Art

[0002] In modern industry, springs, as an important elastic element, are widely used in many fields. With the development of industry, the performance requirements for springs are becoming increasingly stringent, especially for springs in high-temperature working environments. They not only need to have good high-temperature resistance, but also must maintain high precision to ensure the stable operation of the equipment.

[0003] A Chinese patent application with the publication number CN105921649A discloses a method for preparing a high-precision high-temperature alloy spring, comprising: (1) first making a variable diameter mandrel whose working circle diameter is smaller than the supporting circle diameter on both sides thereof, and setting a wire stop pin on the outer side of the supporting circle on one side; (2) clamping the variable diameter mandrel on a lathe, clamping the high-temperature alloy steel wire with a wire clamping plate, and feeding the steel wire head into the wire stop pin; (3) cutting off the excess steel wire to ensure the total number of circles and the effective number of circles, and then loading it into a heat treatment furnace for strengthening treatment; (4) grinding the two end faces of the spring on a grinding wheel to ensure that 3 / 4 of the spring end circle is ground. After each grinding, cool it in water. After grinding, check whether the spring height reaches the specified height; (5) making a pressing sleeve that matches the pressing shaft, and the bottom of the pressing shaft is provided with a disc base integrated with it; (6) clamping the pressing sleeve on a lathe, and installing the ground spring on the pressing shaft. Move the lathe tailstock so that it is against the center of the disc base. After the spring is compressed, stop moving the tailstock and hold it for more than 10 seconds before releasing it. Repeat the operation until the force value of the spring meets the use requirements.

[0004] In the prior art, the traditional spring preparation method exposes many limitations in the production of high-precision high-temperature alloy castings. On the one hand, the material selection is not optimized enough to meet the requirements of high temperature resistance and high precision. On the other hand, there is a lack of effective means to control the spring precision. During the heat treatment process, the spring precision is tested and evaluated, and the optimal temperature zone is selected to adjust the spring precision, so as to improve the reliability and durability of the spring in practical applications.

[0005] To this end, the present invention provides a method for preparing a high-precision high-temperature alloy spring. Summary of the invention

[0006] The object of the present invention is to provide a method for preparing a high-precision high-temperature alloy spring to solve the above-mentioned problems.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] Step 1, obtain the length data of the spring, calculate the spring length and the spring length target value to obtain the spring length difference, mark the spring whose spring length difference does not meet the threshold as an abnormal spring, calculate the mean value of the length difference of the abnormal spring and calculate it with the spring length difference threshold to obtain the abnormal length ratio Cd, calculate the proportion of abnormal springs, obtain the abnormal spring quantity ratio SI, calculate the spring accuracy Jd based on the abnormal spring quantity ratio SI and the abnormal length ratio Cd, the spring accuracy Jd ≥ the spring accuracy threshold Jdy, and generate an abnormal analysis signal;

[0009] Step 2: Based on the abnormal analysis signal, collect data from multiple production batches, obtain the change of spring accuracy Jd under different temperature conditions, draw the temperature-spring accuracy Jd curve, draw the reference line with the spring accuracy threshold Jdy as the reference value, calculate the abnormal curve segment where the spring accuracy Jd exceeds the spring accuracy threshold Jdy, obtain the abnormal area enclosed by the abnormal curve segment and the reference line, obtain the temperature block length value corresponding to the abnormal curve segment, calculate the abnormal area ratio mean value Yc, the temperature block length ratio Wd, and calculate the abnormal degree coefficient Cx, the abnormal degree coefficient Cx ≥ the abnormal degree coefficient threshold Cxz, and generate a degree analysis signal;

[0010] Step 3: Based on the degree analysis signal, obtain the spring accuracy Jd corresponding to different temperatures, use the clustering algorithm to establish the temperature-spring accuracy Jd clustering algorithm model, and calculate the optimal temperature.

[0011] As a further technical solution of the present invention: the spring accuracy Jd is obtained as follows:

[0012] Calculate the spring accuracy Jd based on the number ratio SI and abnormal length ratio Cd of abnormal springs;

[0013] By formula: Get the spring accuracy Jd, where a and b are preset proportional coefficients;

[0014] The abnormal analysis signal is generated in the following manner:

[0015] Compare the spring accuracy Jd with the spring accuracy threshold Jdy;

[0016] If the spring accuracy Jd ≥ the spring accuracy threshold Jdy, an abnormal analysis signal is generated.

[0017] As a further technical solution of the present invention: the method for obtaining the number ratio SI of the abnormal springs is: obtaining the length data of the spring sample, performing a subtraction process between the length of the spring sample and the target value of the spring length, and obtaining the spring length difference;

[0018] comparing the spring length difference to a spring length difference threshold;

[0019] It should be noted that the spring length difference threshold is set by professionals in this field based on experience, and the spring length target value is the expected value of spring production;

[0020] The springs whose spring length difference does not meet the spring length difference threshold range are marked as abnormal springs;

[0021] The number of abnormal springs is obtained, and the ratio of the number of abnormal springs to the number of all tested springs is processed to obtain the abnormal spring number ratio, and the abnormal spring number ratio is marked as S l.

[0022] As a further technical solution of the present invention: the abnormal length ratio Cd is obtained as follows:

[0023] The length difference of the abnormal spring is subtracted from the endpoint value of the adjacent spring length difference threshold range and the absolute value is taken to obtain the critical length difference of the spring;

[0024] Obtain the critical spring length differences of all abnormal springs, and sum the critical spring length differences of all abnormal springs to obtain a total critical spring length difference;

[0025] The total difference of the critical length of the spring is processed by ratio with the number of all abnormal springs to obtain the mean value of the critical difference of the spring length;

[0026] The average value of the critical difference of spring length is ratioed with the length value of the threshold range of the spring length difference to obtain the abnormal length ratio, which is marked as Cd.

[0027] As a further technical solution of the present invention: the abnormality coefficient Cx is obtained as follows:

[0028] Based on the mean value of the abnormal area ratio Yc and the temperature block length ratio Wd, the abnormal degree coefficient Cx is calculated;

[0029] By formula: The abnormality degree coefficient Cx is calculated.

[0030] As a further technical solution of the present invention: the method for obtaining the mean value Yc of the abnormal area ratio is:

[0031] With temperature as the X-axis and spring accuracy Jd as the Y-axis, draw a temperature-spring accuracy Jd curve in a two-dimensional rectangular coordinate system;

[0032] Using the spring accuracy threshold Jdy as a reference value, draw a horizontal reference line in the coordinate system;

[0033] Obtain the curve segment where the spring accuracy Jd exceeds the spring accuracy threshold Jdy, and obtain the abnormal curve segment;

[0034] Obtain the area enclosed by a single abnormal curve segment and the reference line to obtain the abnormal area;

[0035] Get the area enclosed by a single abnormal curve segment and the X-axis to obtain the curve area;

[0036] The abnormal area is processed by ratioing the abnormal area with the curve area to obtain the abnormal area ratio;

[0037] Obtain the abnormal area ratios of all abnormal curve segments and sum them up to obtain the total abnormal area ratio;

[0038] The total abnormal area ratio is processed by ratioing the number of abnormal curve segments to obtain the mean abnormal area ratio, and the mean abnormal area ratio is marked as Yc.

[0039] As a further technical solution of the present invention: the temperature zone length ratio Cd is obtained as follows:

[0040] Obtain the abnormal temperature section corresponding to the abnormal curve section on the X-axis, and calculate the length of the abnormal temperature section;

[0041] The abnormal temperature section length is ratioed to the total temperature section length to obtain the temperature section length ratio, which is marked as Wd.

[0042] As a further technical solution of the present invention: the degree analysis signal is generated in the following manner:

[0043] Compare the abnormality degree coefficient Cx with the abnormality degree coefficient threshold Cxz;

[0044] If the abnormality degree coefficient Cx ≥ the abnormality degree coefficient threshold Cxz, a degree analysis signal is generated.

[0045] As a further technical solution of the present invention: the optimal temperature is obtained by:

[0046] Based on the trained temperature-spring precision Jd clustering algorithm model, calculate the mean spring precision within the temperature range in each cluster;

[0047] Get the average precision of all clusters, sort the average precision from high to low, and get the highest precision;

[0048] Get the temperature range of the cluster with the highest accuracy and get the optimal temperature range;

[0049] The optimal temperature range is summed and averaged to obtain the optimal temperature.

[0050] As a further technical solution of the present invention: the temperature-spring accuracy Jd clustering algorithm model is constructed in the following manner:

[0051] SS1. Preprocess the data of the obtained temperature-spring precision Jd set H, remove outliers, and standardize the data;

[0052] SS2, select K-Means clustering algorithm, use elbow rule to determine the number K, and set K = 8 as the number of clusters;

[0053] SS3, randomly initialize three cluster centers, calculate the distance between the data point and the three cluster centers, take the data pairs as the data points in the cluster centers, and assign the data points to the cluster with the nearest cluster center;

[0054] SS4: Recalculate the center of each cluster, that is, the mean of all data points in the cluster, and repeat the SS3 step until the cluster center no longer changes or reaches the predetermined iteration of 200.

[0055] Beneficial effects of the present invention:

[0056] (1) A high-temperature resistant nickel-based alloy is selected, and aluminum, molybdenum, and niobium elements are added to improve the high-temperature resistance of the spring. In the heat treatment stage, three heat treatment processes are performed to increase the high-temperature resistance and metal fatigue resistance of the spring, so that it can maintain stable performance in a high-temperature environment. The surface of the spring is polished to make the surface smooth and flat, further improving the performance stability of the spring. The chrome plating treatment on the surface of the spring improves the oxidation and corrosion of the spring in a high-temperature environment and corrosive media, which greatly improves the reliable operation ability of the spring under load conditions;

[0057] (2) By obtaining the length data of the spring, the spring length difference is calculated with the target value of the spring length, the abnormal spring quantity ratio and abnormal length ratio are calculated, and the spring accuracy jd is calculated. By analyzing the change of spring accuracy jd at different temperatures, the temperature-spring accuracy jd curve is drawn, and the relevant parameters of the abnormal curve segment are calculated to evaluate the abnormal degree of the spring at different temperatures. The temperature-spring accuracy jd clustering algorithm model is established using the clustering algorithm. The model can calculate the mean value of the spring accuracy in each cluster, thereby determining the optimal temperature range and optimal temperature. By calculating the optimal temperature, the spring can function stably in application scenarios with extremely high precision requirements, thereby improving the operating performance and stability of the entire equipment system. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The present invention will be further described below in conjunction with the accompanying drawings.

[0059] Figure 1 This is a flow chart of a method for preparing a high-precision high-temperature alloy spring according to the present invention;

[0060] Figure 2 It is a flow chart of a spring detection method of the present invention. DETAILED DESCRIPTION

[0061] 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.

[0062] Example 1

[0063] See also Figure 1 As shown, the present invention is a method for preparing a high-precision high-temperature alloy spring.

[0064] S1. Use high temperature resistant nickel-based alloy as raw material, and add aluminum, molybdenum and niobium elements into the material;

[0065] S2, the first heat treatment: heating the alloy to 1100 ℃ -1150 ℃, and keeping it warm for 1-2 hours, and then cooling the alloy quickly; the second heat treatment: heating to 650 ℃ -700 ℃, keeping it warm for 4-8 hours, and then slowly cooling; the third heat treatment: heating to 700 ℃ -750 ℃, keeping it warm for 6-10 hours, and then cooling;

[0066] S3, processing the heat-treated alloy blank through a multi-pass drawing machine to a desired wire diameter, the wire diameter range is between 0.1 and 3 mm, the initial drawing is performed at room temperature, and the precision drawing is performed at low temperature;

[0067] S4, using a CNC spring forming machine to wind the wire into shape;

[0068] S5. Polish the surface of the spring to remove the oxides remaining during processing, and plate chrome on the surface of the spring as a high temperature resistant protective coating;

[0069] Example 2

[0070] Based on Example 1, a method for preparing a high-precision high-temperature alloy spring further includes:

[0071] Step 1: After step S5 is completed, obtain the length data of the spring, calculate the spring length and the spring length target value to obtain the spring length difference, mark the spring whose spring length difference does not meet the threshold as an abnormal spring, calculate the mean of the length difference of the abnormal spring and the spring length difference threshold to obtain the abnormal length ratio Cd, calculate the proportion of abnormal springs, obtain the abnormal spring quantity ratio SI, calculate the spring accuracy Jd based on the abnormal spring quantity ratio SI and the abnormal length ratio Cd, the spring accuracy Jd ≥ the spring accuracy threshold Jdy, and generate an abnormal analysis signal;

[0072] Select the spring prepared in step S5 as a test sample, and mark the test sample as a spring sample;

[0073] Obtain the length data of the spring sample, perform subtraction processing on the spring length of the spring sample and the spring length target value, and obtain the spring length difference;

[0074] comparing the spring length difference to a spring length difference threshold range;

[0075] It should be noted that the spring length difference threshold range is set by professionals in the field based on experience, and the spring length target value is the expected value of spring production;

[0076] The springs whose spring length difference does not meet the spring length difference threshold range are marked as abnormal springs;

[0077] Obtain the number of abnormal springs, and perform ratio processing on the number of abnormal springs and the number of all tested springs to obtain the abnormal spring number ratio, and mark the abnormal spring number ratio as S1;

[0078] The length difference of the abnormal spring is subtracted from the endpoint value of the adjacent spring length difference threshold range and the absolute value is taken to obtain the critical length difference of the spring;

[0079] Obtain the critical spring length differences of all abnormal springs, and sum the critical spring length differences of all abnormal springs to obtain a total critical spring length difference;

[0080] The total difference of the critical length of the spring is processed by ratio with the number of all abnormal springs to obtain the mean value of the critical difference of the spring length;

[0081] The spring length critical difference mean value is processed by ratio with the spring length difference threshold range length value to obtain the abnormal length ratio, and the abnormal length ratio is marked as Cd;

[0082] The length value of the spring length difference threshold range is obtained by taking the difference between the two end points of the spring length difference threshold range and taking the absolute value;

[0083] Calculate the spring accuracy Jd based on the number ratio SI and abnormal length ratio Cd of abnormal springs;

[0084] By formula: Get the spring accuracy Jd, where a and b are preset proportional coefficients;

[0085] Compare the spring accuracy Jd with the spring accuracy threshold Jdy;

[0086] If the spring accuracy Jd ≥ the spring accuracy threshold Jdy, it indicates that the spring accuracy difference is large, and an abnormal analysis signal is generated;

[0087] If the spring accuracy Jd is less than the spring accuracy threshold Jdy, it indicates that the spring accuracy difference is within the expected range, and it is still necessary to continue monitoring the abnormal coefficient changes of the spring;

[0088] Step 2: Based on the abnormal analysis signal, collect data from multiple production batches, obtain the change of spring accuracy Jd under different temperature conditions, draw the temperature-spring accuracy Jd curve, draw the reference line with the spring accuracy threshold Jdy as the reference value, calculate the abnormal curve segment where the spring accuracy Jd exceeds the spring accuracy threshold Jdy, obtain the abnormal area enclosed by the abnormal curve segment and the reference line, obtain the temperature block length value corresponding to the abnormal curve segment, calculate the abnormal area ratio mean value Yc, the temperature block length ratio Wd, and calculate the abnormal degree coefficient Cx, the abnormal degree coefficient Cx ≥ the abnormal degree coefficient threshold Cxz, and generate a degree analysis signal;

[0089] In the third heat treatment stage, data from multiple production batches are collected to analyze the changes in spring accuracy Jd under different temperature conditions;

[0090] With temperature as the X-axis and spring accuracy Jd as the Y-axis, draw a temperature-spring accuracy Jd curve in a two-dimensional rectangular coordinate system;

[0091] Using the spring accuracy threshold Jdy as a reference value, draw a horizontal reference line in the coordinate system;

[0092] Obtain the curve segment where the spring accuracy Jd exceeds the spring accuracy threshold Jdy, and obtain the abnormal curve segment;

[0093] Obtain the area enclosed by a single abnormal curve segment and the reference line to obtain the abnormal area;

[0094] Get the area enclosed by a single abnormal curve segment and the X-axis to obtain the curve area;

[0095] The abnormal area is processed by ratioing the abnormal area with the curve area to obtain the abnormal area ratio;

[0096] Obtain the abnormal area ratios of all abnormal curve segments and sum them up to obtain the total abnormal area ratio;

[0097] The total ratio of abnormal areas is processed by ratio with the number of abnormal curve segments to obtain the mean ratio of abnormal areas, and the mean ratio of abnormal areas is marked as Yc;

[0098] Obtain the abnormal temperature section corresponding to the abnormal curve section on the X-axis, and calculate the length of the abnormal temperature section;

[0099] It should be noted that the abnormal temperature section represents the interval formed between the corresponding temperature points on the X-axis at the two end points of the abnormal curve segment;

[0100] The abnormal temperature section length is processed by ratioing the length of all temperature sections to obtain the temperature block length ratio, which is marked as Wd;

[0101] It should be noted that the length of the entire temperature section is the interval formed between the temperature points corresponding to the two end points of the temperature-spring accuracy Jd curve on the X-axis;

[0102] Based on the mean value of the abnormal area ratio Yc and the temperature block length ratio Wd, the abnormal degree coefficient Cx is calculated;

[0103] By formula: The abnormality coefficient Cx is calculated, where d and f are preset proportional coefficients, ln(f*YC+d*Wd) is the logarithmic function of the base e, and e f+d is an exponential function;

[0104] Compare the abnormality degree coefficient Cx with the abnormality degree coefficient threshold Cxz;

[0105] If the abnormality degree coefficient Cx ≥ abnormality degree coefficient threshold Cxz, it indicates that the abnormality degree coefficient Cx of the spring exceeds the expected value, and a degree analysis signal is generated;

[0106] If the abnormality coefficient Cx is less than the abnormality coefficient threshold Cxz, it indicates that the abnormality coefficient Cx of the spring is within the expected value, but it is still necessary to continue to monitor the change of the abnormality coefficient threshold Cxz; the technical solution described in this embodiment is: obtain the length data of the spring, calculate the spring length and the spring length target value to obtain the spring length difference, mark the spring whose spring length difference does not meet the threshold as an abnormal spring, calculate the mean value of the length difference of the abnormal spring and process it with the spring length difference threshold ratio to obtain the abnormal length ratio Cd, calculate the proportion of abnormal springs, obtain the abnormal spring quantity ratio SI, and calculate the spring accuracy Jd based on the abnormal spring quantity ratio SI and the abnormal length ratio Cd. Jd≥spring precision threshold Jdy, generate an abnormal analysis signal, collect data from multiple production batches based on the abnormal analysis signal, obtain the change of spring precision Jd under different temperature conditions, draw the temperature-spring precision Jd curve, draw the reference line with the spring precision threshold Jdy as the reference value, calculate the abnormal curve segment where the spring precision Jd exceeds the spring precision threshold Jdy, obtain the abnormal area enclosed by the abnormal curve segment and the reference line, obtain the temperature block length value corresponding to the abnormal curve segment, calculate the abnormal area ratio mean Yc and the temperature block length ratio Wd, and calculate the abnormal degree coefficient Cx. The abnormal degree coefficient Cx≥abnormal degree coefficient threshold Cxz, and generate a degree analysis signal.

[0107] Example 3

[0108] Based on Example 2, the method for preparing a high-precision high-temperature alloy spring further includes:

[0109] Step 3: Based on the degree analysis signal, obtain the spring accuracy Jd corresponding to different temperatures, use the clustering algorithm to establish the temperature-spring accuracy Jd clustering algorithm model, and calculate the optimal temperature;

[0110] Get the spring accuracy Jd corresponding to different temperatures, integrate the temperature and the corresponding spring accuracy Jd into a data pair, and integrate all data pairs into a spring accuracy set H;

[0111] Exemplarily, the temperature-spring accuracy Jd data pairs are (700°C, 80%), (750°C, 85%)...; the spring accuracy set H = {(700°C, 80%), (750°C, 85%)...};

[0112] Based on the clustering algorithm, a spring temperature-spring accuracy Jd clustering algorithm model is established;

[0113] The specific temperature-spring accuracy Jd clustering algorithm model is constructed as follows;

[0114] SS1. Preprocess the data of the obtained temperature-spring precision Jd set H, remove outliers, and standardize the data;

[0115] SS2, select K-Means clustering algorithm, use elbow rule to determine the number K, and set K = 8 as the number of clusters;

[0116] It should be noted that the elbow rule is a technique for determining the number of data clusters. In the K-Means clustering algorithm, the optimal K value is determined by analyzing the relationship between the clustering error and the number of clusters K value;

[0117] SS3, randomly initialize three cluster centers, calculate the distance between the data point and the three cluster centers, take the data pairs as the data points in the cluster centers, and assign the data points to the cluster with the nearest cluster center;

[0118] SS4, recalculate the center of each cluster, that is, the mean of all data points in the cluster, and repeat the SS3 step until the cluster center no longer changes or reaches the predetermined iteration of 200 times;

[0119] Based on the trained temperature-spring precision Jd clustering algorithm model, the mean spring precision within the temperature range in each cluster is calculated;

[0120] It should be noted that the trained temperature-spring precision Jd clustering algorithm model takes the temperature range covered by the cluster as a limiting condition within the cluster, and the average precision calculated is the representative data feature within the cluster;

[0121] Get the average precision of all clusters, sort the average precision from high to low, and get the highest precision;

[0122] Get the temperature range of the cluster with the highest accuracy and get the optimal temperature range;

[0123] The optimal temperature range is summed and averaged to obtain the optimal temperature;

[0124] For example, cluster 1: temperature range [700°C, 710°C], average accuracy 85%, cluster 2: temperature range [711°C, 715°C], average accuracy 95%, cluster 3: temperature range [716, 720], average accuracy 90%, the optimal temperature range is [711°C, 715°C], and the optimal temperature is 713°C;

[0125] The technical solution of this embodiment is: based on the degree analysis signal, the spring accuracy Jd corresponding to different temperatures is obtained, a clustering algorithm model of temperature-spring accuracy Jd is established using a clustering algorithm, and the optimal temperature is calculated.

[0126] 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 preparing a high-precision high-temperature alloy spring, characterized in that: Step 1: Obtain the length data of the spring, calculate the spring length and the spring length target value to obtain the spring length difference, mark the spring whose spring length difference does not meet the threshold as an abnormal spring, calculate the mean of the length difference of the abnormal spring and the spring length difference threshold to obtain the abnormal length ratio Cd, calculate the proportion of abnormal springs, and obtain the abnormal spring quantity ratio SI, calculate the spring accuracy Jd based on the abnormal spring quantity ratio SI and the abnormal length ratio Cd, the spring accuracy Jd ≥ the spring accuracy threshold Jdy, and generate an abnormal analysis signal; Step 2: Based on the abnormal analysis signal, collect data from multiple production batches, obtain the change of spring accuracy Jd under different temperature conditions, draw the temperature-spring accuracy Jd curve, draw the reference line with the spring accuracy threshold Jdy as the reference value, calculate the abnormal curve segment where the spring accuracy Jd exceeds the spring accuracy threshold Jdy, obtain the abnormal area enclosed by the abnormal curve segment and the reference line, obtain the temperature block length value corresponding to the abnormal curve segment, calculate the abnormal area ratio mean value Yc, the temperature block length ratio Wd, and calculate the abnormal degree coefficient Cx, the abnormal degree coefficient Cx ≥ the abnormal degree coefficient threshold Cxz, and generate a degree analysis signal; Step 3: Based on the degree analysis signal, obtain the spring accuracy Jd corresponding to different temperatures, use the clustering algorithm to establish the temperature-spring accuracy Jd clustering algorithm model, and calculate the optimal temperature.

2. The method for preparing a high-precision high-temperature alloy spring according to claim 1, characterized in that: The spring accuracy Jd is obtained as follows: Calculate the spring accuracy Jd based on the number ratio SI and the abnormal length ratio Cd of the abnormal springs; By formula: Get the spring accuracy Jd, where a and b are preset proportional coefficients; The abnormal analysis signal is generated in the following manner: Compare the spring accuracy Jd with the spring accuracy threshold Jdy; If the spring accuracy Jd ≥ the spring accuracy threshold Jdy, an abnormal analysis signal is generated.

3. The method for preparing a high-precision high-temperature alloy spring according to claim 2, characterized in that: The number ratio SI of the abnormal springs is obtained as follows: Obtain the length data of the spring sample, perform subtraction processing on the spring length of the spring sample and the spring length target value, and obtain the spring length difference; comparing the spring length difference to a spring length difference threshold; The springs whose spring length difference does not meet the spring length difference threshold range are marked as abnormal springs; The number of abnormal springs is obtained, and the ratio of the number of abnormal springs to the number of all tested springs is processed to obtain the abnormal spring number ratio, and the abnormal spring number ratio is marked as S l.

4. The method for preparing a high-precision high-temperature alloy spring according to claim 2, characterized in that: The abnormal length ratio Cd is obtained as follows: The length difference of the abnormal spring is subtracted from the endpoint value of the adjacent spring length difference threshold range and the absolute value is taken to obtain the critical length difference of the spring; Obtain the critical spring length differences of all abnormal springs, and sum the critical spring length differences of all abnormal springs to obtain a total critical spring length difference; The total difference of the critical length of the spring is processed by ratio with the number of all abnormal springs to obtain the mean value of the critical difference of the spring length; The average value of the critical difference of spring length is ratioed with the length value of the threshold range of the spring length difference to obtain the abnormal length ratio, which is marked as Cd.

5. The method for preparing a high-precision high-temperature alloy spring according to claim 1, characterized in that: The abnormality coefficient Cx is obtained as follows: Based on the mean value of the abnormal area ratio Yc and the temperature block length ratio Wd, the abnormal degree coefficient Cx is calculated; By formula: The abnormality degree coefficient Cx is calculated.

6. The method for preparing a high-precision high-temperature alloy spring according to claim 5, characterized in that: The abnormal area ratio mean value Yc is obtained as follows: With temperature as the X-axis and spring accuracy Jd as the Y-axis, draw a temperature-spring accuracy Jd curve in a two-dimensional rectangular coordinate system; Using the spring accuracy threshold Jdy as a reference value, draw a horizontal reference line in the coordinate system; Obtain the curve segment where the spring accuracy Jd exceeds the spring accuracy threshold Jdy, and obtain the abnormal curve segment; Obtain the area enclosed by a single abnormal curve segment and the reference line to obtain the abnormal area; Get the area enclosed by a single abnormal curve segment and the X-axis to obtain the curve area; The abnormal area is processed by ratioing the abnormal area with the curve area to obtain the abnormal area ratio; Obtain the abnormal area ratios of all abnormal curve segments and sum them up to obtain the total abnormal area ratio; The total abnormal area ratio is processed by ratioing the number of abnormal curve segments to obtain the mean abnormal area ratio, and the mean abnormal area ratio is marked as Yc.

7. The method for preparing a high-precision high-temperature alloy spring according to claim 5, characterized in that: The temperature zone length ratio Cd is obtained as follows: Obtain the abnormal temperature section corresponding to the abnormal curve section on the X-axis, and calculate the length of the abnormal temperature section; The abnormal temperature section length is ratioed to the total temperature section length to obtain the temperature section length ratio, which is marked as Wd.

8. The method for preparing a high-precision high-temperature alloy spring according to claim 1, characterized in that: The degree analysis signal is generated in the following manner: Compare the abnormality degree coefficient Cx with the abnormality degree coefficient threshold Cxz; If the abnormality degree coefficient Cx ≥ the abnormality degree coefficient threshold Cxz, a degree analysis signal is generated.

9. The method for preparing a high-precision high-temperature alloy spring according to claim 8, characterized in that: The optimal temperature is obtained as follows: Based on the trained temperature-spring precision Jd clustering algorithm model, calculate the mean spring precision within the temperature range in each cluster; Get the average precision of all clusters, sort the average precision from high to low, and get the highest precision; Get the temperature range of the cluster with the highest accuracy and get the optimal temperature range; The optimal temperature range is summed and averaged to obtain the optimal temperature.

10. The method for preparing a high-precision high-temperature alloy spring according to claim 9, characterized in that: The temperature-spring precision Jd clustering algorithm model is constructed as follows: SS1. Preprocess the data of the obtained temperature-spring precision Jd set H, remove outliers, and standardize the data; SS2, select K-Means clustering algorithm and use elbow rule to determine the number K; SS3, randomly initialize three cluster centers, calculate the distance between the data point and the three cluster centers, take the data pairs as the data points in the cluster centers, and assign the data points to the cluster with the nearest cluster center; SS4: Recalculate the center of each cluster, that is, the mean of all data points in the cluster, and repeat SS3 until the cluster center no longer changes or the predetermined number of iterations is reached.

Images