Ultrahigh-strength spring steel quenching and tempering treatment process

Abstract

The invention relates to the technical field of spring steel treatment, and particularly discloses an ultrahigh-strength spring steel quenching and tempering treatment process, which is characterized in that a temperature monitoring system is arranged for monitoring a temperature rise stage in spring steel quenching and tempering treatment; the method specifically comprises the following steps that firstly, real-time temperature data of the temperature rising stage in spring steel quenching and tempering treatment are obtained, and analysis and judgment are conducted; according to the method, the temperature data in the temperature rising stage are obtained in real time, the deviation between the temperature data and the standard temperature is calculated, namely the temperature deviation value, the temperature deviation value is compared with the first temperature deviation threshold value and the second temperature deviation threshold value, and a temperature normal signal, a to-be-analyzed signal or a temperature abnormal signal is recognized and generated; therefore, abnormal conditions in the heating process can be found in time and processed, and meanwhile, the to-be-analyzed signals can be further processed and analyzed, so that parameters in the heating process are optimized, and the processing effect is improved.

Description

Technical Field

[0001] The invention relates to the technical field of spring steel processing, and in particular to a quenching and tempering process for ultra-high strength spring steel. Background Art

[0002] Spring steel refers to steel specially used for making springs and elastic elements. Springs use elastic deformation to absorb energy to alleviate vibration and impact, or rely on elastic energy storage to drive. Springs are used under impact, vibration or long-term alternating stress, so spring steel is required to have high elastic limit, high yield strength, high fatigue strength, and sufficient plasticity and toughness to avoid brittle fracture during impact.

[0003] The quenching process of spring steel is: heat the spring steel to the quenching temperature, then keep it warm, and finally oil quench. The tempering process of spring steel is: heat the spring steel to the tempering temperature, then keep it warm, and finally cool it to room temperature.

[0004] However, sufficient time is required for the heating stage of quenching and tempering spring steel. Too long or too short a time will affect the performance of the spring steel. Although the heating situation can be understood in real time during the quenching and tempering process of spring steel, there is a lack of in-depth analysis of the real-time temperature and the process of adjusting the heating process parameters based on the analysis, which in turn affects the stability of product quality.

[0005] In view of this, we proposed a quenching and tempering process for ultra-high strength spring steel. Summary of the invention

[0006] The object of the present invention is to provide a quenching and tempering process for ultra-high strength spring steel to solve the technical problems in the above background.

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

[0008] The present invention provides a quenching and tempering process for ultra-high strength spring steel, wherein a temperature monitoring system is provided to monitor the temperature rise stage during the quenching and tempering process for the spring steel, and specifically comprises the following steps:

[0009] Step 1: Obtain real-time temperature data of the heating stage during the quenching and tempering treatment of spring steel, analyze and judge, and generate a signal to be analyzed;

[0010] Step 2: Based on the signal to be analyzed, the real-time temperature data is analyzed and processed, and the analysis evaluation value is output, and compared with the analysis evaluation threshold to obtain the change law of the real-time temperature;

[0011] Step 3: Based on the changing pattern of real-time temperature, the predicted time for the output temperature to reach the insulation temperature;

[0012] Step 4: Analyze and process the predicted time and temperature deviation values, output the control coefficient, and adjust the heating parameters in the heating stage.

[0013] As a further solution of the present invention: the specific process of step 1 is:

[0014] Divide the total time of the heating stage into several monitoring time points, and obtain the real-time temperature at each monitoring time point;

[0015] Draw the standard temperature curve and the real-time temperature curve in the rectangular coordinate system of the time-temperature plane;

[0016] At each monitoring time point, the difference between the real-time temperature and the standard temperature is calculated, and the absolute value of the difference is taken to obtain the temperature deviation value;

[0017] Comparing the temperature deviation value with a first temperature deviation threshold and a second temperature deviation threshold respectively;

[0018] If the temperature deviation value is greater than the first temperature deviation threshold and the temperature deviation value is less than the second temperature deviation threshold, a signal to be analyzed is generated.

[0019] As a further solution of the present invention: the specific process of obtaining the change law of real-time temperature is:

[0020] Obtaining an analysis evaluation value, and comparing the analysis evaluation value with an analysis evaluation threshold;

[0021] If the analysis evaluation value is less than or equal to the analysis evaluation threshold, a regular change signal is generated;

[0022] If the analysis evaluation value is greater than the analysis evaluation threshold, an irregular change signal is generated.

[0023] As a further solution of the present invention: the process of obtaining the analysis evaluation value is:

[0024] Obtain the percentage value of the large deviation sub-curve segment and the slope deviation degree value;

[0025] Substitute into the formula The analysis evaluation value FX is calculated, where PC represents the proportion of the large deviation sub-curve segment, CD represents the slope deviation degree, and a1 and a2 are preset proportional coefficients.

[0026] As a further solution of the present invention: the process of obtaining the value of the proportion of the large deviation sub-curve segment is as follows:

[0027] Obtaining the monitoring time point corresponding to the generation of the signal to be analyzed, marking the time period from the start time point of the heating phase to the monitoring time point corresponding to the generation of the signal to be analyzed as the time period to be analyzed, and dividing the time period to be analyzed into a number of sub-time periods to be analyzed;

[0028] Obtaining a real-time temperature curve of a time period to be analyzed, marking the curve between each sub-time period to be analyzed in the real-time temperature curve as a sub-curve segment, and calculating the slope value of each sub-curve segment;

[0029] The slope values ​​of each sub-curve segment are summed and averaged to obtain the slope mean of the sub-curve segment, the slope of each sub-curve segment is calculated to be different from the slope mean, and the absolute value of the difference is taken to obtain the slope deviation value;

[0030] The slope deviation value is compared with the slope deviation threshold value, and if the slope deviation value is greater than the slope deviation threshold value, a large deviation signal is generated;

[0031] Based on the sub-curve segments corresponding to the generated large deviation signals, they are marked as large deviation sub-curve segments, the number of large deviation sub-curve segments is counted, and the number of large deviation sub-curve segments is ratioed to the total number of sub-curve segments to obtain the proportion of large deviation sub-curve segments.

[0032] As a further solution of the present invention: the process of obtaining the slope deviation value is as follows:

[0033] The slope deviation value corresponding to the sub-curve segment with the large deviation is extracted, the difference between the slope deviation value and the slope deviation threshold is calculated, the difference is calculated with the slope deviation threshold to obtain the slope deviation ratio, the slope deviation ratios are summed and averaged to obtain the slope deviation degree value.

[0034] As a further solution of the present invention: the process of obtaining the predicted time includes:

[0035] Based on the regular change signal, the slope mean of the sub-curve segment is obtained;

[0036] Substitute into the formula The predicted time YT is calculated, where BW represents the standard temperature for reaching the insulation temperature, DW represents the temperature corresponding to the monitoring time point corresponding to the generation of the signal to be analyzed, KJ represents the mean slope of the sub-curve segment, and DT represents the monitoring time point corresponding to the generation of the signal to be analyzed.

[0037] As a further solution of the present invention: the process of obtaining the predicted time further includes:

[0038] Based on the generated irregular change signal, further analyze the real-time temperature curve of the time period to be analyzed;

[0039] Obtain the temperature data for the period to be analyzed, and divide the temperature data into a training set and a validation set in proportion;

[0040] Select recurrent neural network (RNN) and its variants as the temperature change model;

[0041] Use the training set data to train the RNN and its variants to obtain a trained temperature change model, and then use the validation set data to validate the trained model and evaluate the model performance;

[0042] If the model performance meets the standard expectations, the trained model is the required temperature change model;

[0043] If the model performance does not meet the standard expectations, return to adjust the model parameters or structure, and then retrain and verify until the model performance meets the standard expectations;

[0044] Substitute the standard temperature reaching the insulation temperature into the temperature change model and calculate the predicted time YT.

[0045] As a further solution of the present invention: the process of obtaining the control coefficient is:

[0046] The predicted time and the standard time when the temperature reaches the insulation temperature are calculated, the absolute value of the difference is taken to obtain a time deviation value, and the time deviation value is compared with the time deviation threshold;

[0047] If the time deviation value is less than or equal to the time deviation threshold, a valid signal is generated;

[0048] Based on the effective signal, the time deviation value is calculated by ratio with the predicted time to obtain the time deviation coefficient;

[0049] Obtaining the temperature deviation value and the real-time temperature corresponding to the monitoring time point at which the signal to be analyzed is generated, calculating the ratio of the temperature deviation value to the real-time temperature, and obtaining the temperature deviation coefficient;

[0050] Substitute into the formula TK=s1*SX+s2*WX and calculate the control coefficient TK, where SX represents the time deviation coefficient, WX represents the temperature deviation coefficient, and s1 and s2 are weight coefficients.

[0051] As a further solution of the present invention: the process of adjusting the heating parameters in the heating stage is:

[0052] The heating power corresponding to the monitoring time point at which the signal to be analyzed is generated is obtained, marked as the power to be adjusted, and the power to be adjusted is multiplied by the control coefficient to obtain the power to be adjusted;

[0053] The predicted time is compared with the standard time. If the predicted time is less than the standard time, the power to be adjusted and the power to be adjusted are calculated by difference to obtain the target power. If the predicted time is greater than the standard time, the power to be adjusted and the power to be adjusted are added to obtain the target power. If the predicted time is equal to the standard time, no power adjustment is performed.

[0054] Beneficial effects of the present invention:

[0055] (1) The present invention obtains temperature data in the heating stage in real time, calculates the deviation from the standard temperature, i.e., the temperature deviation value, compares the temperature deviation value with the first temperature deviation threshold and the second temperature deviation threshold, respectively, and identifies and generates a normal temperature signal, a signal to be analyzed, or an abnormal temperature signal, so that abnormal conditions in the heating process can be discovered in time and processed. At the same time, the signal to be analyzed can be further processed and analyzed to optimize the parameters of the heating process and improve the processing effect;

[0056] (2) The present invention analyzes and processes the temperature data to calculate the analysis evaluation value, compares the analysis evaluation value with the analysis evaluation threshold, and determines the temperature change law. Based on the temperature change law, the predicted time for the temperature to reach the insulation temperature can be calculated, and then the temperature monitoring and time prediction can be performed to provide data support for optimizing the quenching and tempering process of spring steel, thereby improving production efficiency and product quality.

[0057] (3) The present invention calculates the time deviation coefficient and the temperature deviation coefficient by analyzing the predicted time and temperature deviation values, calculates the control coefficient according to the time deviation coefficient and the temperature deviation coefficient, and adjusts the heating power based on the control coefficient, so that the subsequent heating process can be optimized through adjustment, and the performance degradation of the spring steel caused by the heating time being too long or too short in the heating stage can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0059] Figure 1 It is a flow chart of a temperature monitoring system in a quenching and tempering process of ultra-high strength spring steel of the present invention;

[0060] Figure 2 It is a flowchart of obtaining analysis and evaluation values ​​in a quenching and tempering process of ultra-high strength spring steel of the present invention;

[0061] Figure 3 It is a flowchart of obtaining a control coefficient in a quenching and tempering process of an ultra-high strength spring steel according to the present invention. DETAILED DESCRIPTION

[0062] 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.

[0063] Embodiment 1:

[0064] See also Figure 1 As shown, the heating process of quenching and tempering of spring steel requires a certain amount of time. A heating rate that is too fast may lead to uneven temperature distribution inside the material, causing some areas to be overheated and some areas to be underheated, which will lead to uneven tissue transformation and affect the performance of the material. A heating rate that is too slow may cause the material to be exposed to high temperature for a long time during the heating process, thereby increasing the risk of oxidation and decarburization, and damaging the surface quality and performance of the material. A quenching and tempering process for ultra-high strength spring steel described in an embodiment of the present invention sets a temperature monitoring system to monitor the heating stage in the quenching and tempering of spring steel, which specifically includes the following steps:

[0065] Step 1: Obtain real-time temperature data during the heating stage of spring steel quenching and tempering treatment, and make analysis and judgment;

[0066] In some embodiments, the total time of the heating stage is divided into a number of monitoring time points, and the real-time temperature at each monitoring time point is obtained;

[0067] Draw the standard temperature curve and the real-time temperature curve in the rectangular coordinate system of the time-temperature plane;

[0068] It should be explained that the standard temperature curve is drawn by summarizing the historical experimental data and is an ideal temperature rise state. The temperature rise time of the standard temperature curve is the standard time required for temperature rise, and the temperature corresponding to each monitoring time point is the standard temperature reached at that monitoring time point.

[0069] At each monitoring time point, the difference between the real-time temperature and the standard temperature is calculated, and the absolute value of the difference is taken to obtain the temperature deviation value;

[0070] The temperature deviation value is compared with a first temperature deviation threshold and a second temperature deviation threshold, respectively, wherein the temperature deviation threshold is set by a person skilled in the art based on a summary of multiple historical experimental data, and the first temperature deviation threshold is less than the second temperature deviation threshold;

[0071] If the temperature deviation value is less than or equal to the first temperature deviation threshold, it means that at the corresponding monitoring time point, the deviation between the real-time temperature and the standard temperature is small, and a normal temperature signal is generated;

[0072] If the temperature deviation value is greater than the first temperature deviation threshold and the temperature deviation value is less than the second temperature deviation threshold, it means that the deviation between the real-time temperature and the standard temperature at the corresponding monitoring time point is large, and a signal to be analyzed is generated;

[0073] If the temperature deviation value is greater than or equal to the second temperature deviation threshold, it means that at the corresponding monitoring time point, the deviation between the real-time temperature and the standard temperature is too large, and a temperature abnormality signal is generated;

[0074] Based on the generated temperature anomaly signal, the technician immediately stops the heating process to avoid further damage to the material and checks whether the heating equipment and process parameters have faults or anomalies, such as short circuit of the heating element, failure of the temperature sensor, etc.;

[0075] The technical solution of the embodiment of the present invention is mainly as follows: by acquiring the temperature data of the heating stage in real time and calculating the deviation from the standard temperature, that is, the temperature deviation value, the temperature deviation value is compared with the first temperature deviation threshold and the second temperature deviation threshold, respectively, to identify and generate a normal temperature signal, a signal to be analyzed or a temperature abnormality signal; when a normal temperature signal is generated, temperature monitoring is continued; when an abnormal temperature signal is generated, the heating process is immediately stopped to find out the cause; when a signal to be analyzed is generated, the temperature data is processed and analyzed, so that abnormal conditions in the heating process can be discovered in time and processed; at the same time, the signal to be analyzed can be further processed and analyzed to optimize the parameters of the heating process and improve the processing effect.

[0076] Embodiment 2:

[0077] Based on Example 1, please refer to Figure 1 and Figure 2 As shown, the quenching and tempering process of the ultra-high strength spring steel described in the embodiment of the present invention further includes the following steps:

[0078] Step 2: Based on the signal to be analyzed, the real-time temperature data is analyzed and processed to obtain the change law of the real-time temperature;

[0079] Among them, the changing rules of real-time temperature include regular changes and irregular changes;

[0080] In some embodiments, a monitoring time point corresponding to the generation of a signal to be analyzed is obtained, a time period from the start time point of the heating phase to the monitoring time point corresponding to the generation of the signal to be analyzed is marked as a time period to be analyzed, and the time period to be analyzed is divided into a plurality of sub-time periods to be analyzed;

[0081] Obtaining a real-time temperature curve of a time period to be analyzed, marking the curve between each sub-time period to be analyzed in the real-time temperature curve as a sub-curve segment, and calculating the slope value of each sub-curve segment;

[0082] The slope values ​​of each sub-curve segment are summed and averaged to obtain the slope mean of the sub-curve segment, the slope of each sub-curve segment is calculated to be different from the slope mean, and the absolute value of the difference is taken to obtain the slope deviation value;

[0083] Compare the slope deviation value with the slope deviation threshold, wherein the slope deviation threshold is set by a person skilled in the art based on a summary of multiple historical experimental data;

[0084] If the slope deviation value is less than or equal to the slope deviation threshold, a small deviation signal is generated; if the slope deviation value is greater than the slope deviation threshold, a large deviation signal is generated;

[0085] Based on the sub-curve segments corresponding to the generated large deviation signals, they are marked as large deviation sub-curve segments, the number of large deviation sub-curve segments is counted, and the number of large deviation sub-curve segments is processed by ratio processing with the total number of sub-curve segments to obtain the proportion of large deviation sub-curve segments;

[0086] Extract the slope deviation value corresponding to the sub-curve segment with the large deviation, calculate the difference between the slope deviation value and the slope deviation threshold, calculate the ratio of the difference to the slope deviation threshold to obtain the slope deviation ratio, sum and average the slope deviation ratios to obtain the slope deviation degree value;

[0087] It should be noted that the temperature in the heating stage rises gradually, so the slopes of the sub-curve segments are all positive;

[0088] Substitute into the formula The analysis evaluation value FX is calculated, where PC represents the percentage of the large deviation sub-curve segment, CD represents the slope deviation degree, a1 and a2 are preset proportional coefficients, where a1 is 2.56 and a2 is 2.44;

[0089] It needs to be explained that the meaning reflected by the analysis evaluation value is: the analysis evaluation value is calculated by the proportion of sub-curve segments with large deviations and the slope deviation degree value, and is used to evaluate the degree of slope deviation of each sub-curve segment in the real-time temperature change curve of the time period to be analyzed, wherein the proportion of sub-curve segments with large deviations is the sub-curve corresponding to the large deviation between the slope and the slope mean value, and the more the number of sub-curve segments, the greater the slope deviation degree of the sub-curve segment, and the slope deviation degree value is the difference between the slope deviation value of the sub-curve segment with large deviations and the slope deviation threshold value, and the greater the difference is, the greater the slope deviation degree of the sub-curve segment;

[0090] Comparing the analytical evaluation value with an analytical evaluation threshold, wherein the analytical evaluation threshold is set by a person skilled in the art based on a summary of multiple historical experimental data;

[0091] If the analysis evaluation value is less than or equal to the analysis evaluation threshold, it means that the sub-curve slope deviation in each analysis sub-period is not large, and the change of real-time temperature is regular, generating a regular change signal;

[0092] If the analysis evaluation value is greater than the analysis evaluation threshold, it means that the sub-curve slope deviation in each analysis sub-period is large, and the change of real-time temperature is irregular, generating an irregular change signal;

[0093] Among them, standards are expected to be set by those skilled in the art based on historical data and experience;

[0094] Step 3: Based on the changing pattern of real-time temperature, the predicted time for the output temperature to reach the insulation temperature;

[0095] It should be noted that the holding temperature refers to the time when the temperature reaches the quenching temperature during the quenching process and the time when the temperature reaches the tempering temperature during the tempering process.

[0096] In some embodiments, based on the regular variation signal, the average slope of the sub-curve segment is obtained;

[0097] Substitute into the formula The predicted time YT is calculated, where BW represents the standard temperature that reaches the insulation temperature, DW represents the temperature corresponding to the monitoring time point corresponding to the generation of the signal to be analyzed, KJ represents the mean slope of the sub-curve segment, and DT represents the monitoring time point corresponding to the generation of the signal to be analyzed;

[0098] Based on the irregular signal, further analyzing the real-time temperature curve of the time period to be analyzed is performed based on the generated irregular change signal;

[0099] In some embodiments, temperature data of the time period to be analyzed is obtained, and the temperature data is divided into a training set and a validation set in proportion (e.g., 70% training set, 30% validation set);

[0100] Use the temperature data in the training set to train the temperature change model;

[0101] The temperature change model selection includes but is not limited to: support vector machine (SVM), neural network (such as multi-layer perceptron MLP, convolutional neural network CNN or recurrent neural network RNN, etc.), decision tree, random forest and integrated learning method (such as gradient boosting machine GBM, XGBoost, etc.);

[0102] In one possible embodiment, a recurrent neural network (RNN) and its variants are selected as the temperature change model;

[0103] It needs to be explained that RNN and its variants are particularly suitable for processing time series data and can capture the time dependency in the data, so they are suitable for the prediction and analysis of temperature changes;

[0104] Use the training set data to train the RNN and its variants to obtain a trained temperature change model, and then use the validation set data to validate the trained model and evaluate the model performance;

[0105] If the model performance meets the standard expectations, the trained model is the required temperature change model;

[0106] If the model performance does not meet the standard expectations, return to adjust the model parameters or structure, and then retrain and verify until the model performance meets the standard expectations;

[0107] Substitute the standard temperature that reaches the insulation temperature into the temperature change model and calculate the predicted time YT;

[0108] The technical solution of the embodiment of the present invention is mainly as follows: based on the signal to be analyzed, the temperature data is analyzed and processed, and an analysis evaluation value is calculated, and the analysis evaluation value is compared with the analysis evaluation threshold. When the analysis evaluation value is less than or equal to the analysis evaluation threshold, the temperature change tends to be linear. When the analysis evaluation value is greater than the analysis evaluation threshold, the temperature change does not show a regular change. By training the temperature data, a temperature change model is obtained, so that based on the temperature change law, the predicted time for the temperature to reach the insulation temperature can be calculated, and then the temperature can be monitored and the time can be predicted, so as to provide data support for optimizing the quenching and tempering process of spring steel and improve production efficiency and product quality.

[0109] Embodiment 3:

[0110] Based on Example 1 and Example 2, please refer to Figure 1 and Figure 3 As shown, the quenching and tempering process of the ultra-high strength spring steel described in the embodiment of the present invention further includes the following steps:

[0111] Step 4: Analyze and process the predicted time and temperature deviation values, output the control coefficient, and adjust the heating parameters in the heating stage;

[0112] Wherein, the heating parameters include heating power;

[0113] In some embodiments, the predicted time and the standard time when the temperature reaches the insulation temperature are calculated, the difference is taken as the absolute value, and the time deviation value is obtained, and the time deviation value is compared with the time deviation threshold, wherein the time deviation threshold is set by a person skilled in the art based on a summary of multiple historical experimental data;

[0114] If the time deviation value is greater than the time deviation threshold, it means that the predicted time is outside the reasonable range, generating an invalid signal;

[0115] Based on the invalid signal, the technicians need to immediately check whether the heating equipment and process parameters are faulty or abnormal;

[0116] If the time deviation value is less than or equal to the time deviation threshold, it means that the predicted time is within a reasonable range and a valid signal is generated;

[0117] Based on the effective signal, the time deviation value is calculated by ratio with the predicted time to obtain the time deviation coefficient;

[0118] Obtaining the temperature deviation value and the real-time temperature corresponding to the monitoring time point at which the signal to be analyzed is generated, calculating the ratio of the temperature deviation value to the real-time temperature, and obtaining the temperature deviation coefficient;

[0119] Substitute into the formula TK=s1*SX+s2*WX, and calculate the control coefficient TK, where SX represents the time deviation coefficient, WX represents the temperature deviation coefficient, s1 and s2 are weight coefficients, s1 is 0.58, and s2 is 0.42;

[0120] The heating power corresponding to the monitoring time point at which the signal to be analyzed is generated is obtained, marked as the power to be adjusted, and the power to be adjusted is multiplied by the control coefficient to obtain the power to be adjusted;

[0121] Compare the predicted time with the standard time. If the predicted time is less than the standard time, the power to be adjusted and the power to be adjusted are calculated by difference to obtain the target power. If the predicted time is greater than the standard time, the power to be adjusted and the power to be adjusted are added to obtain the target power. If the predicted time is equal to the standard time, no power adjustment is performed.

[0122] The technical solution of the embodiment of the present invention is mainly as follows: a time deviation coefficient and a temperature deviation coefficient are calculated by analyzing and predicting the time and temperature deviation values, a control coefficient is calculated according to the time deviation coefficient and the temperature deviation coefficient, and the heating power is adjusted based on the control coefficient, so that the subsequent heating process can be optimized through adjustment, and the performance degradation of the spring steel caused by the heating time being too long or too short in the heating stage can be reduced.

[0123] 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 quenching and tempering process for ultra-high strength spring steel, characterized in that: Setting up a temperature monitoring system to monitor the temperature rise stage during the quenching and tempering treatment of spring steel includes the following steps: Step 1: Obtain real-time temperature data of the heating stage during the quenching and tempering treatment of spring steel, analyze and judge, and generate a signal to be analyzed; Step 2: Based on the signal to be analyzed, the real-time temperature data is analyzed and processed, and the analysis evaluation value is output, and compared with the analysis evaluation threshold to obtain the change law of the real-time temperature; Step 3: Based on the changing pattern of real-time temperature, the predicted time for the output temperature to reach the insulation temperature; Step 4: Analyze and process the predicted time and temperature deviation values, output the control coefficient, and adjust the heating parameters in the heating stage.

2. A quenching and tempering process for ultra-high strength spring steel according to claim 1, characterized in that: The specific process of step one is: Divide the total time of the heating stage into several monitoring time points, and obtain the real-time temperature at each monitoring time point; Draw the standard temperature curve and the real-time temperature curve in the rectangular coordinate system of the time-temperature plane; At each monitoring time point, the difference between the real-time temperature and the standard temperature is calculated, and the absolute value of the difference is taken to obtain the temperature deviation value; Comparing the temperature deviation value with a first temperature deviation threshold and a second temperature deviation threshold respectively; If the temperature deviation value is greater than the first temperature deviation threshold and the temperature deviation value is less than the second temperature deviation threshold, a signal to be analyzed is generated.

3. The quenching and tempering process of ultra-high strength spring steel according to claim 1, characterized in that: The specific process of obtaining the change rule of real-time temperature is as follows: Obtaining an analysis evaluation value, and comparing the analysis evaluation value with an analysis evaluation threshold; If the analysis evaluation value is less than or equal to the analysis evaluation threshold, a regular change signal is generated; If the analysis evaluation value is greater than the analysis evaluation threshold, an irregular change signal is generated.

4. The quenching and tempering process of ultra-high strength spring steel according to claim 1, characterized in that: The process of obtaining the analysis evaluation value is as follows: Obtain the percentage value of the large deviation sub-curve segment and the slope deviation degree value; Substitute into the formula The analysis evaluation value FX is calculated, where PC represents the proportion of the large deviation sub-curve segment, CD represents the slope deviation degree, and a1 and a2 are preset proportional coefficients.

5. A quenching and tempering process for ultra-high strength spring steel according to claim 4, characterized in that: The process of obtaining the value of the proportion of the large deviation sub-curve segment is as follows: Obtaining the monitoring time point corresponding to the generation of the signal to be analyzed, marking the time period from the start time point of the heating phase to the monitoring time point corresponding to the generation of the signal to be analyzed as the time period to be analyzed, and dividing the time period to be analyzed into a number of sub-time periods to be analyzed; Obtaining a real-time temperature curve of a time period to be analyzed, marking the curve between each sub-time period to be analyzed in the real-time temperature curve as a sub-curve segment, and calculating the slope value of each sub-curve segment; The slope values ​​of each sub-curve segment are summed and averaged to obtain the slope mean of the sub-curve segment, the slope of each sub-curve segment is calculated to be different from the slope mean, and the absolute value of the difference is taken to obtain the slope deviation value; The slope deviation value is compared with the slope deviation threshold value, and if the slope deviation value is greater than the slope deviation threshold value, a large deviation signal is generated; Based on the sub-curve segments corresponding to the generated large deviation signals, they are marked as large deviation sub-curve segments, the number of large deviation sub-curve segments is counted, and the number of large deviation sub-curve segments is ratioed to the total number of sub-curve segments to obtain the proportion of large deviation sub-curve segments.

6. A quenching and tempering process for ultra-high strength spring steel according to claim 5, characterized in that: The process of obtaining the slope deviation value is as follows: The slope deviation value corresponding to the sub-curve segment with the large deviation is extracted, the difference between the slope deviation value and the slope deviation threshold is calculated, the difference is calculated with the slope deviation threshold to obtain the slope deviation ratio, the slope deviation ratios are summed and averaged to obtain the slope deviation degree value.

7. The quenching and tempering process of ultra-high strength spring steel according to claim 3, characterized in that: The process of obtaining the predicted time includes: Based on the regular change signal, the slope mean of the sub-curve segment is obtained; Substitute into the formula The predicted time YT is calculated, where BW represents the standard temperature for reaching the insulation temperature, DW represents the temperature corresponding to the monitoring time point corresponding to the generation of the signal to be analyzed, KJ represents the mean slope of the sub-curve segment, and DT represents the monitoring time point corresponding to the generation of the signal to be analyzed.

8. The quenching and tempering process of ultra-high strength spring steel according to claim 3, characterized in that: The process of obtaining the predicted time also includes: Based on the generated irregular change signal, further analyze the real-time temperature curve of the time period to be analyzed; Obtain the temperature data for the period to be analyzed, and divide the temperature data into a training set and a validation set in proportion; Select recurrent neural network (RNN) and its variants as the temperature change model; Use the training set data to train the RNN and its variants to obtain a trained temperature change model, and then use the validation set data to validate the trained model and evaluate the model performance; If the model performance meets the standard expectations, the trained model is the required temperature change model; If the model performance does not meet the standard expectations, return to adjust the model parameters or structure, and then retrain and verify until the model performance meets the standard expectations; Substitute the standard temperature reaching the insulation temperature into the temperature change model and calculate the predicted time YT.

9. The quenching and tempering process of ultra-high strength spring steel according to claim 1, characterized in that: The process of obtaining the control coefficient is as follows: The predicted time and the standard time when the temperature reaches the insulation temperature are calculated, the absolute value of the difference is taken to obtain a time deviation value, and the time deviation value is compared with the time deviation threshold; If the time deviation value is less than or equal to the time deviation threshold, a valid signal is generated; Based on the effective signal, the time deviation value is calculated by ratio with the predicted time to obtain the time deviation coefficient; Obtaining the temperature deviation value and the real-time temperature corresponding to the monitoring time point at which the signal to be analyzed is generated, calculating the ratio of the temperature deviation value to the real-time temperature, and obtaining the temperature deviation coefficient; Substitute into the formula TK=s1*SX+s2*WX and calculate the control coefficient TK, where SX represents the time deviation coefficient, WX represents the temperature deviation coefficient, and s1 and s2 are weight coefficients.

10. A quenching and tempering process for ultra-high strength spring steel according to claim 9, characterized in that: The process of adjusting the heating parameters in the heating stage is as follows: The heating power corresponding to the monitoring time point at which the signal to be analyzed is generated is obtained, marked as the power to be adjusted, and the power to be adjusted is multiplied by the control coefficient to obtain the power to be adjusted; The predicted time is compared with the standard time. If the predicted time is less than the standard time, the power to be adjusted and the power to be adjusted are calculated by difference to obtain the target power. If the predicted time is greater than the standard time, the power to be adjusted and the power to be adjusted are added to obtain the target power. If the predicted time is equal to the standard time, no power adjustment is performed.

Images