Combined machining process optimization method based on residual stress machining process marks
Through simulation processing experiments and regression models, a material residual stress characteristic database was established, and the processing process parameters were reversely optimized, which solved the problem of complex residual stress coupling in the existing technology, improved manufacturing efficiency and prediction accuracy, reduced production costs, and met aviation. High-performance machining requirements for complex components.
Patent Information
- Application Number
- CN202410944125.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-08
AI Technical Summary
In the existing tongue and groove processing combination process, residual stress coupling is complex, resulting in difficult to trace performance errors, low manufacturing efficiency, high production cost, and process parameter optimization depends on experimental iteration, which is not very efficient.
The residual stress characteristic database of materials is established through simulation processing experiments, and the linear correlation between processing process parameters and load is fitted using the regression model, and the combined processing process parameters are optimized in reverse derivation to avoid interference from process parameters and directly predict processing quality.
The optimization of the processing process parameters of the tongue and groove combination is achieved, manufacturing efficiency is improved, production costs is reduced, prediction accuracy and efficiency is enhanced, and a new process combination is inferred to meet the high-performance processing needs of aviation complex components.
Smart Images

Figure CN118917063A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of mortise and tenon processing combined process optimization, and in particular relates to a combined processing process optimization method based on residual stress processing process imprints. Background Art
[0002] With the continuous improvement of performance requirements for high-end power equipment such as the new generation of aircraft engines and gas turbines, the structure and size of components are constantly optimized, and the requirements for processing accuracy and surface quality are gradually increasing. At the same time, the continuous application of new difficult-to-process materials has also made the processing difficulty of mortise and tenon gradually increase. Therefore, a single processing method has entered a bottleneck period and it is difficult to meet the increasingly high demand for improving the quality, increasing efficiency and reducing costs of complex aviation components. The processing method is gradually tending to a combination of multiple processing methods that combine broaching, milling, grinding and special processing, which can give full play to the advantages of various processing methods, thereby achieving more economical, more efficient, higher precision and better quality processing of complex aviation components. However, the material removal mechanism in the combined processing technology is not the same, resulting in superposition of influence areas, coupling of residual stresses, complex change rules, and ultimately difficult to trace the performance errors. In addition, the optimization of combined processing process parameters for residual stress characteristics still mainly relies on experimental rules. Therefore, in the actual production of combined processing, the realization of high performance of parts requires continuous experimental iteration and trial and error manufacturing processing, and the establishment of an empirical relationship between residual stress characteristics and process parameters based on a large amount of experimental data. The empirical relationship is then gradually corrected according to the residual stress results of the parts, and the process parameters of each processing technology in the combined processing are repeatedly adjusted to obtain the desired part performance. This results in low manufacturing efficiency and high production costs. Summary of the invention
[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a combined processing technology optimization method based on residual stress processing process imprints, which is a method for constructing a material residual stress characteristic database through a single-factor simulation processing experiment, a method for realizing the linear correlation between material processing parameters and processing loads through regression model fitting, a method for extracting loads through a sensitivity method and establishing a processing process imprint characteristic relationship that is independent of processing process parameters, a method for reversely deducing and optimizing combined processing parameters through processing process imprint characteristic relationships, and a method for avoiding interference from processing process parameters and directly predicting processing quality with processing loads.
[0004] To achieve the above object, the present invention adopts the following technical solutions:
[0005] The combined processing technology optimization method based on the residual stress processing mark of the present invention is as follows:
[0006] Step 1, use the various processing techniques used in the combined processing technology of mortise and tenon materials to carry out simulation processing experiments respectively. The various processing techniques used in the combined processing technology of mortise and tenon materials include deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting. The process of using deep rolling, electric spark or electrolytic processing to carry out simulation processing experiments is as follows: set the processing parameter range of deep rolling, electric spark or electrolytic processing, and select n parameter levels within the processing parameter range, n ≥ 4, use each parameter level one to carry out simulation processing, remove the preset thickness of mortise and tenon materials, and collect the parameters below each parameter level located at the mortise and tenon material adding The residual stress data at different collection depths below the machining surface are collected; the simulation processing experiment process using cutting, electrochemical mechanical polishing or dealloying cutting processing methods is as follows: the range of each processing parameter of cutting, electrochemical mechanical polishing or dealloying cutting processing methods is set, and n parameter levels 2 are selected within the range of each processing parameter, and the parameter level 2 of each processing parameter is combined to obtain multiple processing parameter combinations, and each processing parameter combination is used for simulation processing to remove the mortise and tenon material of a preset thickness, and the residual stress data at different collection depths below the machining surface of the mortise and tenon material under each processing parameter combination are collected.
[0007] Step 2: Calculate the sensitivity matrix S of the influence of the load type of deep rolling, electric spark or electrochemical machining on the residual stress, where the load type of deep rolling is mechanical load, the load type of electric spark machining is thermal load, and the load type of electrochemical machining is chemical load. The calculation formula of the sensitivity matrix S is:
[0008]
[0009] In the formula, E k is the mechanical load, thermal load or chemical load corresponding to the kth parameter level 1, σ rk is the residual stress under mechanical load, thermal load or chemical load corresponding to the kth parameter level 1;
[0010] Draw a three-dimensional graph of the sensitivity of the load type to the residual stress and the relationship between the acquisition depth and the load size under different load types. Then select the largest sensitivity values in the three three-dimensional graphs for comparison and select the largest sensitivity value. The load type corresponding to the maximum sensitivity value is the load type that has the greatest impact on the residual stress.
[0011] Step 3. Establish a relationship between the load size of the load type with the highest sensitivity to residual stress and its corresponding residual stress, and obtain the processing process imprint characteristic relationship between the load type with the highest sensitivity to residual stress and the residual stress. Specifically, normalize the load size corresponding to the load type with the highest sensitivity to residual stress at each parameter level, and then fit the linear relationship between each load size after normalization and the corresponding residual stress.
[0012] Step 4: Fit the regression equation between the load size and the corresponding processing parameters under deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting processing modes, and establish the regression equation relationship between the load size and the corresponding processing parameters under each processing mode.
[0013] Step 5. Use the inverse problem solving method to determine the processing parameters of each processing method in the processing route of the combined processing technology of the mortise and tenon material, specifically: ① Preset the residual stress at different sampling depths below the processing surface of the mortise and tenon material; ② According to the preset residual stress, the total load size of the load type with the highest sensitivity to the residual stress is obtained through the processing process imprint characteristic relationship, and then the total load size is distributed to each processing technology with this load type in ascending order of the complexity of each processing technology in the mortise and tenon processing route; ③ If there is no total load size remaining after executing step ②, increase the preset value of the residual stress at different sampling depths below the processing surface of the mortise and tenon material, and return to step ② to execute, otherwise the remaining load size after distributing the total load size to each processing technology with this load type is then distributed to each processing technology without this load type in ascending order of the complexity of each processing technology in the mortise and tenon processing route; ④ If the total load size is not left after executing step ②, then increase the preset value of the residual stress at different sampling depths below the processing surface of the mortise and tenon material, and return to step ② to execute, otherwise the total load size is distributed to each processing technology with this load type, and then the remaining load size is distributed to each processing technology without this load type in ascending order of the complexity of each processing technology in the mortise and tenon processing route; After executing step ③, if there are processing processes in the mortise and tenon processing route that are not assigned to loads, the preset values of residual stress at different sampling depths below the processing surface of the mortise and tenon material are increased, and the process returns to step ② for execution, otherwise the assignment is terminated; the load size assigned to each processing process is the maximum load size calculated by the regression equation corresponding to the processing process according to the processing parameter range of the processing process, and the processing parameter value corresponding to the calculated maximum load size is the optimal processing parameter value of the processing process, thereby obtaining the optimal processing parameter value of each processing process in the mortise and tenon processing route; among them, the complexity of deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting gradually increases, the load type of the cutting processing method is mechanical-thermal coupling load, the load type of the electrochemical mechanical polishing processing method is mechanical-chemical coupling load, and the load type of the dealloying cutting processing method is mechanical-thermal-chemical coupling load.
[0014] Preferably, the load size is characterized by the size of processing energy, wherein the processing energy of mechanical load is mechanical energy, the processing energy of thermal load is thermal energy, the processing energy of chemical load is chemical energy, the processing energy of mechanical-thermal coupling load is coupling energy composed of mechanical energy and thermal energy, the processing energy of mechanical-chemical coupling load is coupling energy composed of mechanical energy and chemical energy, and the processing energy of mechanical-thermal-chemical coupling load is coupling energy composed of mechanical energy, maximum thermal energy and chemical energy.
[0015] Preferably, the processing parameters of the deep rolling processing method are rolling force, the processing parameters of the electrospark machining method are machining voltage, the processing parameters of the electrolytic machining method are electrolyte concentration, the processing parameters of the cutting processing method are cutting speed and cutting depth, the processing parameters of the electrochemical mechanical polishing processing method are electrolyte concentration and abrasive particle size, and the processing parameters of the dealloying cutting processing method are electrolyte concentration, cutting speed and cutting depth.
[0016] Preferably, in step 1, ABAQUS software is used to perform a simulation processing experiment, and residual stress data is collected through an ABAQUS subroutine.
[0017] Preferably, the normalization process in step 3 adopts the Z-Score normalization formula, and the Z-Score normalization formula is:
[0018]
[0019] Where x′ is the normalized load size, X is the load size before normalization, μ is the average load size corresponding to each parameter level before normalization, and σ is the standard deviation of the load size corresponding to each parameter level before normalization.
[0020] Preferably, the process of fitting the linear relationship between each load size and the corresponding residual stress after normalization in step 3 is:
[0021] (1) Draw a scatter plot of each load size and corresponding residual stress after normalization;
[0022] (2) Analyze the trend of the scatter plot and determine the form of the regression equation based on the trend of the scatter plot;
[0023] (3) According to the determined regression equation form, the corresponding regression model is called to fit the normalized load magnitudes and the corresponding residual stresses, thereby obtaining the linear relationship between the normalized load magnitudes and the corresponding residual stresses.
[0024] Preferably, the feasibility of determining the processing parameters using the inverse problem solving method is judged by calculating the upper limit of the relative error of deep rolling, electric spark or electrochemical machining in the collected depth distribution set. The larger the upper limit of the relative error, the lower the feasibility of determining the processing parameters using the inverse problem solving method, and vice versa.
[0025] More preferably, the expression of the upper limit of relative error is
[0026]
[0027] In the formula, + r is the upper limit of relative error, z is the collection depth distribution set, I is the insensitivity matrix of the influence of load type on residual stress of deep rolling, electric spark or electrochemical machining, err is the absolute error of the residual stress value of the same collection depth of the tongue and groove material calculated by the relationship between the simulation machining experiment and the machining process imprint characteristics at the same parameter level of deep rolling, electric spark or electrochemical machining;
[0028] Among them, the expression of the insensitivity matrix I is
[0029] I=(S T S) -1 S T .
[0030] The present invention has the following beneficial effects:
[0031] The present invention reversely derives the optimal processing parameters of each processing method in the processing route of the mortise and tenon combination processing technology by establishing the processing process imprint characteristic relationship between the load type with the highest sensitivity to residual stress and the residual stress, as well as the regression equation relationship between the load size and the processing parameters under each processing method, thereby realizing the optimization of the processing parameters of each processing method in the processing route of the mortise and tenon combination processing technology. Compared with the prior art, there is no need to repeatedly adjust the processing parameters of each processing method in the combined processing according to the residual stress results, thereby improving manufacturing efficiency and reducing production costs. Furthermore, residual stress can be predicted through the processing process imprint characteristic relationship and the corresponding regression equation relationship, avoiding the interference of processing parameters, improving prediction accuracy and efficiency, and inferring new process combinations, thereby providing useful exploration for high-performance processing of complex aviation components oriented to service performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION
[0033] The present invention is further described below in conjunction with the accompanying drawings.
[0034] like Figure 1As shown, the combined processing technology optimization method based on the residual stress processing imprint of the present invention is as follows:
[0035] Step 1. Use each processing technology used in the combined processing technology of mortise and tenon materials to carry out simulation processing experiments. The processing technologies used in the combined processing technology of mortise and tenon materials include deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting processing methods. The process of using deep rolling, electric spark or electrolysis processing methods to carry out simulation processing experiments is as follows: set the processing parameter range of deep rolling, electric spark or electrolysis processing methods, and select n parameter levels within the processing parameter range, n ≥ 4, use each parameter level to carry out simulation processing, remove the preset thickness of mortise and tenon materials, and collect different collection depths below the processing surface of the mortise and tenon materials at each parameter level. The residual stress data at the location (stress-strain distribution data can also be obtained); the simulation processing experiment process using cutting, electrochemical mechanical polishing or dealloying cutting processing methods is as follows: setting the range of each processing parameter of the cutting, electrochemical mechanical polishing or dealloying cutting processing methods, and selecting n parameter levels 2 within the range of each processing parameter, combining the parameter level 2 of each processing parameter to obtain multiple processing parameter combinations, using each processing parameter combination to perform simulation processing, removing the mortise and tenon material of a preset thickness, and collecting the residual stress data at different collection depths below the processing surface of the mortise and tenon material under each processing parameter combination (stress-strain distribution data can also be obtained).
[0036] Among them, the load type of the deep rolling processing method is mechanical load, and the processing energy data is mechanical energy; the load type of the electrospark processing method is thermal load, and the processing energy data is thermal energy; the load type of the electrolytic processing method is chemical load, and the processing energy data is chemical energy; the load type of the cutting processing method is mechanical-thermal coupling load, and the processing energy data is coupled energy composed of mechanical energy and thermal energy; the load type of the electrochemical mechanical polishing processing method is mechanical-chemical coupling load, and the processing energy data is coupled energy composed of mechanical energy and chemical energy; the load type of the dealloying cutting processing method is mechanical-thermal-chemical coupling load, and the processing energy data is coupled energy composed of mechanical energy, thermal energy and chemical energy.
[0037] In this embodiment, the processing parameters of the deep rolling processing method are rolling force, the processing parameters of the electrospark processing method are processing voltage, the processing parameters of the electrolytic processing method are electrolyte concentration, the processing parameters of the cutting processing method are cutting speed and cutting depth, the processing parameters of the electrochemical mechanical polishing processing method are electrolyte concentration and abrasive particle size, and the processing parameters of the dealloying cutting processing method are electrolyte concentration, cutting speed and cutting depth. In this embodiment, ABAQUS software is used to carry out simulation processing experiments, and the residual stress data at the collection depths of 0μm, 3μm, 6μm, 9μm, 12μm and 15μm below the processing surface of the mortise and tenon material are collected through the ABAQUS subroutine.
[0038] In order to verify the accuracy of the simulation processing experiment, the real processing experiment was carried out using each parameter level and each processing parameter combination to remove the preset thickness of the mortise and tenon material, and the residual stress data at different collection depths below the processing surface of the mortise and tenon material under each parameter level and each processing parameter combination in the real processing experiment were collected. The relative errors of each residual stress data under the simulation processing experiment and the corresponding residual stress data under the real processing experiment were calculated and analyzed, and the relative errors of each residual stress data were smaller than the corresponding preset relative errors, which further proved that the data accuracy of the simulation processing experiment was high.
[0039] Step 2: Taking the load size (which can be represented by the processing energy size) as input and the residual stress as the output response, calculate the sensitivity matrix S of the influence of the load type of deep rolling, electric spark or electrochemical processing on the residual stress. The calculation formula of the sensitivity matrix S is:
[0040]
[0041] In the formula, E k is the mechanical load, thermal load or chemical load corresponding to the kth parameter level 1, σ rk is the residual stress under mechanical load, thermal load or chemical load corresponding to the kth parameter level 1;
[0042] Among them, the sensitivity value is proportional to the rate of change of residual stress to load size, that is, a small change in load size will cause a change in residual stress, which quantitatively describes the influence of mechanical load, thermal load or chemical load on residual stress. The greater the influence of load size on the change of residual stress, the greater the sensitivity value of the response.
[0043] Draw a three-dimensional graph of the sensitivity of the load type to the residual stress and the relationship between the acquisition depth and the load size under different load types. Then select the largest sensitivity values in the three three-dimensional graphs for comparison and select the largest sensitivity value. The load type corresponding to the maximum sensitivity value is the load type that has the greatest impact on the residual stress.
[0044] Step 3. Establish a relationship between the load size of the load type with the highest sensitivity to residual stress and its corresponding residual stress, and obtain the processing process imprint characteristic relationship between the load type with the highest sensitivity to residual stress and the residual stress. Specifically, normalize the load size corresponding to the load type with the highest sensitivity to residual stress at each parameter level, and then fit the linear relationship between each load size after normalization and the corresponding residual stress.
[0045] Among them, the normalization process uses the Z-Score normalization formula, which is:
[0046]
[0047] Where x′ is the normalized load size, X is the load size before normalization, μ is the average load size corresponding to each parameter level before normalization, and σ is the standard deviation of the load size corresponding to each parameter level before normalization.
[0048] The process of fitting the linear relationship between each load size and the corresponding residual stress after normalization is:
[0049] (1) Draw a scatter plot of the normalized load magnitude and the corresponding residual stress data;
[0050] (2) Analyze the trend of the scatter plot and determine the form of the regression equation based on the trend of the scatter plot;
[0051] (3) According to the determined regression equation form, the corresponding regression model is called to fit the normalized load magnitudes and the corresponding residual stresses, thereby obtaining the linear relationship between the normalized load magnitudes and the corresponding residual stresses.
[0052] Step 4: Fit the regression equation between the load size and the corresponding processing parameters under each processing method, and establish the regression equation relationship between the load size and the corresponding processing parameters under each processing method.
[0053] Step 5. Use the inverse problem solving method to determine the processing parameters of each processing method in the processing route of the combined processing technology of the mortise and tenon material, specifically: ① Preset the residual stress at different sampling depths below the processing surface of the mortise and tenon material; ② According to the preset residual stress, the total load size of the load type with the highest sensitivity to the residual stress is obtained through the processing process imprint characteristic relationship, and then the load size is arranged in ascending order of the complexity of each processing technology in the mortise and tenon processing route. Allocate it to each processing technology with this load type in sequence; ③ If there is no total load size left after executing step ②, increase the preset value of the residual stress at different sampling depths below the processing surface of the mortise and tenon material, and return to step ② to execute, otherwise the total load size is allocated to each processing technology with this load type The remaining load size after the process is then distributed to each processing technology that does not have this load type in ascending order of the complexity of each processing technology in the mortise and tenon processing route; ④ If after executing step ③, there is a processing technology in the mortise and tenon processing route that has not been allocated a load, then increase the preset value of the residual stress at different acquisition depths below the processing surface of the mortise and tenon material, and return to step ② to execute, otherwise the allocation is terminated; the load size allocated to each processing technology is the maximum load size calculated by the regression equation corresponding to the processing technology according to the processing parameter range of the processing technology, and the processing parameter value corresponding to the calculated maximum load size is the optimal processing parameter value of the processing technology, and then the optimal processing parameter value of each processing technology in the mortise and tenon processing route is obtained. Among them, each processing technology in the mortise and tenon processing route is determined, and the complexity of deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting gradually increases. It should be noted that only when the total load size is allocated to all processing technologies in the mortise and tenon processing route, the preset residual stress is reasonable, otherwise the preset residual stress is too small, and the processing according to the mortise and tenon processing route cannot meet the preset residual stress requirements.
[0054] For example, if the processing techniques in the mortise and tenon processing route include dealloying cutting, electrochemical polishing and deep punching, and the mechanical load has the greatest influence on the residual stress after the combined processing of the mortise and tenon materials, the complexity of the dealloying cutting process in the mortise and tenon processing route is higher than that of the electrochemical polishing process, and the complexity of the electrochemical polishing process is higher than that of the deep punching process. When allocating, the mechanical load is first allocated to the deep punching, and the maximum mechanical load within the rolling force parameter range is calculated according to the rolling force parameter range of the deep punching and the regression equation of the mechanical load and the rolling force under the deep punching processing mode, and the rolling force corresponding to the maximum mechanical load is the optimal rolling force value within the rolling force parameter range. Then, the remaining mechanical load is allocated to the electrochemical polishing processing mode according to a similar allocation method to the deep punching processing mode, and the optimal processing parameter value of the electrochemical polishing processing is obtained. Finally, the remaining mechanical load is allocated to the dealloying cutting processing mode according to a similar allocation method, and the optimal processing parameter value of the dealloying cutting processing is obtained, thereby obtaining the optimal parameter values of each processing technique in the mortise and tenon processing route.
[0055] Step 6: Calculate the upper limit of relative error of deep rolling, electric spark or electrochemical machining in the collected depth distribution set, and determine the feasibility of determining machining parameters using the inverse problem solving method.
[0056] The expression of the upper limit of relative error is
[0057]
[0058] In the formula, + r is the upper limit of relative error, z is the collection depth distribution set, I is the insensitivity matrix of the influence of load type on residual stress of deep rolling, electric spark or electrochemical machining, err is the absolute error of the residual stress value of the same collection depth of the tongue and groove material calculated by the relationship between the simulation machining experiment and the machining process imprint characteristics at the same parameter level of deep rolling, electric spark or electrochemical machining;
[0059] Among them, the expression of the insensitivity matrix I is
[0060] I=(S T S) -1 S T
[0061] Among them, the upper limit of the relative error is determined by the insensitivity matrix I and the absolute error result of the direct problem, and the error of the inverse problem solution is jointly determined by the accuracy of the direct problem prediction model (i.e., the relationship between the load type with the highest sensitivity to residual stress and the residual stress, that is, the relationship model between the load size of the load type with the highest sensitivity to residual stress and its corresponding residual stress) and the sensitivity matrix S. Therefore, the feasibility of the direct problem model to predict residual stress and the feasibility of solving the inverse problem to optimize the processing load parameters can be verified by the error of the inverse problem solution.
[0062] If the upper limit of the relative error is larger, it means that the feasibility of determining the processing parameters by the inverse problem solving method is smaller, and vice versa. If the upper limit of the relative error is smaller than the preset upper limit of the relative error, it means that the feasibility of determining the processing parameters by the inverse problem solving method meets the requirements.
Claims
1. A combined processing technology optimization method based on residual stress processing imprint, characterized in that: The details are as follows: Step 1, use the various processing techniques used in the combined processing technology of mortise and tenon materials to carry out simulation processing experiments respectively. The various processing techniques used in the combined processing technology of mortise and tenon materials include deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting. The process of using deep rolling, electric spark or electrolytic processing to carry out simulation processing experiments is as follows: set the processing parameter range of deep rolling, electric spark or electrolytic processing, and select n parameter levels within the processing parameter range, n ≥ 4, use each parameter level one to carry out simulation processing, remove the preset thickness of mortise and tenon materials, and collect the parameters below each parameter level located at the mortise and tenon material adding Residual stress data at different acquisition depths below the machining surface; the simulation processing experiment process using cutting, electrochemical mechanical polishing or dealloying cutting processing methods is as follows: setting the range of each processing parameter of the cutting, electrochemical mechanical polishing or dealloying cutting processing method, and selecting n parameter levels 2 within the range of each processing parameter, combining the parameter level 2 of each processing parameter to obtain multiple processing parameter combinations, using each processing parameter combination to perform simulation processing, removing the mortise and tenon material of a preset thickness, and collecting residual stress data at different acquisition depths below the machining surface of the mortise and tenon material under each processing parameter combination; Step 2: Calculate the sensitivity matrix S of the influence of the load type of deep rolling, electric spark or electrochemical machining on the residual stress, where the load type of deep rolling is mechanical load, the load type of electric spark machining is thermal load, and the load type of electrochemical machining is chemical load. The calculation formula of the sensitivity matrix S is: In the formula, E k is the mechanical load, thermal load or chemical load corresponding to the kth parameter level 1, σ rk is the residual stress under mechanical load, thermal load or chemical load corresponding to the kth parameter level 1; Draw a three-dimensional graph of the sensitivity of the load type to the residual stress under different load types, the relationship between the acquisition depth and the load size, and then select the largest sensitivity values in the three three-dimensional graphs for comparison, and select the largest sensitivity value. The load type corresponding to the maximum sensitivity value is the load type that has the greatest impact on the residual stress; Step 3, establish a relationship between the load size of the load type with the highest sensitivity to residual stress and the corresponding residual stress, and obtain the processing process imprint characteristic relationship between the load type with the highest sensitivity to residual stress and the residual stress, specifically: normalize the load size corresponding to the load type with the highest sensitivity to residual stress at each parameter level, and then fit the linear relationship between each load size after normalization and the corresponding residual stress; Step 4, performing regression equation fitting on the load size and corresponding processing parameters under deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting processing modes, and establishing the regression equation relationship between the load size and the corresponding processing parameters under each processing mode; Step 5. Use the inverse problem solving method to determine the processing parameters of each processing method in the processing route of the combined processing technology of the mortise and tenon material, specifically: ① Preset the residual stress at different sampling depths below the processing surface of the mortise and tenon material; ② According to the preset residual stress, the total load size of the load type with the highest sensitivity to the residual stress is obtained through the processing process imprint characteristic relationship, and then the total load size is distributed to each processing technology with this load type in ascending order of the complexity of each processing technology in the mortise and tenon processing route; ③ If there is no total load size remaining after executing step ②, increase the preset value of the residual stress at different sampling depths below the processing surface of the mortise and tenon material, and return to step ② to execute, otherwise the remaining load size after distributing the total load size to each processing technology with this load type is then distributed to each processing technology without this load type in ascending order of the complexity of each processing technology in the mortise and tenon processing route; ④ If the total load size is not left after executing step ②, then increase the preset value of the residual stress at different sampling depths below the processing surface of the mortise and tenon material, and return to step ② to execute, otherwise the total load size is distributed to each processing technology with this load type, and then the remaining load size is distributed to each processing technology without this load type in ascending order of the complexity of each processing technology in the mortise and tenon processing route; After executing step ③, if there are processing processes in the mortise and tenon processing route that are not assigned to loads, the preset values of residual stress at different sampling depths below the processing surface of the mortise and tenon material are increased, and the process returns to step ② for execution, otherwise the assignment is terminated; the load size assigned to each processing process is the maximum load size calculated by the regression equation corresponding to the processing process according to the processing parameter range of the processing process, and the processing parameter value corresponding to the calculated maximum load size is the optimal processing parameter value of the processing process, thereby obtaining the optimal processing parameter value of each processing process in the mortise and tenon processing route; among them, the complexity of deep rolling, electric spark, electrolysis, cutting, electrochemical mechanical polishing and dealloying cutting gradually increases, the load type of the cutting processing method is mechanical-thermal coupling load, the load type of the electrochemical mechanical polishing processing method is mechanical-chemical coupling load, and the load type of the dealloying cutting processing method is mechanical-thermal-chemical coupling load.
2. The combined processing technology optimization method based on residual stress processing imprint according to claim 1 is characterized in that: The load size is characterized by the processing energy size, wherein the processing energy of the mechanical load is mechanical energy, the processing energy of the thermal load is thermal energy, the processing energy of the chemical load is chemical energy, the processing energy of the mechanical-thermal coupling load is the coupling energy composed of mechanical energy and thermal energy, the processing energy of the mechanical-chemical coupling load is the coupling energy composed of mechanical energy and chemical energy, and the processing energy of the mechanical-thermal-chemical coupling load is the coupling energy composed of mechanical energy, maximum thermal energy and chemical energy.
3. The combined processing technology optimization method based on residual stress processing imprint according to claim 1 is characterized in that: The processing parameters of the deep rolling processing method are rolling force, the processing parameters of the electrospark processing method are processing voltage, the processing parameters of the electrolytic processing method are electrolyte concentration, the processing parameters of the cutting processing method are cutting speed and cutting depth, the processing parameters of the electrochemical mechanical polishing processing method are electrolyte concentration and abrasive particle size, and the processing parameters of the dealloying cutting processing method are electrolyte concentration, cutting speed and cutting depth.
4. The combined processing technology optimization method based on residual stress processing imprint according to claim 1 is characterized in that: In the step 1, ABAQUS software is used to perform a simulation processing experiment, and residual stress data is collected through an ABAQUS subroutine.
5. The combined processing technology optimization method based on residual stress processing imprint according to claim 1 is characterized in that: The normalization process in step 3 adopts the Z-Score normalization formula, which is: Where x′ is the normalized load size, X is the load size before normalization, μ is the average load size corresponding to each parameter level before normalization, and σ is the standard deviation of the load size corresponding to each parameter level before normalization.
6. The combined processing technology optimization method based on residual stress processing imprint according to claim 1 is characterized in that: The process of fitting the linear relationship between each load size and the corresponding residual stress after normalization in step 3 is: (1) Draw a scatter plot of the normalized load magnitude and the corresponding residual stress; (2) Analyze the trend of the scatter plot and determine the form of the regression equation based on the trend of the scatter plot; (3) According to the determined regression equation form, the corresponding regression model is called to fit the normalized load magnitudes and the corresponding residual stresses, thereby obtaining the linear relationship between the normalized load magnitudes and the corresponding residual stresses.
7. The combined processing technology optimization method based on residual stress processing imprint according to claim 1 is characterized in that: By calculating the upper limit of relative error of deep rolling, electric spark or electrochemical machining in the collected depth distribution set, the feasibility of using the inverse problem solving method to determine the machining parameters is judged. The larger the upper limit of relative error, the lower the feasibility of using the inverse problem solving method to determine the machining parameters, and vice versa.
8. The combined processing technology optimization method based on residual stress processing imprint according to claim 7 is characterized in that: The expression of the upper limit of relative error is: In the formula, + r is the upper limit of relative error, z is the collection depth distribution set, I is the insensitivity matrix of the influence of load type on residual stress of deep rolling, electric spark or electrochemical machining, err is the absolute error of the residual stress value of the same collection depth of the tongue and groove material calculated by the relationship between the simulation machining experiment and the machining process imprint characteristics at the same parameter level of deep rolling, electric spark or electrochemical machining; Among them, the expression of the insensitivity matrix I is I=(S T S) -1 S T 。
