A deep small hole machining method combining multi-objective optimization and adaptive control

Abstract

The present invention discloses a deep small hole machining method combining multi-objective optimization and adaptive control, which includes: designing a two-factor three-level orthogonal experiment table, conducting experiments according to process parameters, and statistically analyzing experimental data; collecting torque signals in real time, studying the characteristics of torque signals, and determining the torque change rate threshold θ during the safe machining process of deep small holes; establishing and verifying a response surface quadratic function model; optimizing the response surface quadratic function model to obtain the best combination of process parameters under constraint conditions; using the best combination of process parameters as the initial data for machining, driving the machine tool to start drilling, and monitoring the torque signal in real time during the machining process to adaptively generate control signals; when the torque change rate is greater than the preset change rate threshold θ, controlling the machine tool to retract the tool, chip removal and cooling, and continuing machining after cooling is completed until machining is completed. The present invention can machine deep small holes with high quality, high efficiency and stability, the machining process is stable and efficient, and the machining result has high precision.

Description

Technical Field

[0001] The present invention relates to the technical field of process parameter and machining process optimization, and specifically relates to a deep small hole machining method combining multi-objective optimization and adaptive control. Background Art

[0002] With the rapid development of the manufacturing industry, deep small hole drilling plays an increasingly important role in the field of mechanical manufacturing. At the same time, deep small hole structures are widely used in the core components of fields such as aerospace, precision instruments, and mechanical equipment. For example, the air film cooling holes of aero-engine blades, the drawing holes of wire drawing dies, the fine deep group holes on tire molds, and the nozzle needle valve pairs of engine fuel injectors all require deep small hole machining. In machining, holes with a machined hole diameter D ≤ 3mm and a ratio of hole depth L to hole diameter D, L / D ≥ 10, are generally called deep small hole machining.

[0003] Since drilling itself is a semi-closed machining process, and this closedness is particularly prominent during deep small hole drilling, problems such as difficult chip evacuation, poor tool rigidity, and low strength make the drilling process unstable, resulting in rapid tool wear and breakage, and damage to the workpiece. Therefore, during the deep small hole machining process, people often adopt conservative process parameters to increase the tool life and ensure the smoothness of the machining process. However, conservative process parameters will reduce the machining accuracy and efficiency, and cannot effectively respond to various intermediate states during the machining process.

[0004] The invention with the publication number CN118126772A proposes an optimized cold pressing extraction process for edible vegetable oil processing. However, the experimental model of this invention is a simple empirical model constructed based on experimental data, which only reflects the influence of current processing parameters on the response variable and fails to accurately describe the prediction of the machining process. In deep small hole machining, it is necessary to accurately predict the machining process to avoid tool breakage during machining and affect the overall machining. Therefore, this experimental model is only applicable to edible vegetable oil processing and cannot be applied to deep small hole machining. The invention with the publication number CN118246277A proposes a design method for a sheet stamping die based on multi-objective optimization technology, and analyzes based on a response surface regression model. However, on the one hand, this model is a multiple linear response surface regression model, while in deep small hole machining, the influence of factors on the response variable is non-linear, and this model does not introduce model errors, so the prediction of the machining process is not accurate enough. Therefore, this model cannot be applied to deep small hole machining. On the other hand, it needs to rely on a finite element simulation model, which has a large amount of calculation and low practicality. Summary of the Invention

[0005] Aiming at the problems in the existing deep small hole machining process, such as difficult chip removal, easy tool breakage, unstable machining process, low machining accuracy, and low machining efficiency, the purpose of the present invention is to provide a deep small hole machining method combining multi-objective optimization and adaptive control, which can perform high-quality, high-efficiency, and stable machining on deep small holes, with a smooth and efficient machining process and high machining accuracy.

[0006] To achieve the above technical objectives, the technical solution adopted by the present invention is as follows:

[0007] A deep small hole machining method combining multi-objective optimization and adaptive control, the deep small hole machining method includes the following steps:

[0008] S1: Design a two-factor three-level orthogonal experiment table, conduct experiments according to process parameters, and statistically analyze the experimental data; among them, the two factors are the spindle speed V c and the feed rate V f , the machining environment for each group of experiments is completely the same, and each group of experiments uses a brand-new tool of the same model;

[0009] S2: Real-time collect torque signals, study the characteristics of torque signals, and determine the torque change rate threshold θ during the safe machining process of deep small holes;

[0010] S3: Conduct response surface analysis based on the experimental data in step S1, establish a response surface quadratic function model and verify it;

[0011] S4: Combine the torque change rate threshold θ to optimize the response surface quadratic function model, and obtain the best process parameter combination under the constraint conditions;

[0012] S5: Use the best process parameter combination in step S4 as the initial data for machining, drive the machine tool to start drilling, and real-time monitor the torque signal during the machining process to adaptively generate control signals; when the torque change rate is greater than the preset torque change rate threshold θ, control the machine tool to retract the tool, chip removal, and cooling, and continue machining after cooling is completed until machining is completed.

[0013] Step S2 further includes:

[0014] Install a torque sensor on the tool spindle, and use the torque sensor to real-time collect torque signals;

[0015] Perform noise reduction and filtering processing on the collected torque signals, and organize the processed data into a torque-time relationship chart;

[0016] Observe the change of the torque change rate. When tool breakage occurs, calculate the torque change rate at this time, and determine the torque change rate threshold θ during the safe machining process of deep small holes according to the torque change rate at this time.

[0017] Step S3 further includes:

[0018] S31. Based on Design-Expert software, perform response surface analysis according to the experimental data in Step S1, and respectively establish response surface quadratic function models for response variables y 1 , y 2 and variables V c , V f ;

[0019]

[0020] wherein, y 1 represents the maximum torque change rate, y 2 represents the roundness, Vc represents the spindle speed, V f represents the feed rate, β i and Y i are model coefficients, i = 0, 1, 2, 3, 4, 5, ε 1 and ε 2 are model errors;

[0021] S32. Analyze the confidence level P value and lack of fit error of the established response surface quadratic function model, and verify the response surface quadratic function model.

[0022] Step S4 further includes:

[0023] Construct constraint conditions by integrating the machine tool specifications, tool parameters, torque slope threshold, and machining accuracy, and optimize the design of the response surface quadratic function model to obtain the optimal process parameter combination of spindle speed V c and feed rate Vf:

[0024] max y 1 , min y 2

[0025] s.t. V cmin ≤ V c ≤ V cmax

[0026] V fmin ≤ V f ≤ V fmax

[0027]

[0028] y 1 < θ

[0029] y 2 < 1.5μm

[0030] In the formula, V cmin and V cmaxare the minimum drilling speed and the maximum drilling speed of the machine tool respectively; V fmin and Vf max are the minimum feed speed and the maximum feed speed of the machine tool respectively; KT is the correction coefficient, C T is the tool life coefficient, and m and q are the influence indexes of the spindle speed V c and the feed speed V f on the tool life respectively.

[0031] Step S4 further includes:

[0032] S41. The drill bit quickly approaches the workpiece for drilling, the torque generated during the machining process is collected in real time, the torque change rate is calculated, and it is compared with the torque change rate threshold θ. The difference between the calculated torque change rate and the torque change rate threshold θ is used as a control signal, which is amplified by the driver and then drives the spindle servo motor; when the torque change rate is greater than the torque change rate threshold, it transfers to step S42;

[0033] S42. Automatically retract the tool for chip removal, and at the same time set a delay time t to cool the tool. After the cooling is completed, return to step S41 to continue the machining until the machining is completed.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] First, the deep small hole machining method combining multi-objective optimization and adaptive control of the present invention accurately determines the torque change rate threshold for deep small hole machining based on the time-domain analysis of the torque signal, providing data support for tool protection.

[0036] Second, the deep small hole machining method combining multi-objective optimization and adaptive control of the present invention, based on the response surface analysis principle, establishes a response surface quadratic function model of the maximum torque change rate, roundness, spindle speed and feed speed, providing an accurate mathematical model for optimization design. The genetic algorithm designed is used to adaptively optimize the deep small hole machining process parameters, and the optimized process parameters are used as the initial data for machining. This data can achieve high-quality, high-efficiency and stable machining of deep small holes.

[0037] Third, the deep small hole machining method combining multi-objective optimization and adaptive control of the present invention, through the adaptive control system, monitors the change of the spindle torque in real time. As long as the torque change rate is within the machining safety range, the optimized process parameters can be adopted for machining. Once the torque change rate exceeds the machining safety range, the adaptive control system will control the machine tool to retract the tool for chip removal and cool it. This system effectively protects the tool, improves the tool life, avoids damage to the workpiece caused by tool breakage, and at the same time realizes the intelligence of the machining process. Description of the Drawings

[0038] Figure 1It is the flow chart of the deep small hole machining method combining multi-objective optimization and adaptive control of the present invention;

[0039] Figure 2 It is the process parameter optimization flow chart of the deep small hole machining method combining multi-objective optimization and adaptive control of the present invention;

[0040] Figure 3 It is the optimization algorithm diagram of the deep small hole machining method combining multi-objective optimization and adaptive control of the present invention;

[0041] Figure 4 It is the working principle diagram of the adaptive control system of the deep small hole machining method combining multi-objective optimization and adaptive control of the present invention. Specific Embodiments

[0042] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0043] See Figure 1 , a deep small hole machining method combining multi-objective optimization and adaptive control, the deep small hole machining method includes the following steps:

[0044] S1: Design a two-factor three-level orthogonal experiment table, conduct experiments according to the process parameters, and statistically analyze the experimental data; among them, the two factors are the spindle speed V c and the feed rate V f , the machining environment of each group of experiments is exactly the same, and each group of experiments uses a brand-new tool of the same model;

[0045] S2: Real-time collect torque signals, study the characteristics of the torque signals, and determine the torque change rate threshold θ during the safe machining process of the deep small holes;

[0046] S3: Conduct response surface analysis based on the experimental data in step S1, establish a response surface quadratic function model and verify it;

[0047] S4: Combine the torque change rate threshold θ, optimize and design the response surface quadratic function model, and obtain the best process parameter combination under the constraint conditions;

[0048] S5: Use the best process parameter combination in step S4 as the initial data for machining, drive the machine tool to start drilling, and real-time monitor the torque signal during the machining process to adaptively generate control signals; when the torque change rate is greater than the preset torque change rate threshold, control the machine tool to retract the tool, chip removal and cooling, and continue machining after the cooling is completed until the machining is completed.

[0049] In step S1, the Box-Behnken central composite design method is used to design a two-factor and three-level orthogonal experiment. This method can obtain the mathematical model between process parameters and the target value with as few test times as possible and ensure the accuracy of the mathematical model. In this experiment, the spindle speed (V c ), which has a greater impact on drilling, and the feed rate (V f ) are used as the experimental factors for the two process parameters, and the maximum torque change rate (y 1 ) and roundness (y 2 ) are used as indicators for the experiment, and the experimental results are statistically analyzed.

[0050] In step S2, the torque signal is obtained by real-time acquisition using a torque sensor installed on the tool spindle. The specific process includes the following steps:

[0051] S21. Collect data: Real-time collect the torque signal and measure and record it through the sensor and data collector;

[0052] S22. Data processing: Denoise and filter the collected data, and organize and plot the collected data into a graph;

[0053] S23. Data analysis: Observe the change trend of torque over time. When the change rate of torque changes sharply, tool breakage occurs. Calculate the value when the change rate of torque starts to change sharply, and determine the torque change rate threshold θ for the safe machining process of deep small holes based on this value.

[0054] Step S3 is to perform response surface analysis according to the experimental results in step S1, perform regression analysis on the experimental data to obtain a regression equation, and use the response surface for analysis to obtain and analyze and verify the response surface quadratic function model. The specific process includes the following steps:

[0055] S31. Establish a model: Based on the Design-Expert software, perform response surface analysis according to the experimental data in step S1, and establish response surface quadratic function models of the response variables y 1 , y 2 and the variables V c , V f respectively;

[0056]

[0057] Among them, y 1 represents the maximum torque change rate, y 2 represents roundness, V c represents the spindle speed, V f represents the feed rate, β i and γ i are model coefficients, i = 0, 1, 2, 3, 4, 5, ε 1and ε 2 is the model error, which is obtained through the experimental data analysis in step S1.

[0058] S32. Conduct variance analysis: Analyze the confidence level P-value and lack of fit of the established model. The P-value represents the significance level of this model. The lower the P-value, the higher the accuracy of the established mathematical model. If the P-value of lack of fit is less than 0.05, it indicates a large difference and the fitted model equation is invalid. The larger the P-value (not exceeding 1), the better the effect of the fitted model equation. According to the comparison between the predicted values and the experimental measurement values of the regression model, verify the stability and reliability of the model.

[0059] In step S4, use the gamultiobj function in the MATLAB toolbox to perform multi-objective optimization to obtain the optimal process parameter combination under the constraint conditions, which is used as the initial data for processing to achieve pre-planning.

[0060] Such as Figure 2 and Figure 3 , perform multi-objective optimization design based on the genetic algorithm, where the optimization objective in the optimization process is the response surface quadratic function model established in step S3:

[0061]

[0062] The optimization objects are the spindle speed Vc and the feed rate Vf;

[0063] The constraint conditions are the machine tool specification constraints: To ensure the safety of drilling, the spindle speed and the feed rate should be kept within the parameter range allowed by the machine tool, as shown in the following formula:

[0064] V cmin ≤V c ≤V cmax

[0065] V fmin ≤V f ≤V fmax

[0066] In the formula, V cmin and V cmax are the minimum drilling speed and the maximum drilling speed of the machine tool respectively; V fmin and V fmax are the minimum feed rate and the maximum feed rate of the machine tool respectively.

[0067] Tool constraint: During machining, a higher spindle speed and a larger feed rate can improve machining efficiency, but the tool life will be reduced accordingly. Therefore, during optimization, it is necessary to ensure the tool life, as shown in the following formula:

[0068]

[0069] Among them, K T is the correction coefficient, C T is the tool life coefficient, and m and q are the influence indexes of the spindle speed V c and the feed rate V f on the tool life, respectively.

[0070] Torque constraint: During the machining process, to ensure that the tool is not severely damaged and the machining process proceeds safely and smoothly, the maximum torque change rate shall not exceed the torque change rate threshold θ set in step S2, as shown in the following formula:

[0071] y 1 < θ.

[0072] Roundness constraint: To ensure the optimized machining accuracy, the constraint conditions are set in combination with the actual production requirements, as shown in the following formula:

[0073] y 2 < 1.5μm.

[0074] According to the above optimization objectives, optimization objects, and constraint conditions, a genetic algorithm is designed to adaptively optimize the process parameters of deep small hole machining. The specific algorithm is as follows:

[0075] max y 1 ,min y 2

[0076] s.t. V cmin ≤ V c ≤ V cmax

[0077] V fmin ≤ V f ≤ V fmax

[0078]

[0079] y 1 < θ

[0080] y 2 < 1.5μm

[0081] Such as Figure 4, in step S5, an adaptive control system is used to control the machining process. The adaptive control system controls the spindle servo motor to work based on a torque sensor and an adaptive controller, mainly consisting of a machine tool, a tool, a torque sensor, a controller, a driver, a spindle servo motor, etc. The working principle of the system is as follows: The torque generated during the machining process is transmitted through the tool spindle to the torque sensor. The torque change rate is obtained through a calculation formula. The measured value is amplified by an amplifier and converted to the adaptive controller through an A / D converter for comparison with the torque change rate threshold θ set in step S2. The difference between the measured value and the threshold is used as a control signal, which is amplified by the driver to drive the spindle servo motor. When the torque change rate is too large, the system can automatically retract the tool for chip removal, and at the same time, a delay time t is set to cool the tool, and then continue machining. This cycle continues until the machining is completed. This system simulates manual operation, and its machining process is: the drill bit quickly approaches the workpiece → drilling → torque measurement → comparison with the torque change rate threshold → pushing the drill for chip removal → re-drilling until the hole machining is completed.

[0082] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript.

[0083] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions run by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0084] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the process inFigure 1 one or more processes and / or blocks Figure 1 the functions specified in one or more blocks.

[0085] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions running on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 one or more processes and / or blocks Figure 1 or more processes and / or the functions specified in one or more blocks.

[0086] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0087] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A deep small hole machining method combining multi-objective optimization with adaptive control, characterized in that: The deep small hole processing method comprises the following steps: S1: Design a two-factor three-level orthogonal experimental table, conduct experiments based on process parameters, and statistically analyze experimental data; among them, the two factors are spindle speed V c and feed speed V f ,The processing environment of each set of experiments is exactly the same, and each set of experiments uses a new tool of the same type; S2: Real-time acquisition of torque signals, study of torque signal characteristics, and determination of torque change rate threshold θ during deep small hole safe machining; S3: Perform response surface analysis based on the experimental data of step S1, establish and verify the response surface quadratic function model; S4: Combined with the torque change rate threshold θ, the response surface quadratic function model is optimized to obtain the best process parameter combination under the constraint conditions; S5: using the optimal process parameter combination in step S4 as the initial processing data, driving the machine tool to start drilling processing, monitoring the torque signal in real time during the processing, and adaptively generating a control signal; when the torque change rate is greater than the preset torque change rate threshold θ, controlling the machine tool to retract the tool, remove chips and cool, and continuing the processing after cooling is completed until the processing is completed; Step S3 further comprises: S31, based on the Design-Expert software, perform response surface analysis according to the experimental data of step S1, and establish the response variables y1, y2 and variable V respectively. c 、V f The response surface quadratic function model of Among them, y1 represents the maximum torque change rate, y2 represents the roundness, V c Indicates the spindle speed, V f Indicates feed speed, β i and γ i is the model coefficient, i=0,1,2,3,4,5, ε1 and ε2 are the model errors; S32. Analyze the confidence value P and lack-of-fit error Lack of fit of the established response surface quadratic function model, and verify the response surface quadratic function model.

2. The deep small hole machining method combining multi-objective optimization and adaptive control according to claim 1 is characterized in that: Step S2 further comprises: A torque sensor is installed on the tool spindle, and the torque sensor is used to collect torque signals in real time; The collected torque signal is subjected to noise reduction and filtering, and the processed data is organized and plotted into a torque-time relationship chart; The change of torque change rate is observed. When tool breakage occurs, the torque change rate at this time is calculated, and the torque change rate threshold θ of the deep small hole safe machining process is determined according to the torque change rate at this time.

3. The deep small hole machining method combining multi-objective optimization and adaptive control according to claim 1 is characterized in that: Step S4 further comprises: The constraints are constructed by integrating machine tool specifications, tool parameters, torque slope threshold and machining accuracy. The constraints include: V cmin ≤V c ≤V cmax In fmin ≤V f ≤V fmax y1<θ y2<1.5μm Where V cmin and V cmax are the minimum drilling speed and the maximum drilling speed of the machine tool respectively; V fmin and V fmax are the minimum feed speed and the maximum feed speed of the machine tool respectively; K T is the correction factor, C T is the tool life coefficient, m and q are the spindle speed V c and feed speed V f Impact index on tool life; Under the condition of satisfying the constraints, y1 is maximized and y2 is minimized, and the response surface quadratic function model is optimized to obtain the spindle speed V c and feed speed V f The best process parameter combination.

4. The deep small hole machining method combining multi-objective optimization and adaptive control according to claim 1 is characterized in that: Step S4 further comprises: S41, the drill bit quickly approaches the workpiece for drilling, collects the torque generated during the machining process in real time, calculates the torque change rate, compares it with the torque change rate threshold θ, and uses the difference between the calculated torque change rate and the torque change rate threshold θ as a control signal, which is amplified by the driver to drive the spindle servo motor; when the torque change rate is greater than the torque change rate threshold, proceed to step S42; S42, automatically retract the tool to remove chips, and set a delay time t to cool the tool. After cooling, return to step S41 to continue processing until the processing is completed.

Images