An adaptive control system and method for complex structure machining process

Abstract

The application discloses a kind of adaptive control system and method for complex structural part machining process, the system includes stable parameter offline planning module, machine tool numerical control system reconstruction module, maximum threshold learning and storage module and machining parameter adaptive control module.The application is by implanting driver in the numerical control system of machine tool, and adding driver instruction in original G code program, the segment division of machining program is carried out, according to the data acquisition and machining parameter target setting of the machining process of divided program segment, the machining parameter regulation of each program segment is realized by fuzzy control method, so as to realize the adaptive regulation of machining parameter in the overall machining process of complex structural part.It has important practical significance to improve the intelligentization of complex structural part machining process, improve the machining efficiency of complex structural part, and ensure the machining quality of complex structural part.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, specifically to an adaptive control system and method for the processing of complex structural parts. Background Technology

[0002] With the development of modern industry, the demand for processing complex structural components is increasing, especially in the fields of aviation, aerospace, automotive, and precision machinery. These complex structural components are often characterized by complex structures, high processing precision, and diverse materials, posing a great challenge to traditional processing technologies.

[0003] Most existing machining methods rely on fixed machining parameters and preset machining paths, failing to adapt to real-time changes during the machining process. In current machining processes, technicians pre-determine machining parameters such as depth of cut, spindle speed, and feed rate through simulation or past machining experience, inputting these parameters into the G-code machining program before machining begins. During machining, the CNC machine tool follows the programmed G-code to process complex structural parts. However, during machining, due to the influence of various factors such as material hardness, cutting force, and cutting temperature, cutting parameters (such as feed rate and spindle speed) need continuous adjustment to maintain optimal machining results. Traditional machining methods often cannot capture these parameter changes in real time, leading to low machining efficiency and unstable machining quality. Therefore, monitoring the machining process of complex structural parts and adaptively controlling machining parameters is of great significance for achieving intelligent manufacturing, improving the machining efficiency of complex structural parts, and ensuring machining quality.

[0004] Existing machining parameter control technologies have achieved intelligent regulation of machining parameters to a certain extent. In the patent "An Adaptive Regulation Method for CNC Machine Tool Process Parameters Considering Tool State" (application number: CN202310077644.X), tool state prediction and spindle power prediction models are constructed respectively, and an adaptive regulation model for process parameters considering tool state is established based on these models, thus achieving adaptive regulation of machining parameters. In the patent "An Optimization Method for Adaptive Fuzzy Control Rules of CNC Machining Parameters" (application number: CN201310081486.1), the power bond graph method is used to optimize the fuzzy control rules of adaptive control in CNC machining, improving control performance and stability. In the patent "An Adaptive Control Method for Machine Tools Based on GA-BP Neural Network Algorithm" (application number: CN201910732917.3), a neural network algorithm is used to optimize and adaptively adjust feed rate and spindle speed in real time, effectively improving machining efficiency and quality.

[0005] Analysis of machining parameter control technology reveals that: (1) Existing machining parameter control technology has constructed various neural network models, and all of them require a large number of cross-experiments to improve the accuracy of the models. They are not suitable for complex structural parts with high added value and long processing time in the aerospace field; (2) Existing machining parameter control technology is mainly applicable to machining processes where the initial machining parameters remain unchanged. The parameter control target in the machining process is singular and does not consider the scenario where the original CNC program setting machining parameters change with the machining characteristics during the machining process of complex parts. The machining parameter target cannot change with the changes in machining characteristics, which means that existing machining parameter control technology cannot be applied to the machining process of complex parts where the setting machining parameters change with the machining characteristics. Summary of the Invention

[0006] To address the shortcomings of existing machining parameter control technologies, such as their singular control objectives, inability to adapt machining parameters to changes in machining characteristics, and unsuitability for complex structural components in the aerospace field with high added value and long machining times, this invention proposes an adaptive control system and method for the machining process of complex structural components. This method involves embedding a driver program into the machine tool's CNC system and adding driver program instructions to the original G-code program. The machining program is divided into segments, and data acquisition and machining parameter target setting are performed according to these segments. Fuzzy control is then used to adjust the machining parameters of each segment, thereby achieving adaptive control of machining parameters throughout the overall machining process of complex structural components. This has significant practical implications for improving the intelligence of complex structural component machining processes, increasing machining efficiency, and ensuring machining quality.

[0007] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows:

[0008] In a first aspect, the present invention discloses an adaptive control system for the machining process of complex structural parts, the system comprising a stable parameter offline planning module, a machine tool CNC system modification module, a maximum threshold learning and storage module, and a machining parameter adaptive control module;

[0009] The offline planning module for stable parameters is used to construct an offline simulation model of stable machining parameters for complex structural parts, and to simulate the machining process of complex structural parts based on the principle of constant cutting force. The original G-code program is written according to the stable machining parameters set in the simulation.

[0010] The machine tool CNC system modification module is used to implant driver programs into the machine tool CNC system and define the R parameters of the CNC system. The implanted driver programs are used to implement the segmentation of the original G code program during machining, the selection and activation of controlled machining parameters, and the switching control of adaptive control functions. The R parameters are used to store the initial machining parameters and establish the connection between the machine tool CNC system, the driver programs, and the original G code program.

[0011] The maximum threshold learning and storage module uses the driver program embedded in the CNC system of the machine tool to divide the original G code program into segments, collects data of the machining process according to the divided program segments, extracts the maximum peak value of the signal corresponding to each program segment after division, and stores the maximum peak value of the corresponding machining process in the database according to the program name and the segment number after division.

[0012] The adaptive control module for machining parameters uses a driver program embedded in the CNC system of the machine tool to divide the original G-code program into segments, select control parameters, activate the adaptive control function, retrieve the corresponding maximum peak value stored in the database according to the program name and segment number, collect data during the machining process, calculate and compare the relationship between the real-time acquired signal and the maximum peak value, and perform adaptive control of machining parameters based on the comparison result until the machining program ends.

[0013] Furthermore, the stable parameter offline planning module includes a model building unit, a material property definition unit, a model import unit, an initial cutting parameter setting unit, a simulation calculation unit, a machining parameter determination unit, and a G-code writing unit;

[0014] The model building unit is used to create a three-dimensional geometric model of the complex structural part to be machined and a geometric model of the tool used in actual machining.

[0015] The material property definition unit defines the material properties of complex structural components and cutting tools, respectively;

[0016] The model import unit imports the created geometric models of complex structural parts and cutting tools into the ABAQUS finite element analysis software. After meshing the model, boundary conditions are set according to the actual machining conditions of the complex structural parts.

[0017] The initial cutting parameter setting unit determines the machining parameters, including the initial cutting speed, feed rate, and depth of cut, for the complex structural parts to be machined based on previous machining experience or process manuals, and determines the target cutting force according to the principle of constant cutting force.

[0018] The simulation calculation unit starts the process simulation module of the ABAQUS finite element analysis software to perform simulation calculations of the cutting process and solve for the physical quantities of stress, strain and cutting force during the cutting process.

[0019] The machining parameter determination unit optimizes and adjusts the cutting parameters according to the changes in the cutting force using a genetic algorithm, so that the cutting force gradually approaches the target cutting force. When the cutting force is within a preset fluctuation range relative to the target cutting force, the cutting parameters at the corresponding moment are output as stable machining parameters.

[0020] The G-code writing unit writes the original G-code program based on the stable processing parameters output by the processing parameter determination unit.

[0021] Furthermore, the machine tool CNC system modification module implants a driver program into the user loop bar under the NC program bar of the machine tool CNC system. The implanted driver program includes two function programs: ADACTRL(a,b,c) and ADAOFF. ADACTRL(a,b,c) indicates the start of a new segment, and ADAOFF indicates the end of the adaptive control process. In ADACTRL(a,b,c), 'a' indicates whether adaptive feed is enabled, where a value of 0 indicates adaptive feed is disabled, and a value of 1 indicates adaptive feed is enabled; 'b' indicates whether adaptive spindle speed is enabled, where a value of 0 indicates adaptive spindle speed is disabled, and a value of 1 indicates adaptive spindle speed is enabled; 'c' indicates the segment number, where c is a natural number greater than or equal to 1.

[0022] Furthermore, the maximum threshold learning and storage module includes a segmentation unit, a processing data acquisition unit, and a peak extraction unit;

[0023] The segmentation unit traverses the original G-code program and determines whether the initial machining parameters, including the depth of cut, feed rate, and spindle speed, have changed. When the initial machining parameters change, a driver function instruction is added to divide the segments.

[0024] The processing data acquisition unit acquires processing data according to the predefined program segments, and performs noise reduction and filtering on the acquired processing data.

[0025] The peak extraction unit extracts the maximum peak value of the signal corresponding to each program segment after division, and stores the maximum peak value of the corresponding processing process in the database according to the program name and the segment number after division.

[0026] Furthermore, the adaptive control module for processing parameters includes a program modification unit, a processing control unit, a peak value comparison unit, and an adaptive control unit;

[0027] The program modification unit uses driver function instructions embedded in the machine tool CNC system to divide the original G-code program into segments, select control parameters, and activate the adaptive control function.

[0028] The machining control unit controls the machine tool CNC system to run the modified G-code program to process the complex structural parts being processed.

[0029] The peak comparison unit retrieves the maximum peak value corresponding to the control parameters stored in the database according to the program name and the segment number, and at the same time collects the real-time data of the control parameters during the processing, and calculates and compares the relationship between the real-time data of the control parameters and the maximum peak value of the control parameters in the database.

[0030] The adaptive control unit uses fuzzy control to adaptively adjust the control parameters based on the comparison results output by the peak comparison unit, so that the real-time acquired control parameters approach their corresponding maximum peak values ​​until the machining program ends and the machining control unit is called to terminate the machining process.

[0031] Furthermore, the processing parameter adaptive control module and the maximum threshold learning and storage module maintain consistency in the division position and number of segments of the original G-code program fragments.

[0032] Furthermore, the adaptive control unit multiplies the corresponding maximum peak value retrieved from the database by a coefficient and then compares it with the real-time data of the control parameters.

[0033] Secondly, this invention discloses an adaptive control method for the machining process of complex structural parts, characterized in that the method is executed based on the system described above; the method includes the following steps:

[0034] S1. Construct an offline simulation model of stable machining parameters for complex structural parts, and simulate the machining process of complex structural parts based on the principle of constant cutting force. Write the original G-code program according to the stable machining parameters set in the simulation.

[0035] S2, embed the driver program into the CNC system of the machine tool and define the R parameters of the CNC system; wherein, the embedded driver program is used to realize the segmentation of the original G code program during the machining process, the selection and activation of the controlled machining parameters, and the switching control of the adaptive control function; the R parameters are used to store the initial machining parameters and establish the connection between the CNC system of the machine tool, the driver program, and the original G code program;

[0036] S3. The driver program embedded in the CNC system of the machine tool is used to divide the original G code program into segments. Data acquisition of the machining process is carried out according to the divided program segments. The maximum peak value of the signal corresponding to each program segment is extracted and stored in the database according to the program name and the segment number.

[0037] S4. The driver program embedded in the CNC system of the machine tool is used to divide the original G-code program into segments, select control parameters, activate the adaptive control function, retrieve the corresponding maximum peak value stored in the database according to the program name and segment number, collect data of the machining process, calculate and compare the relationship between the real-time acquired signal and the maximum peak value, and perform adaptive control of the machining parameters based on the comparison result until the machining program ends.

[0038] Step S1 further includes:

[0039] S11, using UG software to create a three-dimensional geometric model of a complex structural component to be manufactured;

[0040] S12, Based on the type of tool used in actual machining, establish the geometric model of the tool, including the cutting edge shape, tool diameter, tool length and geometric parameters of the cutting edge;

[0041] S13 defines the material properties of the complex structural parts to be machined and the cutting tools, including elasticity model, Poisson's ratio, density, and yield strength.

[0042] S14. Import the created geometric models of the complex structural parts and tools to be machined into the ABAQUS finite element analysis software and mesh them.

[0043] S15, set boundary conditions according to the actual machining of complex structural parts. In the finite element analysis software, the clamping and positioning in the actual machining process is simulated by setting displacement constraints. At the same time, the contact conditions between the tool and the complex structural parts to be machined are defined, including the contact type and friction coefficient settings.

[0044] S16. Based on past machining experience or process manuals, determine the initial cutting speed, feed rate, and depth of cut machining parameters for the complex structural parts to be machined, and set them in the finite element analysis software.

[0045] S17. Based on the principle of constant cutting force, determine the target cutting force. The formula for calculating the cutting force is as follows:

[0046]

[0047] In the formula, a is the cutting force coefficient; p This refers to the depth of cut. For the cutting depth a p The index; f is the feed rate; f is the exponent of the feed rate; v is the cutting speed; It is the exponent of the cutting speed v;

[0048] S18, start the process simulation module of ABAQUS finite element analysis software to perform simulation calculations of the cutting process and solve for the physical quantities of stress, strain and cutting force during the cutting process;

[0049] S19. Based on the changes in cutting force, the cutting parameters are optimized and adjusted through a genetic algorithm to make the cutting force gradually approach the target cutting force. When the fluctuation range of the cutting force relative to the target cutting force is within ±5%, the cutting parameters at this time are taken as stable machining parameters.

[0050] Furthermore, in step S4, the process of adaptively adjusting the control parameters using fuzzy control to make the real-time acquired control parameters approach their corresponding maximum peak values ​​includes the following steps:

[0051] S41, based on the power sensor, acquires the spindle power signal in real time during the machining process;

[0052] S42, calculate the difference ΔE between the real-time acquired spindle power signal and the maximum peak threshold, and the rate of change of the power signal difference.

[0053] S43, for ΔE, The output control quantity U is fuzzified;

[0054] S44, Determine the fuzzy control rules;

[0055] S45, perform fuzzy synthesis operation to obtain the fuzzy relation matrix R:

[0056] R = U(E × EI) × U;

[0057] In the formula, × represents the membership degree operator in fuzzy mathematics theory;

[0058] Based on the membership tables of each change level of E and EI, the corresponding U is calculated according to the control rules, and then the corresponding control table is obtained by weighted average calculation.

[0059] S46, given inputs E and EI, the corresponding output controlled variable U is obtained by querying the control table and calculating the weighted average.

[0060] S47 defuzzifies the output controlled variable U by multiplying it with a given quantization factor, and outputs the changes of the controlled variable during the processing of complex parts.

[0061] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0062] First, the adaptive control system and method of the present invention for the machining process of complex structural parts can realize real-time monitoring and adaptive adjustment of process parameters during the machining process of complex structural parts. It has good applicability and high controllability. At the same time, it can provide visualized machining parameter change curves, which helps to understand the real-time machining process of complex structural parts, optimize machining efficiency, and record machining process data, which helps to process information statistics and process review analysis.

[0063] Secondly, the adaptive control system and method of the present invention for the machining process of complex structural parts achieves adaptive control of process parameters during the machining process of complex structural parts by embedding driver programs into the CNC system of the machine tool and adding driver program instructions to the original G-code program, dividing the machining program into segments, and collecting data and setting machining parameter targets according to the divided program segments.

[0064] Third, the adaptive control system and method of the present invention for the machining process of complex structural parts realizes the control of process parameters in the machining process of complex structural parts through fuzzy control method. Compared with traditional machining methods, this method is more scientific and reasonable, and can greatly improve machining efficiency and give full play to the performance of machine tools. Attached Figure Description

[0065] Figure 1 This is a schematic diagram illustrating the structural principle of the adaptive control system for complex structural parts processing according to the present invention.

[0066] Figure 2 A flowchart for obtaining stable processing parameters through software pre-simulation;

[0067] Figure 3 This is a schematic diagram illustrating the segmentation of the original G-code program in the maximum threshold learning and storage module during the actual processing of a certain part.

[0068] Figure 4 This is a schematic diagram illustrating the segmentation of the original G-code program in the adaptive control module for machining parameters during the actual machining process of a certain part.

[0069] Figure 5 This is a fuzzy control flowchart for the machining parameters of complex parts in the adaptive control module for machining parameters. Detailed Implementation

[0070] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0071] Example 1

[0072] like Figure 1As shown, the adaptive control system for complex structural parts machining in this embodiment includes a stable parameter offline planning module, a machine tool CNC system modification module, a maximum threshold learning and storage module, and a machining parameter adaptive control module.

[0073] In adaptive control systems for machining complex structural parts, the offline planning module for stable parameters constructs an offline simulation model of stable machining parameters for the complex structural part based on its machining characteristics, process requirements, material information, and tool information. The simulation is then performed based on the principle of constant cutting force to obtain stable machining parameters set through simulation. When the machining process for the workpiece is mature, initial stable machining parameters can be set based on past machining experience and machine tool performance.

[0074] Specifically, when machining a completely new complex structural component, it is necessary to construct an offline simulation model of the stable machining parameters of the complex structural component. Based on the principle of constant cutting force, the machining process of the complex structural component is simulated to obtain the stable machining parameters set through simulation. For example... Figure 2 As shown, the specific steps are as follows:

[0075] (1) Establish the geometric model of the workpiece to be processed: Use UG software to accurately create the three-dimensional geometric model of the complex structural parts, and ensure that the size, shape and features (holes, slots, curved surfaces) of the model are consistent with the actual workpiece. Pay attention to the integrity of the model during the modeling process and avoid assembly defects such as missing surfaces and overlapping surfaces.

[0076] (2) Cutting tool modeling: Based on the type of tool used in actual machining, establish the geometric model of the tool, which mainly includes the cutting edge shape, tool diameter, tool length and geometric parameters of the cutting edge.

[0077] (3) Material property definition: Define the material properties of the workpiece and the cutting tool respectively, mainly including elasticity model, Poisson's ratio, density, yield strength, etc.

[0078] (4) Model import and mesh generation: Import the created geometric models of the workpiece and tool into the ABAQUS finite element analysis software and perform mesh generation. During the mesh generation process, special attention should be paid to using a finer mesh in areas with drastic stress changes (such as near the machined surface, corners, etc.).

[0079] (5) Determine boundary conditions and loads: Set boundary conditions according to the actual machining of complex structural parts. In the finite element analysis software, simulate the clamping and positioning in the actual machining process by setting displacement constraints. At the same time, define the contact conditions between the tool and the complex structural parts to be machined, including the contact type and friction coefficient settings.

[0080] (6) Initial cutting parameter settings: Based on past machining experience or process manuals, determine the initial cutting speed, feed rate and depth of cut and other machining parameters for the complex structural parts to be machined, and set them in the finite element analysis software.

[0081] (7) Define the target value of cutting force: Based on the principle of constant cutting force, determine the target cutting force. The formula for calculating the cutting force is as follows:

[0082]

[0083] In the formula, This is the cutting force coefficient, which is related to various factors such as workpiece material, tool material, and cutting conditions; a p This refers to the depth of cut. For the cutting depth a p The index represents the degree of influence of the cutting depth on the cutting force, and is generally between 0.8 and 1.0; f is the feed rate; is the exponent of the feed rate f, with a value generally between 0.6 and 0.8, indicating that the feed rate has less influence on the cutting force than the depth of cut; v is the cutting speed. The index of cutting speed v is typically between -0.1 and 0.1, indicating that the cutting speed has a relatively small impact on the cutting force.

[0084] (8) Perform finite element simulation calculation: After the parameter settings are completed, start the process simulation module of the ABAQUS finite element analysis software to perform simulation calculation of the cutting process and solve the physical quantities such as stress, strain and cutting force in the cutting process.

[0085] (9) Determination of stable machining parameters: Based on the changes in cutting force, the cutting parameters are optimized and adjusted by genetic algorithm so that the cutting force gradually approaches the target cutting force. When the cutting force fluctuates less near the target cutting force (fluctuation range within ±5%), the cutting parameters at this time are the stable machining parameters.

[0086] The machine tool CNC system retrofit module includes embedding driver programs into the machine tool's CNC system and defining the R parameters of the CNC system. Specifically, driver programs are embedded into the "User Cycle" section under the "NC Program" column of the machine tool's CNC system. The embedded driver programs are user-defined functions used to implement the segmentation of the original G-code program during machining, the selection and activation of controlled machining parameters, and the on / off control of adaptive control functions. The embedded driver programs define R parameters to store machining parameters such as initial feed rate and spindle speed. By adding function instructions of the driver programs to the original G-code program, the connection between the machine tool CNC system, the driver programs, and the original G-code program is established.

[0087] For example, the implanted driver program includes two functions: ADACTRL(a,b,c) and ADAOFF. ADACTRL(a,b,c) indicates the start of a new segment, and ADAOFF indicates the end of the adaptive control process. In ADACTRL(a,b,c), 'a' indicates whether adaptive feed is enabled (0 for disabled, 1 for enabled); 'b' indicates whether adaptive spindle speed is enabled (0 for disabled, 1 for enabled); and 'c' indicates the segment number, starting from 1, increasing sequentially from 1 to 1, 2, 3, 4, 5...

[0088] like Figure 3 As shown, box 2 represents the R parameter instruction added to the original G code program, which is assigned the value R188 = 1000, thereby transmitting the original machining parameters set in the G code to the driver program embedded in the CNC system.

[0089] In the maximum threshold learning and storage module, the driver function instructions embedded in the CNC system are used to divide the original G-code program into segments. Data acquisition of the processing is carried out according to the segmented program segments. The acquired signals are processed by noise reduction, filtering and feature extraction. The maximum peak value of the signal corresponding to each segment is extracted and stored in the database according to the program name and the segment name.

[0090] like Figure 3 As shown, boxes 1, 3, and 4 represent the segmentation of the original G-code program, with parameters a and b both being 0, indicating that the adaptive control function for processing parameters is not activated and the program is in a learning state. c represents 1 and 2 respectively, indicating that the original G-code processing program has been divided into segment 1 and segment 2. Box 5 indicates the end of segmentation. Data acquisition is performed according to the segmented program segments, and the acquired signals undergo noise reduction, filtering, and feature extraction processing. The maximum peak value of the signal corresponding to each segment is extracted, and the maximum peak value of the corresponding processing process is stored in the database according to the program name and the segment name.

[0091] Preferably, the basis for segmenting the original G-code program by adding a driver function embedded in the CNC system is whether the initial machining parameters such as depth of cut, feed rate, and spindle speed have changed. If the initial machining parameters have changed, a driver function instruction is added to segment the program.

[0092] In the adaptive control module for machining parameters, the driver function instructions embedded in the CNC system are used to segment the original G-code program, select control parameters, and activate control switches, such as... Figure 4As shown, boxes 1, 3, and 4 divide the program into segments. In boxes 3 and 4, the parameter 'a' of the instruction ADACTRL(a,b,c) is always 1, indicating that the selected control parameter is the feed rate, activating the control switch. The parameters 'c' are 1 and 2 respectively, indicating that the original G-code machining program has been divided into segment 1 and segment 2. Box 5 indicates the end of the program segment division. Box 2 is used to establish the connection between the initial machining parameters in the G-code program and the driver program embedded in the CNC system. Based on the program name and the segment number, the corresponding maximum peak value stored in the database is retrieved. Data acquisition during the machining process is performed, and the relationship between the real-time acquired signal and the maximum peak value is calculated and compared. Based on this, adaptive control of the machining parameters is implemented until the machining program ends.

[0093] In this embodiment, in the adaptive control module for processing parameters, the division position and number of segments of the original G-code program fragments are consistent with the division position and number of segments of the G-code program in the corresponding maximum threshold learning and storage module. For example... Figure 3 and Figure 4 As shown, the segmentation positions and number of segments in both programs are consistent.

[0094] Example 2

[0095] like Figure 1 As shown in the figure, this embodiment discloses an adaptive control method for the machining process of complex structural parts, which specifically includes the following steps:

[0096] S1. Construct an offline simulation model of stable machining parameters for complex structural parts, and simulate the machining process of complex structural parts based on the principle of constant cutting force to obtain stable machining parameters.

[0097] S2. Based on the stable machining parameters obtained from simulation, write the original G-code machining program.

[0098] S3 is a driver program developed to implement the switching control of the original G-code program segmentation, selection and activation of controlled machining parameters, and adaptive control functions during the machining process, and then embedded into the CNC system of the machine tool.

[0099] S4 defines the R parameter, which is used to store machining parameters such as the initial feed rate and spindle speed.

[0100] S5, Modify the prepared original G-code program, and divide the original G-code program into segments.

[0101] S6 runs the modified G-code program to collect the spindle power signal in real time during the machining process.

[0102] S7. The acquired signal is subjected to noise reduction and filtering. Peak features are extracted from each segment according to the divided program segments to obtain the maximum peak threshold of each segment. Preferably, step S7 can be repeated multiple times for the same processing procedure. By acquiring the process signal multiple times for the same segment of the same processing program, and then performing noise reduction, filtering and feature extraction processing, a more accurate and better-adaptive maximum peak threshold can be obtained.

[0103] S8 stores the extracted maximum peak threshold in the database according to the program name and the segment name after division.

[0104] S9 uses the driver function instructions embedded in the CNC system to divide the original G-code program into segments, selects the controlled machining parameters, and activates the control switch.

[0105] S10, start machining, and collect the spindle power signal in real time during the machining process.

[0106] S11, automatically retrieve the corresponding maximum peak threshold stored in the database based on the program name and the segment number.

[0107] S12: Calculate and compare the real-time acquired machining process signal with the corresponding maximum peak threshold retrieved from the database. Based on fuzzy control, adjust the machining parameters in real-time to bring the real-time monitoring signal closer to the corresponding maximum peak threshold, ensuring that the real-time monitoring signal fluctuates around the maximum peak threshold. For example, the controlled machining parameter in step S12 can be the spindle speed or feed rate. Fuzzy control is used to adaptively control the machining parameters to ensure that the real-time monitoring signal fluctuates around the maximum peak threshold. Preferably, the peak threshold characteristics retrieved from the database can be multiplied by coefficients to appropriately increase the peak threshold, thereby maximizing the performance of the machine tool.

[0108] S13, step S12 continues until the machining program ends.

[0109] like Figure 5 As shown, the fuzzy control of the feed rate in the adaptive control module for machining parameters in step S12 includes the following steps:

[0110] S1201: Real-time acquisition of spindle power signal during machining based on power sensor;

[0111] S1202: Solve for the difference ΔE between the real-time acquired spindle power signal and the maximum peak threshold (spindle power signal difference) and the rate of change of the power signal difference.

[0112] S1203: For ΔE, The output control quantity U is fuzzified;

[0113] The monitored spindle power signal difference variation range was divided into 14 levels: {-6, -5, -4, -3, -2, -1, -0, +0, 1, 2, 3, 4, 5, 6}. The spindle power signal difference variation range was further divided into 8 fuzzy sets: negative large (NB), negative medium (NM), negative small (NS), negative zero (NO), positive zero (PO), positive small (PS), positive medium (PM), and positive large (PB). The resulting fuzzy table of spindle power signal difference is shown in Table 1.

[0114] Table 1

[0115]

[0116] Similarly, the rate of change of the spindle power signal difference is divided into 13 levels: {-6, -5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5, 6}. The rate of change of the spindle power signal difference is also divided into 7 fuzzy sets: negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB). The fuzzy table of the rate of change of the spindle power signal difference is shown in Table 2.

[0117] Table 2

[0118]

[0119]

[0120] Similarly, let “U” represent the variable of “output control quantity”, and divide it into 15 change levels: {-7, -6, -5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5, 6, 7}. Then divide it into 7 fuzzy sets: negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB). The fuzzy table of the output change quantity is shown in Table 3.

[0121] Table 3

[0122]

[0123] S1204: Determine the fuzzy control rules;

[0124] The fuzzy control rules of this invention are described using fuzzy conditional statements as follows:

[0125] (1)if(E=NB or NM)and(EI=NB or NM)then U=PB

[0126] (2)if(E=NB or NM)and(EI=NS or ZO)then U=PB

[0127] (3)if(E=NB or NM)and(EI=PS)then U=PM

[0128] (4)if(E=NB or NM)and(EI=PM or PB)then U=ZO

[0129] (5)if(E=NS)and(EI=NB or NM)then U=PM

[0130] (6)if(E=NS)and(EI=NS or ZO)then U=PM

[0131] (7)if(E=NS)and(EI=PS)then U=ZO

[0132] (8)if(E=NS)and(EI=PM or PB)then U=NS

[0133] (9)if(E=NO or PO)and(EI=NB or NM)then U=PM

[0134] (10)if(E=NO or PO)and(EI=NS)then U=PS

[0135] (11)if(E=NO or PO)and(EI=ZO)then U=ZO

[0136] (12)if(E=NO or PO)and(EI=PS)then U=NS

[0137] (13)if(E=NO or PO)and(EI=PM or PB)then U=NM

[0138] (14)if(E=PS)and(EI=NB or NM)then U=PS

[0139] (15)if(E=PS)and(EI=NS)then U=ZO

[0140] (16)if(E=PS)and(EI=ZO or PS)then U=NM

[0141] (17)if(E=PS)and(EI=PM or PB)then U=NM

[0142] (18)if(E=PM or PB)and(EI=NB or NM)then U=ZO

[0143] (19)if(E=PM or PB)and(EI=NS)then U=NM

[0144] (20)if(E=PM or PB)and(EI=ZO or PS)then U=NB

[0145] (21)if(E=PM or PB)and(EI=PM or PB)then U=NB

[0146] The above fuzzy conditional statements can be summarized as fuzzy control rules as shown in Table 4.

[0147] Table 4

[0148]

[0149] S1205: Perform fuzzy synthesis operation, calculate the fuzzy relation matrix, and generate a control table;

[0150] The fuzzy relation R can be obtained according to the fuzzy rules:

[0151] R = U(E × EI) × U

[0152] In the formula, × represents the membership degree operator in fuzzy mathematics theory.

[0153] After obtaining the fuzzy relation R, the corresponding U is obtained according to the membership tables of each change level of E and EI, and then the corresponding control table is obtained by weighted average calculation (see Table 5).

[0154] Table 5

[0155]

[0156]

[0157] S1206: Perform fuzzy decision on the output quantity based on the given input; after giving inputs E and EI, obtain the corresponding controlled output quantity U by querying the control table and calculating the weighted average.

[0158] S1207: By multiplying with a given quantization factor, the output controlled variable U is defuzzified, and the changes of the controlled variable during the processing of complex parts are output.

[0159] In step S12, the fuzzy control method for spindle speed in the adaptive control module for machining parameters is the same as the fuzzy control method for feed rate.

[0160] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take 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 this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0161] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, produce instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0162] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0163] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, causing a series of operational steps to be executed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that run on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0164] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0165] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. An adaptive control system for the machining process of complex structural parts, characterized in that, The system includes a stable parameter offline planning module, a machine tool CNC system modification module, a maximum threshold learning and storage module, and a machining parameter adaptive control module. The offline planning module for stable parameters is used to construct an offline simulation model of stable machining parameters for complex structural parts, and to simulate the machining process of complex structural parts based on the principle of constant cutting force. The original G-code program is written according to the stable machining parameters set in the simulation. The machine tool CNC system modification module is used to implant driver programs into the machine tool CNC system and define the R parameters of the CNC system. The implanted driver programs are used to implement the segmentation of the original G code program during machining, the selection and activation of controlled machining parameters, and the switching control of adaptive control functions. The R parameters are used to store the initial machining parameters and establish the connection between the machine tool CNC system, the driver programs, and the original G code program. The maximum threshold learning and storage module uses the driver program embedded in the CNC system of the machine tool to divide the original G code program into segments, collects data of the machining process according to the divided program segments, extracts the maximum peak value of the signal corresponding to each program segment after division, and stores the maximum peak value of the corresponding machining process in the database according to the program name and the segment number after division. The maximum threshold learning and storage module includes a segmentation unit, a processing data acquisition unit, and a peak extraction unit; The segmentation unit traverses the original G-code program and determines whether the initial machining parameters, including the depth of cut, feed rate, and spindle speed, have changed. When the initial machining parameters change, a driver function instruction is added to divide the segments. The processing data acquisition unit acquires processing data according to the predefined program segments, and performs noise reduction and filtering on the acquired processing data. The peak extraction unit extracts the maximum peak value of the signal corresponding to each program segment after division, and stores the maximum peak value of the corresponding processing process in the database according to the program name and the segment number after division. The adaptive control module for machining parameters uses a driver program embedded in the CNC system of the machine tool to divide the original G-code program into segments, select control parameters, activate the adaptive control function, retrieve the corresponding maximum peak value stored in the database according to the program name and segment number, collect data during the machining process, calculate and compare the relationship between the real-time acquired signal and the maximum peak value, and perform adaptive control of machining parameters based on the comparison result until the machining program ends.

2. The adaptive control system for complex structural component machining processes according to claim 1, characterized in that, The stable parameter offline planning module includes a model building unit, a material property definition unit, a model import unit, an initial cutting parameter setting unit, a simulation calculation unit, a machining parameter determination unit, and a G-code writing unit; The model building unit is used to create a three-dimensional geometric model of the complex structural part to be machined and a geometric model of the tool used in actual machining. The material property definition unit defines the material properties of complex structural components and cutting tools, respectively; The model import unit imports the created geometric models of complex structural parts and cutting tools into the ABAQUS finite element analysis software. After meshing the model, boundary conditions are set according to the actual machining conditions of the complex structural parts. The initial cutting parameter setting unit determines the machining parameters, including the initial cutting speed, feed rate, and depth of cut, for the complex structural parts to be machined based on previous machining experience or process manuals, and determines the target cutting force according to the principle of constant cutting force. The simulation calculation unit starts the process simulation module of the ABAQUS finite element analysis software to perform simulation calculations of the cutting process and solve for the physical quantities of stress, strain and cutting force during the cutting process. The machining parameter determination unit optimizes and adjusts the cutting parameters according to the changes in the cutting force using a genetic algorithm, so that the cutting force gradually approaches the target cutting force. When the cutting force is within a preset fluctuation range relative to the target cutting force, the cutting parameters at the corresponding moment are output as stable machining parameters. The G-code writing unit writes the original G-code program based on the stable processing parameters output by the processing parameter determination unit.

3. The adaptive control system for complex structural component machining processes according to claim 1, characterized in that, The machine tool CNC system modification module implants a driver program into the user loop bar under the NC program bar of the machine tool CNC system. The implanted driver program includes two function programs: ADACTRL(a,b,c) and ADAOFF. ADACTRL(a,b,c) indicates the start of a new segment, and ADAOFF indicates the end of the adaptive control process. In ADACTRL(a,b,c), 'a' indicates whether adaptive feed is enabled, where a value of 0 indicates adaptive feed is disabled, and a value of 1 indicates adaptive feed is enabled; 'b' indicates whether adaptive spindle speed is enabled, where a value of 0 indicates adaptive spindle speed is disabled, and a value of 1 indicates adaptive spindle speed is enabled; 'c' indicates the segment number, where c is a natural number greater than or equal to 1.

4. The adaptive control system for complex structural component machining processes according to claim 1, characterized in that, The adaptive control module for processing parameters includes a program modification unit, a processing control unit, a peak comparison unit, and an adaptive control unit. The program modification unit uses driver function instructions embedded in the machine tool CNC system to divide the original G-code program into segments, select control parameters, and activate the adaptive control function. The machining control unit controls the machine tool CNC system to run the modified G-code program to process the complex structural parts being processed. The peak comparison unit retrieves the maximum peak value corresponding to the control parameters stored in the database according to the program name and the segment number, and at the same time collects the real-time data of the control parameters during the processing, and calculates and compares the relationship between the real-time data of the control parameters and the maximum peak value of the control parameters in the database. The adaptive control unit uses fuzzy control to adaptively adjust the control parameters based on the comparison results output by the peak comparison unit, so that the real-time acquired control parameters approach their corresponding maximum peak values ​​until the machining program ends and the machining control unit is called to terminate the machining process.

5. The adaptive control system for complex structural component machining processes according to claim 1, characterized in that, The adaptive control module for processing parameters and the maximum threshold learning and storage module maintain consistency in the division position and number of segments of the original G-code program.

6. The adaptive control system for complex structural component machining processes according to claim 4, characterized in that, The adaptive control unit multiplies the corresponding maximum peak value retrieved from the database by a coefficient and then compares it with the real-time data of the control parameters.

7. An adaptive control method for the machining process of complex structural parts, characterized in that, The method is performed based on the system described in any one of claims 1-6; the method includes the following steps: S1. Construct an offline simulation model of stable machining parameters for complex structural parts, and simulate the machining process of complex structural parts based on the principle of constant cutting force. Write the original G-code program according to the stable machining parameters set in the simulation. S2, embed the driver program into the CNC system of the machine tool and define the R parameters of the CNC system; wherein, the embedded driver program is used to realize the segmentation of the original G code program during the machining process, the selection and activation of the controlled machining parameters, and the switching control of the adaptive control function; the R parameters are used to store the initial machining parameters and establish the connection between the CNC system of the machine tool, the driver program, and the original G code program; S3. The driver program embedded in the CNC system of the machine tool is used to divide the original G code program into segments. Data acquisition of the machining process is carried out according to the divided program segments. The maximum peak value of the signal corresponding to each program segment is extracted and stored in the database according to the program name and the segment number. S4. The driver program embedded in the CNC system of the machine tool is used to divide the original G-code program into segments, select control parameters, activate the adaptive control function, retrieve the corresponding maximum peak value stored in the database according to the program name and segment number, collect data of the machining process, calculate and compare the relationship between the real-time acquired signal and the maximum peak value, and perform adaptive control of the machining parameters based on the comparison result until the machining program ends.

8. The adaptive control method for machining complex structural parts according to claim 7, characterized in that, Step S1 further includes: S11, using UG software to create a three-dimensional geometric model of a complex structural component to be manufactured; S12, Based on the type of tool used in actual machining, establish the geometric model of the tool, including the cutting edge shape, tool diameter, tool length and geometric parameters of the cutting edge; S13 defines the material properties of the complex structural parts to be machined and the cutting tools, including elasticity model, Poisson's ratio, density, and yield strength. S14. Import the created geometric models of the complex structural parts and tools to be machined into the ABAQUS finite element analysis software and mesh them. S15, set boundary conditions according to the actual machining of complex structural parts. In the finite element analysis software, the clamping and positioning in the actual machining process is simulated by setting displacement constraints. At the same time, the contact conditions between the tool and the complex structural parts to be machined are defined, including the contact type and friction coefficient settings. S16. Based on past machining experience or process manuals, determine the initial cutting speed, feed rate, and depth of cut machining parameters for the complex structural parts to be machined, and set them in the finite element analysis software. S17. Based on the principle of constant cutting force, determine the target cutting force. The formula for calculating the cutting force is as follows: In the formula, This is the cutting force coefficient; This refers to the depth of cut. For cutting depth The index; f is the feed rate; f is the exponent of the feed rate; v is the cutting speed; It is the exponent of the cutting speed v; S18, start the process simulation module of ABAQUS finite element analysis software to perform simulation calculations of the cutting process and solve for the physical quantities of stress, strain and cutting force during the cutting process; S19. Based on the changes in cutting force, the cutting parameters are optimized and adjusted through a genetic algorithm to make the cutting force gradually approach the target cutting force. When the fluctuation range of the cutting force relative to the target cutting force is within ±5%, the cutting parameters at this time are taken as stable machining parameters.

9. The adaptive control method for machining complex structural parts according to claim 7, characterized in that, In step S4, the process of adaptively adjusting the control parameters using fuzzy control to make the real-time acquired control parameters approach their corresponding maximum peak values ​​includes the following steps: S41, based on the power sensor, acquires the spindle power signal in real time during the machining process; S42, calculate the difference between the real-time acquired spindle power signal and the maximum peak threshold. Rate of change of power signal difference ; S43, to , The output control quantity U is fuzzified; S44, Determine the fuzzy control rules; S45, perform fuzzy synthesis operation to obtain the fuzzy relation matrix R: R = U(E × EI) × U; In the formula, × represents the membership degree operator in fuzzy mathematics theory; Based on the membership tables of each change level of E and EI, the corresponding U is calculated according to the control rules, and then the corresponding control table is obtained by weighted average calculation. S46, given inputs E and EI, the corresponding output controlled variable U is obtained by querying the control table and calculating the weighted average. S47 defuzzifies the output controlled variable U by multiplying it with a given quantization factor, and outputs the changes of the controlled variable during the processing of complex parts.

Images