Dynamic cooperative control method and device for multi-axis composite machining of high-aspect-ratio parts

Abstract

The application discloses a dynamic cooperative control method and device for multi-axis composite machining of high-aspect-ratio parts, relates to the related technical field of part cooperative machining, and comprises the following steps: obtaining a multi-channel digital potentiometer array; matching a target multi-axis composite machining process and sending the target multi-axis composite machining process to a central controller to control a main shaft and multiple feed shafts to perform multi-axis composite machining; collecting machining data time sequences through a multi-modal sensor to obtain multi-modal machining interaction state characteristics; obtaining machining process control parameters, main shaft servo adjustment parameters and multiple feed shaft servo adjustment parameters; and performing dynamic cooperative control on the multi-axis composite machining process of the high-aspect-ratio part. The application solves the technical problems of control lag caused by the dramatic dynamic change of the cutting process, poor cooperation of the shafts, and insufficient machining precision and stability in the prior art, achieves precise cooperative dynamic control of the multi-axis composite machining of the high-aspect-ratio part, and significantly improves the machining precision, surface quality and production efficiency.

Description

Technical Field

[0001] This application relates to the technical field of collaborative machining of parts, specifically to a dynamic collaborative control method and device for multi-axis composite machining of parts with high aspect ratio. Background Technology

[0002] As modern manufacturing continues to develop towards higher precision, higher efficiency, and more complex structures, the demand for machining high aspect ratio parts, such as turbine blades in aerospace, miniature shaft parts in precision medical devices, and deep-hole structures in energy equipment, is increasing. These high aspect ratio parts typically exhibit complex structures, thin walls that are easily deformed, and significant machining challenges. Traditional multi-axis CNC machining relies on predefined machining paths and process parameters. However, in machining high aspect ratio parts, due to large material removal volumes, drastic changes in cutting forces, and significant vibration and thermal deformation, static control modes struggle to adapt to the dynamic changes during machining. This can easily lead to decreased machining accuracy, poor surface quality, and even tool damage. Particularly in machining deep cavities and micro-holes, the interaction between the tool and workpiece is complex and variable, making true collaborative optimization impossible with existing machining control systems. Furthermore, multi-axis servo control uses analog voltage interfaces for adjustment, and its response speed and accuracy are limited by hardware, making it difficult to achieve high-frequency, multi-channel real-time parameter adjustments. This, in turn, affects the dynamic response capability, control accuracy, and process adaptability of high aspect ratio parts in multi-axis composite machining.

[0003] Therefore, current technologies suffer from technical problems such as drastic dynamic changes in the cutting process leading to control lag, poor coordination of motion between axes, and insufficient machining accuracy and stability. Summary of the Invention

[0004] This application provides a dynamic collaborative control method and device for multi-axis composite machining of high aspect ratio parts, which solves the technical problems in the prior art, such as control lag caused by drastic dynamic changes in the cutting process, poor coordination of motion of each axis, and insufficient machining accuracy and stability. It achieves the technical effect of realizing precise collaborative dynamic control for multi-axis composite machining of high aspect ratio parts, and significantly improving machining accuracy, surface quality and production efficiency.

[0005] This application provides a dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts. The method includes: deploying multi-channel digital potentiometers at the analog adjustment interfaces of the spindle servo driver and multiple feed axis servo drivers to obtain a multi-channel digital potentiometer array; matching a target multi-axis composite machining process based on the machining target of the high aspect ratio part, and sending the matched target multi-axis composite machining process to a central controller to control the spindle and multiple feed axes to perform multi-axis composite machining; during the multi-axis composite machining process, acquiring machining data in timing using multi-modal sensors, and processing the acquired multi-modal machining data. The sequence is used for state recognition to obtain multimodal machining interaction state characteristics; based on the multimodal machining interaction state characteristics, dual control quantity recognition is performed to obtain machining process control parameters, spindle servo adjustment parameters, and multiple feed axis servo adjustment parameters; the machining process control parameters are transmitted to the machine tool CNC system through the central controller, and the spindle servo adjustment parameters and multiple feed axis servo adjustment parameters are converted into digital potentiometer codes, calibrated by table lookup, and then written in batches via SPI and distributed to the multi-channel digital potentiometer array to perform dynamic collaborative control of the multi-axis composite machining process of high aspect ratio parts.

[0006] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts also performs the following processing: wherein the multi-channel digital potentiometer array communicates with the central controller via an SPI bus; each digital potentiometer in the multi-channel digital potentiometer array is a non-volatile device with a resolution of 10 to 12 bits.

[0007] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts further performs the following processing: obtaining the machining target of the high aspect ratio part; using the machining target as an index, searching in the multi-axis composite machining library to obtain a set of candidate multi-axis composite machining processes; performing process reliability scoring on the set of candidate multi-axis composite machining processes respectively, and taking the candidate multi-axis composite machining process corresponding to the maximum score as the target multi-axis composite machining process.

[0008] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts further performs the following processing: in a preset acquisition window, the multi-modal sensor is used to acquire machining data in a time sequence to obtain a multi-modal machining data sequence; state recognition is performed on the multi-modal machining data sequence to obtain a multi-modal machining state feature sequence; the multi-modal machining state feature sequence is iteratively interacted with in chronological order to determine the multi-modal machining interaction state features.

[0009] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts further performs the following processing: iteratively interacting the first multi-modal machining state features and the second multi-modal machining state features in the multi-modal machining state feature sequence to determine the first-stage multi-modal machining interaction state features; iteratively interacting the third multi-modal machining state features in the multi-modal machining state feature sequence based on the first-stage multi-modal machining interaction state features to obtain the second-stage multi-modal machining interaction state features, and so on, to determine the multi-modal machining interaction state features.

[0010] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts further performs the following processing: performing inner product mapping on the first multimodal machining state features and the second multimodal machining state features to determine the first-stage adjacency state transition matrix; and using the first-stage adjacency state transition matrix to perform graph convolution interaction on the second multimodal machining state features to obtain the first-stage multimodal machining interaction state features.

[0011] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts further performs the following processing: transmitting the multi-modal machining interaction state features to a dual control quantity recognizer to obtain initial machining process control parameters, initial spindle servo adjustment parameters, and multiple initial feed axis servo adjustment parameters; coordinating and fine-tuning the initial machining process control parameters and initial spindle servo adjustment parameters in conjunction with the machining target to obtain machining process control parameters and spindle servo adjustment parameters; and guiding and coordinating fine-tuning the multiple initial feed axis servo adjustment parameters based on the spindle servo adjustment parameters to obtain multiple feed axis servo adjustment parameters.

[0012] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts further performs the following processing: machining simulation is performed on the initial machining process control parameters and initial spindle servo adjustment parameters, and the simulated deviation amount and simulated deviation direction are determined based on the machining target; using the simulated deviation direction as the fine-tuning direction and the simulated deviation amount as the fine-tuning target, the initial machining process control parameters and initial spindle servo adjustment parameters are coordinated and fine-tuned multiple times according to a preset step size to obtain the machining process control parameters and spindle servo adjustment parameters.

[0013] In a possible implementation, the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts also performs the following processing: the dual control quantity identifier includes a machining process control branch and a spindle servo adjustment branch connected in parallel.

[0014] This application also provides a dynamic collaborative control device for multi-axis composite machining of high aspect ratio parts. The device includes: a potentiometer array acquisition module, used to deploy multi-channel digital potentiometers at the analog adjustment interfaces of the spindle servo driver and multiple feed axis servo drivers respectively, to obtain a multi-channel digital potentiometer array; a multi-axis composite machining module, used to match target multi-axis composite machining processes based on the machining target of the high aspect ratio part, and send the matched target multi-axis composite machining processes to the central controller to control the spindle and multiple feed axes to perform multi-axis composite machining; and a status recognition module, used to acquire machining data timing through multi-modal sensors during multi-axis composite machining, and to analyze the acquired data. The system performs state recognition on the modal machining data sequence to obtain multimodal machining interaction state characteristics; the adjustment parameter acquisition module is used to perform dual control quantity recognition based on the multimodal machining interaction state characteristics to obtain machining process control parameters, spindle servo adjustment parameters, and multiple feed axis servo adjustment parameters; the dynamic collaborative control module is used to transmit the machining process control parameters to the machine tool CNC system through the central controller, and convert the spindle servo adjustment parameters and multiple feed axis servo adjustment parameters into digital potentiometer codes respectively, and after table lookup calibration, write them in batches through SPI and distribute them to the multi-channel digital potentiometer array to perform dynamic collaborative control of the multi-axis composite machining process of high aspect ratio parts.

[0015] This application proposes a dynamic collaborative control method and apparatus for multi-axis composite machining of high aspect ratio parts. The method involves obtaining a multi-channel digital potentiometer array; matching the target multi-axis composite machining process and sending it to a central controller to control the spindle and multiple feed axes for multi-axis composite machining; acquiring machining data timing through multi-modal sensors to obtain multi-modal machining interaction state characteristics; obtaining machining process control parameters, spindle servo adjustment parameters, and multiple feed axis servo adjustment parameters; and performing dynamic collaborative control of the multi-axis composite machining process for high aspect ratio parts. This solves the technical problems in existing technologies, such as drastic dynamic changes in the cutting process leading to control lag, poor coordination of axis movements, and insufficient machining accuracy and stability. It achieves precise collaborative dynamic control for multi-axis composite machining of high aspect ratio parts, significantly improving machining accuracy, surface quality, and production efficiency. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic flowchart of a dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the dynamic collaborative control device for multi-axis composite machining of high aspect ratio parts provided in the embodiments of this application.

[0019] Figure labeling: Potentiometer array acquisition module 10, multi-axis composite machining module 20, status recognition module 30, adjustment parameter acquisition module 40, dynamic collaborative control module 50. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or apparatuses. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0023] This application provides a dynamic collaborative control method for multi-axis composite machining of parts with high aspect ratios, such as... Figure 1 As shown, the method includes: Step S100: Multi-channel digital potentiometers are installed at the analog adjustment interfaces of the spindle servo driver and multiple feed axis servo drivers respectively to obtain a multi-channel digital potentiometer array.

[0024] Preferably, the analog adjustment interface is an analog voltage input port in the industrial servo drive that provides variable analog voltage for adjusting key parameters within the servo drive. Multi-channel digital potentiometers are deployed at the analog adjustment interfaces of the spindle servo drive and multiple feed axis servo drives. Each digital potentiometer is an integrated circuit, essentially a variable resistor whose resistance value can be controlled by digital signals. Each digital potentiometer chip integrates multiple independent potentiometers. Specifically, the external voltage input line of the analog adjustment interface requiring dynamic adjustment in each servo drive is disconnected, and one channel of the digital potentiometer is connected in series in the loop. That is, the two fixed terminals of this channel are connected to a reference voltage, and the output voltage of the sliding terminal is connected to the analog input port of the servo drive. When the central controller sends a digital code value to the digital potentiometer via the SPI bus, its internal resistance ratio changes, thereby outputting a corresponding, precise analog voltage at the sliding terminal. Multiple analog interfaces of the spindle servo drive and multiple analog interfaces of each feed axis servo drive are connected in series with digital potentiometers, ultimately forming a multi-channel digital potentiometer array. Rapid batch synchronous updates are achieved through a digitally programmable precision voltage source, thereby ensuring multi-axis collaboration.

[0025] Furthermore, step S100 also includes the multi-channel digital potentiometer array communicating with the central controller via an SPI bus; each digital potentiometer in the multi-channel digital potentiometer array is a non-volatile device with a resolution of 10 to 12 bits.

[0026] Preferably, the multi-channel digital potentiometer array communicates with the central controller via an SPI bus. The SPI bus is a high-speed, full-duplex, synchronous serial communication protocol containing four main signal lines: SCLK (synchronous clock, generated by the central controller); MOSI (master device central controller output, slave device digital potentiometer input); MISO (master device input, slave device output); and SS / CS (chip select signal, controlled by the central controller to determine the slave device for communication). Specifically, the central controller's SPI interface acts as the master device, and all digital potentiometer chips are connected in parallel on the same SCLK and MOSI buses. Each digital potentiometer chip has an independent chip select pin connected to a GPIO pin of the central controller. When the central controller needs to update the resistance value of a potentiometer, it pulls down the chip select signal of the corresponding potentiometer chip to wake it up. Through the MOSI line, synchronized with the SCLK clock, it sends a data frame containing the target address and control instructions, along with a digital code corresponding to the specific resistance value. The digital potentiometer chip receives and latches this data, adjusting its output resistance / voltage. Finally, the central controller pulls up the chip select signal to end the communication.

[0027] Preferably, each digital potentiometer in the multi-channel digital potentiometer array is a non-volatile device with a resolution of 10 to 12 bits, where resolution refers to the number of resistance or output voltage levels that the digital potentiometer can distinguish; 10 bits of resolution means there are 2 10 =1024 levels, 12-bit resolution means there are 2 12 =4096 levels, meaning that if the total resistance of the digital potentiometer is R, the minimum step value for adjusting the resistor of a 10-bit resolution non-volatile device is approximately R / 1024, and for a 12-bit device it is R / 4096. This ensures that the central controller can make fine and smooth adjustments to the servo parameters. Non-volatile devices mean that the stored resistance value code will not be lost after the device is powered off. This avoids possible operational abnormalities or safety issues caused by the servo driver parameters being reset to default values ​​after an unexpected power outage and restart, thus ensuring the reliability and stability of the system.

[0028] Step S200: Based on the machining target of the high aspect ratio part, match the target multi-axis composite machining process, and send the matched target multi-axis composite machining process to the central controller to control the spindle and multiple feed axes to perform multi-axis composite machining.

[0029] Step S200 further includes step S210, obtaining the machining target of the high aspect ratio part; step S220, using the machining target as an index, searching in the multi-axis composite machining library to obtain a set of candidate multi-axis composite machining processes; step S230, scoring the process reliability of the candidate multi-axis composite machining process set, and taking the candidate multi-axis composite machining process corresponding to the maximum score as the target multi-axis composite machining process.

[0030] Preferably, the machining target of the high aspect ratio part is obtained, which includes multiple specific searchable parameters, such as the aspect ratio of the final part, hole diameter tolerance, surface roughness, geometric accuracy, material information, quality and efficiency requirements, etc.; then, using the machining target as an index, a search is performed in the multi-axis composite machining database, which stores a large amount of historical multi-axis machining data. The index field of each data entry is the parameters included in the machining target, and the content field is the corresponding verified successful machining scheme, including specific toolpath, spindle speed, feed per tooth and other cutting parameters, tool type / geometric angle, and key servo control parameters. The machining target of the current part is matched with the index fields in the database to determine multiple composite machining schemes that meet the conditions, forming a candidate multi-axis composite machining process set; then... Each candidate multi-axis composite machining process in the candidate multi-axis composite machining process set is evaluated for process reliability. This evaluation comprehensively considers parameter matching degree, historical success rate, risk index, and efficiency index. A process reliability score is calculated for each candidate multi-axis composite machining process. Parameter matching degree is the degree of consistency between the parameters included in the evaluation scheme and the current machining target parameters; the better the consistency, the higher the score. Historical success rate is the pass rate of parts processed by previous schemes; the higher the pass rate, the higher the score. Risk index is whether the evaluation scheme includes limit cutting parameters; the lower the risk, the higher the score. Efficiency index is the scheme with shorter processing time while meeting quality requirements. Finally, the candidate multi-axis composite machining process with the highest score is selected as the target multi-axis composite machining process to ensure improved process planning efficiency.

[0031] Preferably, the matched target multi-axis composite machining process is sent to the central controller. The central controller receives and parses it into motion control commands, which include the tool's motion trajectory in space, such as a complex five-axis linkage interpolation path, process parameters such as spindle speed, feed rate, and depth of cut, auxiliary commands such as tool change and coolant switching, and preset servo gain and feedforward parameters for different machining stages. Then, the central controller controls the spindle and multiple feed axes to perform multi-axis composite machining. Specifically, the complex tool trajectory is decomposed into a sequence of position commands executed by each motion axis. At the same time, based on the parsed path and feed rate, real-time interpolation calculations are performed to generate accurate, time-stamped target position points for each feed axis to ensure coordinated movement of multiple axes and synthesis of the required tool path. The calculated position commands are then periodically sent to the servo drives of each feed axis and the spindle servo drive for multi-axis collaborative machining.

[0032] In step S300, during the multi-axis composite machining process, machining data time sequence is acquired through multi-modal sensors, and the acquired multi-modal machining data sequence is state-identified to obtain multi-modal machining interaction state characteristics.

[0033] Step S300 further includes step S310, in a preset acquisition window, using the multimodal sensor to acquire processing data in a time sequence to obtain a multimodal processing data sequence; step S320, performing state recognition on the multimodal processing data sequence to obtain a multimodal processing state feature sequence; step S330, iterating and interacting with the multimodal processing state feature sequence in chronological order to determine the multimodal processing interaction state features.

[0034] Preferably, in the multi-axis composite machining process, a preset acquisition window is set based on the machining cycle or the spindle rotation cycle. Within the preset acquisition window, multi-modal sensors are used to acquire machining data in a time sequence. That is, multiple different types of sensors are used to simultaneously acquire physical quantities of the multi-axis composite machining process. For example, accelerometers acquire vibration signals, force sensors acquire cutting force signals, acoustic emission sensors acquire high-frequency stress wave signals generated by material deformation and fracture, and current sensors acquire drive current signals of the spindle and feed axis motors. This results in multiple parallel data sequences with timestamps, thus determining the multi-modal machining data sequence. Then, state identification is performed on the multi-modal machining data sequence. That is, signal processing and feature extraction are performed on each type of sensor data sequence to determine key indicators that can characterize the current machining state. For example, a fast Fourier transform is performed on the vibration signal data to obtain the spectrum. Then, the amplitude of the dominant frequency and the presence of sidebands are identified from the spectrum. Next, the root mean square value and other time-domain and frequency-domain features of the vibration signal are calculated to obtain the vibration state feature sequence, thus obtaining the multi-modal machining state feature sequence. Each point is a feature set of different sensor data at the same time. Then, the multimodal processing state feature sequence is iteratively interacted in chronological order, that is, the evolution and mutual influence of each feature over time are analyzed based on neural networks or time series modeling. Finally, the fused feature is determined as the multimodal processing interaction state feature, which integrates a comprehensive state description of the dynamic evolution history within the entire preset time window.

[0035] Furthermore, step S330 also includes step S331, iteratively interacting the first multimodal processing state feature and the second multimodal processing state feature in the multimodal processing state feature sequence to determine the first-stage multimodal processing interaction state feature; step S332, iteratively interacting the third multimodal processing state feature in the multimodal processing state feature sequence based on the first-stage multimodal processing interaction state feature to obtain the second-stage multimodal processing interaction state feature, and so on, to determine the multimodal processing interaction state feature.

[0036] Preferably, graph convolution is used to iteratively interact the first and second multimodal processing state features. Specifically, the correlation between the first and second multimodal processing state features is calculated and quantified into an adjacency state transition matrix, which is then applied to the second multimodal processing state features to obtain enhanced first-stage multimodal processing interaction state features, including the temporal evolution information from the first to the second multimodal processing state features. Then, based on the first-stage multimodal processing interaction state features, the third multimodal processing state features in the multimodal processing state feature sequence are iteratively interacted, integrating them into the third multimodal processing state features to obtain second-stage multimodal processing interaction state features, including the first, second, and third multimodal processing state features and their temporal evolution information. This process continues, accumulating and encoding the dynamic evolution information of the processing state over time layer by layer into the final feature vector, ultimately determining the multimodal processing interaction state features.

[0037] Furthermore, step S331 also includes step A, performing inner product mapping on the first multimodal processing state features and the second multimodal processing state features to determine the first stage adjacency state transition matrix; and step B, using the first stage adjacency state transition matrix to perform graph convolution interaction on the second multimodal processing state features to obtain the first stage multimodal processing interaction state features.

[0038] Preferably, an inner product mapping is performed on the first and second multimodal processing state features, which measures the similarity between the sub-features contained in the two multimodal processing state features. A higher value indicates a higher similarity. This generates a first-stage adjacency state transition matrix to quantify the evolution of processing states. Each element in the matrix represents the influence weight or transition strength of a sub-feature of the first multimodal processing state feature on a sub-feature of the second multimodal processing state feature, characterizing the dynamic evolution pattern of the features. Then, the first-stage adjacency state transition matrix is ​​used to perform a graphical mapping on the second multimodal processing state features. Convolutional interaction, where graph convolution is a neural network layer that operates on graph-structured data. Specifically, the time points corresponding to the first and second multimodal processing state features are regarded as two nodes, and the connecting lines are directed edges. The weights of the edges are defined by the adjacency state transition matrix, that is, the new node feature = adjacency state transition matrix × original node feature × weight matrix. The weight matrix is ​​used to perform linear transformation in the feature space to generate state features that contain both current feature information and historical feature evolution information, and finally obtain the first-stage multimodal processing interaction state features.

[0039] Step S400: Based on the multimodal machining interaction state characteristics, perform dual control quantity identification to obtain machining process control parameters, spindle servo adjustment parameters, and multiple feed axis servo adjustment parameters.

[0040] Step S400 further includes step S410, transmitting the multimodal machining interaction state features to the dual control quantity recognizer to obtain initial machining process control parameters, initial spindle servo adjustment parameters, and multiple initial feed axis servo adjustment parameters; step S420, combining the machining target, performing coordinated fine-tuning on the initial machining process control parameters and initial spindle servo adjustment parameters to obtain machining process control parameters and spindle servo adjustment parameters; step S430, based on the spindle servo adjustment parameters, guiding coordinated fine-tuning on the multiple initial feed axis servo adjustment parameters to obtain multiple feed axis servo adjustment parameters.

[0041] Furthermore, step S400 also includes a dual control quantity identifier comprising a machining process control branch and a spindle servo adjustment branch connected in parallel.

[0042] Preferably, dual control quantity recognition is performed based on the characteristics of multimodal machining interaction states. The dual control quantity recognizer is a trained neural network, including a machining process control branch and a spindle servo adjustment branch connected in parallel. It can output a comprehensive adjustment scheme according to the current complex machining state. The machining process control branch and the spindle servo adjustment branch are calculated independently and simultaneously within the dual control quantity recognizer. The machining process control branch is used to adjust cutting process-related parameters according to the machining state, such as spindle speed setpoint, feed rate, and depth of cut. The spindle servo adjustment branch is used to adjust the dynamic response parameters inside the spindle servo driver, such as proportional gain, integral gain, and speed feedforward gain, and synchronously outputs the response parameters of the feed axis servo driver.

[0043] Preferably, the multimodal machining interaction state characteristics are transmitted to a dual control quantity recognizer for dual control quantity recognition, and two sets of preliminary control parameters are output simultaneously, including initial machining process control parameters, initial spindle servo adjustment parameters, and multiple initial feed axis servo adjustment parameters. The initial machining process control parameters may include suggested spindle speed correction, feed rate correction, etc. The initial spindle servo adjustment parameters and initial feed axis servo adjustment parameters are parameters sent to the digital potentiometer array to adjust the PID gain, feedforward gain, etc. inside the servo driver to change the dynamic response characteristics of the axis.

[0044] Preferably, the initial machining process control parameters and initial spindle servo adjustment parameters are finely tuned in conjunction with the machining target. This involves inputting the initial machining process control parameters and initial spindle servo adjustment parameters into a data-driven prediction model for machining simulation. The simulation results are then compared with the machining target to calculate the simulation deviation and its direction. Finally, the initial machining process control parameters and initial spindle servo adjustment parameters are adjusted alternately and simultaneously according to a preset step size, and machining simulation is performed again until the simulation prediction results meet the machining target. This process ensures that the process strategy and the dynamic performance of the spindle are optimally matched.

[0045] Preferably, the initial feed axis servo adjustment parameters are guided and fine-tuned based on the spindle servo adjustment parameters. Specifically, the initial feed axes are adjusted synchronously with the spindle servo adjustment parameters. According to the final determined spindle servo adjustment parameters, the initial servo parameters of each feed axis are fine-tuned according to their mapping relationship. For example, if the spindle is set to a high-rigidity mode, the feed axis also needs to increase its position loop gain accordingly to more closely follow the command trajectory and maintain synchronization with the spindle's heavy cutting. Conversely, if the spindle is in a vibration-suppressing compliant mode, the feed axis also needs to reduce its gain to avoid chattering caused by coupling with the spindle's vibration frequency. This achieves deep servo-level coordination between the spindle and feed axes, thereby avoiding machining errors or vibrations caused by dynamic response mismatch, and ultimately obtaining multiple feed axis servo adjustment parameters. The machining process control parameters, spindle servo adjustment parameters, and feed axis servo adjustment parameters output through collaborative fine-tuning serve as the control scheme for multi-axis composite machining.

[0046] Furthermore, step S420 also includes step S421, performing machining simulation on the initial machining process control parameters and initial spindle servo adjustment parameters, and determining the simulation deviation amount and simulation deviation direction based on the machining target; step S422, using the simulation deviation direction as the fine-tuning direction and the simulation deviation amount as the fine-tuning target, performing multiple coordinated fine-tuning adjustments on the initial machining process control parameters and initial spindle servo adjustment parameters according to a preset step size, to obtain the machining process control parameters and spindle servo adjustment parameters.

[0047] Preferably, the initial machining process control parameters and initial spindle servo adjustment parameters are simulated, that is, they are input into an internally preset machining process prediction model to simulate the dynamic behavior of the machine tool, the cutting process and the final machining result under the output initial machining parameters. For example, the surface roughness of the output part, the magnitude of the cutting force, whether chatter will occur, etc. are predicted. Then, the simulated machining result is compared with the machining target to obtain the simulation deviation, which represents the deviation between the simulated machining result and the machining target, and the direction of the simulation deviation is determined.

[0048] Preferably, the initial machining process control parameters and initial spindle servo adjustment parameters are fine-tuned multiple times according to a preset step size, with the simulated deviation direction as the fine-tuning direction and the simulated deviation amount as the fine-tuning target. The preset step size refers to the step size of each adjustment, such as fine-tuning with a small increment to avoid excessive adjustment leading to system instability. Specifically, based on the deviation direction and the step size, the initial machining process control parameters and initial spindle servo adjustment parameters are fine-tuned for the first time, for example, reducing the feed rate from 200 mm / min to 195 mm / min and increasing the spindle gain from 0.35 to 0.36. The fine-tuned parameter set is then input into the machining process prediction model to calculate the new deviation amount and direction. A second fine-tuning is then performed, and this process is iterated multiple times until the simulated machining result meets the machining target, at which point the process terminates, obtaining the final machining process control parameters and spindle servo adjustment parameters.

[0049] In step S500, the machining process control parameters are transmitted to the machine tool CNC system through the central controller, and the spindle servo adjustment parameters and multiple feed axis servo adjustment parameters are converted into digital potentiometer codes respectively. After table lookup calibration, they are written in batches via SPI and distributed to the multi-channel digital potentiometer array to perform dynamic collaborative control of the multi-axis composite machining process of high aspect ratio parts.

[0050] Preferably, the central controller sends the machining process control parameters as setpoints to the CNC controller of the machine tool CNC system via an industrial communication network or bus to change the position commands of each feed axis and the speed commands of the spindle. Then, the spindle servo adjustment parameters and the multiple feed axis servo adjustment parameters are converted into digital potentiometer code values. That is, based on the functional relationship between the code values ​​and the servo adjustment parameters, the spindle servo adjustment parameters and the multiple feed axis servo adjustment parameters are converted into corresponding digital code values. However, in practical applications, there may be a slight nonlinear error in the resistance-code value relationship of the digital potentiometer, or there may be a gain deviation in the potentiometer output to the driver interface of the entire circuit. Calibration is performed through a calibration lookup table, which stores the precise correspondence between the theoretical value and the actual measured output voltage of each digital potentiometer channel. The converted digital potentiometer code values ​​are corrected to updated code values ​​that can accurately generate the target voltage by matching the calibration lookup table, so as to ensure the accuracy of control. Finally, the update code values ​​are batch-written and distributed to the multi-channel digital potentiometer array via SPI. Specifically, the SPI master controller of the central controller first pulls down the chip select signal of the first digital potentiometer chip, sends the update code value via the MOSI line, sets the resistance value of each channel on that digital potentiometer chip, then pulls up its chip select, and immediately pulls down the chip select of the next digital potentiometer chip, continuing to send update code values. Because SPI communication is extremely fast and continuous, the update operations of all digital potentiometers take effect almost simultaneously, ensuring that the dynamic characteristics of all axes change in a coordinated manner. This enables dynamic coordinated control of the multi-axis composite machining process for high aspect ratio parts. The CNC machine tool spindle and multiple feed axes execute new machining paths with highly coordinated dynamic performance. Based on the real-time sensed state, macroscopic process and microscopic servo response are adjusted simultaneously and synchronously, thereby achieving high-precision dynamic control, improving machining accuracy, surface quality, and production efficiency.

[0051] In the above text, refer to Figure 1 A dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts according to embodiments of the present invention is described in detail. Next, reference will be made to... Figure 2 A dynamic collaborative control device for multi-axis composite machining of high aspect ratio parts is described according to an embodiment of the present invention.

[0052] The dynamic collaborative control device for multi-axis composite machining of high aspect ratio parts according to embodiments of the present invention solves the technical problems in the prior art, such as control lag caused by drastic dynamic changes in the cutting process, poor coordination of the motion of each axis, and insufficient machining accuracy and stability. It achieves precise collaborative dynamic control for multi-axis composite machining of high aspect ratio parts, significantly improving machining accuracy, surface quality, and production efficiency. Figure 2As shown, the dynamic collaborative control device for multi-axis composite machining of high aspect ratio parts includes: potentiometer array acquisition module 10, multi-axis composite machining module 20, status recognition module 30, adjustment parameter acquisition module 40, and dynamic collaborative control module 50.

[0053] The potentiometer array acquisition module 10 is used to deploy multi-channel digital potentiometers at the analog adjustment interfaces of the spindle servo drive and multiple feed axis servo drives to obtain a multi-channel digital potentiometer array; the multi-axis composite machining module 20 is used to match the machining target of high aspect ratio parts with target multi-axis composite machining processes, and send the matched target multi-axis composite machining processes to the central controller to control the spindle and multiple feed axes to perform multi-axis composite machining; the status recognition module 30 is used to acquire machining data timing sequence through multi-modal sensors during multi-axis composite machining, and to perform status recognition on the acquired multi-modal machining data sequence to obtain... Multimodal machining interaction state characteristics; adjustment parameter acquisition module 40, used to identify dual control quantities based on the multimodal machining interaction state characteristics, and obtain machining process control parameters, spindle servo adjustment parameters and multiple feed axis servo adjustment parameters; dynamic collaborative control module 50, used to transmit the machining process control parameters to the machine tool CNC system through the central controller, and convert the spindle servo adjustment parameters and multiple feed axis servo adjustment parameters into digital potentiometer codes respectively, and after table lookup calibration, write them in batches through SPI and distribute them to the multi-channel digital potentiometer array, so as to perform dynamic collaborative control of the multi-axis composite machining process of high aspect ratio parts.

[0054] The specific configuration of the potentiometer array acquisition module 10 will be described in detail below. The potentiometer array acquisition module 10 further includes: wherein the multi-channel digital potentiometer array communicates with the central controller via an SPI bus; each digital potentiometer in the multi-channel digital potentiometer array is a non-volatile device with a resolution of 10 to 12 bits.

[0055] The specific configuration of the multi-axis composite machining module 20 will be described in detail below. The multi-axis composite machining module 20 further includes: acquiring the machining target of a high aspect ratio part; using the machining target as an index, searching in a multi-axis composite machining library to obtain a set of candidate multi-axis composite machining processes; performing process reliability scoring on each of the candidate multi-axis composite machining processes, and using the candidate multi-axis composite machining process corresponding to the maximum score as the target multi-axis composite machining process.

[0056] The specific configuration of the state recognition module 30 will be described in detail below. The state recognition module 30 further includes: acquiring processing data in a time sequence using the multimodal sensor within a preset acquisition window to obtain a multimodal processing data sequence; performing state recognition on the multimodal processing data sequence to obtain a multimodal processing state feature sequence; and iteratively interacting with the multimodal processing state feature sequence in chronological order to determine the multimodal processing interaction state features.

[0057] The specific configuration of the state recognition module 30 will be described in detail below. The state recognition module 30 further includes: iteratively interacting with the first and second multimodal processing state features in the multimodal processing state feature sequence to determine a first-stage multimodal processing interaction state feature; iteratively interacting with the third multimodal processing state feature in the multimodal processing state feature sequence based on the first-stage multimodal processing interaction state feature to obtain a second-stage multimodal processing interaction state feature, and so on, to determine the multimodal processing interaction state feature.

[0058] The specific configuration of the state recognition module 30 will be described in detail below. The state recognition module 30 further includes: performing inner product mapping on the first multimodal processing state features and the second multimodal processing state features to determine the first-stage adjacency state transition matrix; and using the first-stage adjacency state transition matrix to perform graph convolution interaction on the second multimodal processing state features to obtain the first-stage multimodal processing interaction state features.

[0059] The specific configuration of the adjustment parameter acquisition module 40 will be described in detail below. The adjustment parameter acquisition module 40 further includes: transmitting the multimodal machining interaction state features to a dual control quantity recognizer to obtain initial machining process control parameters, initial spindle servo adjustment parameters, and multiple initial feed axis servo adjustment parameters; coordinating and fine-tuning the initial machining process control parameters and initial spindle servo adjustment parameters in conjunction with the machining target to obtain machining process control parameters and spindle servo adjustment parameters; and guiding and coordinating fine-tuning the multiple initial feed axis servo adjustment parameters based on the spindle servo adjustment parameters to obtain multiple feed axis servo adjustment parameters.

[0060] The specific configuration of the adjustment parameter acquisition module 40 will be described in detail below. The adjustment parameter acquisition module 40 further includes: performing machining simulation on the initial machining process control parameters and initial spindle servo adjustment parameters, and determining the simulation deviation amount and direction based on the machining target; using the simulation deviation direction as the fine-tuning direction and the simulation deviation amount as the fine-tuning target, and according to a preset step size, performing multiple coordinated fine-tuning adjustments on the initial machining process control parameters and initial spindle servo adjustment parameters to obtain the machining process control parameters and spindle servo adjustment parameters.

[0061] The specific configuration of the adjustment parameter acquisition module 40 will be described in detail below. The adjustment parameter acquisition module 40 further includes a dual control quantity recognizer comprising a machining process control branch and a spindle servo adjustment branch connected in parallel.

[0062] The dynamic collaborative control device for multi-axis composite machining of high aspect ratio parts provided in the embodiments of the present invention can execute the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0063] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.

[0064] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A dynamic collaborative control method for multi-axis composite machining of parts with high depth-to-diameter ratio, characterized in that, The method includes: Multi-channel digital potentiometers are installed at the analog adjustment interfaces of the spindle servo drive and multiple feed axis servo drives to obtain a multi-channel digital potentiometer array. Based on the machining target matching of high aspect ratio parts, the target multi-axis composite machining process is matched, and the matched target multi-axis composite machining process is sent to the central controller to control the spindle and multiple feed axes to perform multi-axis composite machining. In the multi-axis composite machining process, machining data is acquired in time sequence through multi-modal sensors, and the acquired multi-modal machining data sequence is identified to obtain the multi-modal machining interaction state characteristics. Based on the multimodal machining interaction state characteristics, dual control quantity identification is performed to obtain machining process control parameters, spindle servo adjustment parameters and multiple feed axis servo adjustment parameters; The central controller transmits the machining process control parameters to the machine tool CNC system, and converts the spindle servo adjustment parameters and multiple feed axis servo adjustment parameters into digital potentiometer codes. After table lookup calibration, the parameters are written in batches via SPI and distributed to the multi-channel digital potentiometer array to achieve dynamic and coordinated control of the multi-axis composite machining process of high aspect ratio parts.

2. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 1, characterized in that, in, The multi-channel digital potentiometer array communicates with the central controller via the SPI bus. Each digital potentiometer in the multi-channel digital potentiometer array is a non-volatile device with a resolution of 10 to 12 bits.

3. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 1, characterized in that, Multi-axis composite machining processes based on target matching for high aspect ratio parts include: Obtain the machining target for parts with a high aspect ratio; Using the processing target as an index, a search is performed in the multi-axis composite machining library to obtain a set of candidate multi-axis composite machining processes; The reliability scores of the candidate multi-axis composite machining processes are evaluated, and the candidate multi-axis composite machining process with the highest score is selected as the target multi-axis composite machining process.

4. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 1, characterized in that, In multi-axis composite machining, machining data is acquired in time sequence using multi-modal sensors, and the acquired multi-modal machining data sequences are subjected to state recognition to obtain multi-modal machining interaction state characteristics, including: Within a preset acquisition window, the multimodal sensor is used to acquire processing data in a time sequence to obtain a multimodal processing data sequence. The multimodal processing data sequence is subjected to state identification to obtain a multimodal processing state feature sequence; The multimodal processing state feature sequence is iteratively interacted with in chronological order to determine the multimodal processing interaction state features.

5. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 4, characterized in that, The multimodal processing state feature sequence is iteratively interacted with in chronological order to determine the multimodal processing interaction state features, including: The first multimodal processing state feature and the second multimodal processing state feature in the multimodal processing state feature sequence are iteratively interacted to determine the first stage multimodal processing interaction state features; Based on the first-stage multimodal processing interaction state features, the third multimodal processing state features in the multimodal processing state feature sequence are iteratively interacted to obtain the second-stage multimodal processing interaction state features, and so on, to determine the multimodal processing interaction state features.

6. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 5, characterized in that, Iterative interaction is performed on the first and second multimodal processing state features in the multimodal processing state feature sequence to determine the first-stage multimodal processing interaction state features, including: Perform inner product mapping on the first multimodal processing state features and the second multimodal processing state features to determine the first-stage adjacent state transition matrix; The first-stage multimodal processing state features are obtained by performing graph convolution interaction on the second multimodal processing state features using the first-stage adjacency state transition matrix.

7. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 1, characterized in that, Based on the multimodal machining interaction state characteristics, dual control quantity identification is performed to obtain machining process control parameters, spindle servo adjustment parameters, and multiple feed axis servo adjustment parameters, including: The multimodal machining interaction state features are transmitted to the dual control quantity recognizer to obtain the initial machining process control parameters, the initial spindle servo adjustment parameters, and multiple initial feed axis servo adjustment parameters. Based on the machining target, the initial machining process control parameters and the initial spindle servo adjustment parameters are finely adjusted in a coordinated manner to obtain the machining process control parameters and the spindle servo adjustment parameters. Based on the spindle servo adjustment parameters, the multiple initial feed axis servo adjustment parameters are guided and finely adjusted to obtain multiple feed axis servo adjustment parameters.

8. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 7, characterized in that, Based on the stated machining objective, the initial machining process control parameters and initial spindle servo adjustment parameters are coordinated and fine-tuned to obtain the machining process control parameters and spindle servo adjustment parameters, including: The initial machining process control parameters and initial spindle servo adjustment parameters are used to simulate machining, and the simulation deviation amount and simulation deviation direction are determined based on the machining target. Using the simulated deviation direction as the fine-tuning direction and the simulated deviation amount as the fine-tuning target, the initial machining process control parameters and the initial spindle servo adjustment parameters are finely adjusted multiple times according to a preset step size to obtain the machining process control parameters and the spindle servo adjustment parameters.

9. The dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in claim 7, characterized in that, The dual control quantity identifier includes a machining process control branch and a spindle servo adjustment branch connected in parallel.

10. A dynamic collaborative control device for multi-axis composite machining of parts with high depth-to-diameter ratio, characterized in that, The apparatus is used to implement the dynamic collaborative control method for multi-axis composite machining of high aspect ratio parts as described in any one of claims 1 to 9, the apparatus comprising: The potentiometer array acquisition module is used to deploy multi-channel digital potentiometers at the analog adjustment interfaces of the spindle servo drive and multiple feed axis servo drives to obtain a multi-channel digital potentiometer array. The multi-axis composite machining module is used to match the target multi-axis composite machining process based on the machining target of high aspect ratio parts, and send the matched target multi-axis composite machining process to the central controller to control the spindle and multiple feed axes to perform multi-axis composite machining. The state recognition module is used to acquire machining data in a time sequence through multi-modal sensors during multi-axis composite machining, and to identify the state of the acquired multi-modal machining data sequence to obtain the multi-modal machining interaction state characteristics. The adjustment parameter acquisition module is used to perform dual control quantity identification based on the multimodal machining interaction state characteristics to obtain machining process control parameters, spindle servo adjustment parameters and multiple feed axis servo adjustment parameters. The dynamic collaborative control module is used to transmit the machining process control parameters to the machine tool CNC system through the central controller, and convert the spindle servo adjustment parameters and multiple feed axis servo adjustment parameters into digital potentiometer codes respectively. After table lookup calibration, the parameters are written in batches via SPI and distributed to the multi-channel digital potentiometer array to perform dynamic collaborative control of the multi-axis composite machining process of high aspect ratio parts.

Images