Machine tool adaptive control system based on digital twinning and implementation method thereof

Abstract

The invention relates to a machine tool adaptive control system based on digital twinning and an implementation method thereof, and the system comprises a numerical control machine tool, a data collection system, a digital twinning model, a main shaft load adaptive control module, and a man-machine interaction system. Real-time monitoring and self-adaptive regulation and control of the machining process are achieved by constructing a digital twin model of the numerical control machine tool and a main shaft load self-adaptive control module. According to the machine tool self-adaptive control system and the implementation method thereof, self-adaptive control of the numerical control machine tool can be realized, so that the machining precision, efficiency and stability of the numerical control machine tool are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a machine tool adaptive control system based on digital twin and an implementation method thereof. Background Art

[0002] With the rapid development of intelligent manufacturing technology, the application of digital twin technology in the manufacturing industry has become more and more extensive. Digital twin technology creates a virtual model of the physical machine tool, synchronizes and monitors the status, behavior and performance of the physical machine tool in real time, so as to achieve precise management and control of the production process. At present, the application of digital twin technology in the field of machine tools is mainly concentrated in machine tool status monitoring, fault diagnosis, production optimization and other aspects. However, there are still some problems with the existing technology, which limits its further application in machine tool adaptive control.

[0003] (1) Lack of unified standards and specifications

[0004] At present, the application of digital twin technology in the field of machine tools has not yet formed a unified standard and specification. Different research and application cases vary significantly in data collection, model construction, interaction methods, etc., resulting in poor interoperability and scalability of the system. At the same time, the development of digital twin models is often customized and lacks versatility, which limits its promotion and application in different types of machine tools and processing technologies.

[0005] (2) It is difficult to balance model accuracy and real-time performance

[0006] The existing digital twin technology models usually use refined physical models or complex data-driven models, which can provide high accuracy, but often require large amounts of calculations and are difficult to meet real-time requirements. Therefore, they cannot effectively respond to the dynamic changes of machine tools during the processing process, causing the prediction and optimization results of the digital twin model to lag behind the actual processing status, making it difficult to achieve real-time adaptive control of the processing process.

[0007] (3) Insufficient adaptive control capability

[0008] The current digital twin technology is mainly focused on machine tool status monitoring and production optimization, and lacks adaptive control capabilities. When unexpected conditions occur during the processing (such as changes in workpiece materials, tool wear, etc.), the existing digital twin system cannot make timely adjustments, resulting in the impact on machine tool processing accuracy and efficiency. This is mainly because the existing digital twin system lacks a closed-loop control mechanism based on feedback and cannot achieve real-time adaptive control in virtual-real interaction.

[0009] (4) The multi-source heterogeneity of data and models is difficult to integrate

[0010] There are many types of data involved in the machine tool processing process (such as vibration signals, temperature signals, force signals, etc.), and they come from different sensor devices. These data are often heterogeneous and complex. How to effectively integrate multi-source heterogeneous data and make full use of them in digital twin models is still a difficulty in existing technologies. At the same time, when combining data with physical models, existing technologies have problems such as slow model update speed and poor generalization ability of data-driven models, which cannot accurately reflect the actual situation of the processing process. Summary of the invention

[0011] The purpose of the present invention is to provide a machine tool adaptive control system based on digital twin and its implementation method, which can realize adaptive control of CNC machine tools, thereby improving the accuracy, efficiency and stability of CNC machine tool processing.

[0012] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a machine tool adaptive control system based on digital twin, including a CNC machine tool, a data acquisition system, a digital twin model, a spindle load adaptive control module and a human-computer interaction system. By constructing a digital twin model of the CNC machine tool and a spindle load adaptive control module, real-time monitoring and adaptive regulation of the machining process can be realized.

[0013] Furthermore, the CNC machine tool is the controlled object, and its basic structure includes a spindle, a tool, a workpiece and a machine bed. The spindle load and processing parameter data are derived from the CNC machine tool and collected in real time through a data acquisition system;

[0014] The data acquisition system includes multiple sensors of various types, a data processing unit, a database and a data transmission unit, which are used to collect real-time data of the CNC machine tool and store the collected data in the database. The digital twin model realizes communication by reading the data in the database;

[0015] The digital twin model consists of a physical model and a data-driven model; the physical model combines the structure and dynamic characteristics of the machine tool to reflect the physical behavior of the machine tool during the processing; the data-driven model uses a support vector machine algorithm based on the collected multi-source data to establish a mapping relationship between the spindle load and the processing parameters; combining the physical model and the data-driven model to form a real-time updated digital twin model to realize the monitoring and simulation of the processing process;

[0016] The spindle load adaptive control module monitors the spindle load and processing status in real time and uses a fuzzy adaptive control algorithm to adjust the feed speed of the machine tool; when it is detected that the spindle load exceeds the set threshold or abnormal fluctuation occurs, the spindle load adaptive control module adaptively adjusts the processing parameters through a feedback mechanism to ensure the stability of the processing process and the processing quality;

[0017] The human-machine interaction system includes a visual monitoring interface and a simulation panel, so as to view the real-time status of the machine tool and the cutting force prediction results of the machining process, as well as the adjustment process of the adaptive control through the visual monitoring interface.

[0018] Further, the sensors of the data acquisition system are installed at different positions on the CNC machine tool to collect the spindle current and voltage, temperature changes and related operation data during the machining process of the machine tool, and send the collected real-time data to the data processing unit, and the data acquisition device uploads the collected data to the database;

[0019] The digital twin model communicates with the database through the data transmission unit to obtain the operation data of the machine tool in real time, and uses the physical model and the data-driven model to map the machining process in real time; the digital twin model sends the real-time data and the prediction results of the cutting force to the spindle load adaptive control module;

[0020] When the spindle load fluctuates abnormally or exceeds the set threshold, the spindle load adaptive control module uses a fuzzy adaptive control algorithm to adaptively adjust the feed speed of the machine tool according to the feedback information of the digital twin model; the spindle load adaptive control module dynamically adjusts the processing parameters of the CNC machine tool, including feed speed, spindle speed, and cutting depth, according to the feedback information of the digital twin model and real-time collected data. The adjusted processing parameters act on the machine tool through the actuator to achieve adaptive control of the processing process;

[0021] The human-machine interaction system performs a multi-directional three-dimensional visualization display of the machine tool through a visualization monitoring interface, and uses Unity's WebGL technology to realize remote monitoring of the machine tool's operating status.

[0022] The present invention also provides a method for implementing the above-mentioned machine tool adaptive control system based on digital twin, comprising the following steps:

[0023] Step S1: constructing a geometric model of the CNC machine tool based on the geometric features of the CNC machine tool and the constraint relationship between each component;

[0024] Step S2: obtaining data of the machine tool spindle during cutting, preprocessing the obtained data and building an agent model;

[0025] Step S3: Building a data acquisition system for the machine tool operation status and current, voltage and temperature data;

[0026] Step S4: Establishing digital twin models of the spindle, tool, and workpiece;

[0027] Step S5: Establish a spindle load adaptive control module to adjust the machine tool feed speed in real time according to the real-time data of the machine tool and the digital twin model;

[0028] Step S6: Building a human-computer interaction system.

[0029] Furthermore, the implementation method of step S1 is:

[0030] First, use 3D scanning technology or CAD model import to obtain the geometric data of key machine tool components and workpieces to ensure that the geometric accuracy of the model is consistent with the actual machine tool;

[0031] Secondly, use 3D modeling software to build a 3D geometric model of the machine tool and workpiece; the model includes various components of the machine tool and can represent the contact point between the tool and the workpiece, the spindle position, and the workpiece position;

[0032] Finally, the 3D geometric model is converted into STP format and imported into the 3D rendering software. The 3D geometric model is lightweight, and materials are set for each component of the 3D geometric model for appearance rendering. At the same time, the parent-child relationship of each component is set according to the movement form of the machine tool and exported to FBX format.

[0033] Furthermore, the implementation method of step S2 is:

[0034] By performing dynamic simulation on the cutting process of the spindle or collecting machine tool operation data, the cutting force of the spindle at different cutting depths, feed speeds and spindle speeds is obtained, and the obtained data is divided into training data and verification data; in order to reduce the impact of different dimensions and speed up the training speed, the data is normalized:

[0035]

[0036] Then, a proxy model is constructed using a support vector machine model with RBF as the kernel function. The proxy model is trained using normalized data, and the hyperparameters of SVM are optimized using the K-means Optimizer (KO). After training, a proxy model for cutting force prediction is obtained.

[0037] Furthermore, the implementation method of step S3 is:

[0038] First, select the appropriate sensor and development method based on the specific situation of the data machine tool and the signals to be collected;

[0039] Secondly, according to the working principle of the sensor and the data collection requirements, the sensor is installed at the key position of the machine tool; after the installation is completed, the sensitivity, response speed and sampling frequency of the sensor are debugged to ensure the accuracy and real-time performance of data collection;

[0040] Finally, the data collected by the sensor is transmitted to the data processing unit; the data processing unit filters, denoises and extracts features of the collected data, and uploads the processed data to the database, and the database communicates with the digital twin model through the data transmission unit.

[0041] Furthermore, the implementation method of step S4 is:

[0042] Import the FBX file of the geometric model of the CNC machine tool generated in step S1 into the Unity3D software, develop scripts for the linear motion of the machine tool x, y, and z axes and the rotary motion of the tool based on the DOTween animation plug-in, optimize the tool cutting workpiece effect through dynamic mesh modification technology, and then add mesh colliders to the parts of the machine tool, and the collision detection method is continuous detection; at the same time, read the database data in real time to drive the movement of the digital twin model to realize the real-time mapping of the digital twin model to the physical machine tool; and the script retains the development and editing interface to facilitate the promotion and secondary development of the script;

[0043] Secondly, the code running on the machine tool is parsed through a script, the commands in the G code are converted into Unity coordinates, and the feed speed, spindle speed, and cutting depth data of the machine tool are obtained. Then, the agent model trained in step S2 is used to simulate the machine tool processing process and predict the cutting force of the spindle in real time, providing a basis for optimizing the machine tool processing process.

[0044] Furthermore, the implementation method of step S5 is:

[0045] A two-dimensional fuzzy controller is used to implement the fuzzy adaptive machining control algorithm; a fuzzy controller is constructed in the Fuzzy toolbox of MATLAB, in which the fuzzy sets of Ep, Cp and △u are (NB, NM, NS, 0, PS, PM, PB), the range of fuzzy subsets is (-3, -2, -1, 0, 1, 2, 3), and the membership functions are triangular function and trapezoidal function; after setting the fuzzy rule table, the Mamdani method is used for fuzzy reasoning, and the center of gravity method is used for anti-fuzzy operation; the predicted cutting torque is compared with the target torque, and the data of how much the feed speed can be changed to reach the target torque is obtained after the fuzzy controller, thereby achieving stability and efficiency of the machining process.

[0046] Furthermore, the implementation method of step S6 is:

[0047] By using the UI tools and XCharts plug-ins in Unity3D software, a human-computer interaction system is built to achieve three-dimensional visualization of the processing process and real-time display of operating data. At the same time, remote monitoring is achieved based on Unity WebGL technology.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] (1) Improved the stability and accuracy of the machining process

[0050] The present invention introduces spindle load adaptive control technology, which can monitor the spindle load in real time and dynamically adjust the feed speed according to the change of load. When the spindle load fluctuates or changes abnormally, the system can respond quickly, adjust the processing parameters to the optimal value, and maintain the stability of the spindle load, thereby effectively reducing the processing errors and workpiece surface defects caused by load fluctuations, and improving the stability and accuracy of the processing process.

[0051] (2) Improved system response speed and robustness

[0052] By introducing fuzzy adaptive control, the present invention can quickly respond to sudden changes in the spindle load during the machining process, dynamically adjust the machining parameters, and make the system more robust. In addition, by combining the proxy model with real-time feedback of actual machining data, the system can quickly correct the prediction error and ensure that the machining process is always in the optimal state.

[0053] (3) Improved processing efficiency and equipment utilization

[0054] The present invention can adaptively adjust the machining parameters and optimize the feed speed parameters according to the real-time situation of the machining state and the spindle load, thereby avoiding machine tool shutdown, tool wear or workpiece damage caused by improper setting of the machining parameters, thereby improving the overall machining efficiency and equipment utilization of the machine tool, extending the service life of the equipment and the tool, and reducing maintenance costs.

[0055] (4) Good human-computer interaction experience and intelligence level

[0056] The present invention provides an intuitive visualization interface and a human-machine interaction system, through which operators can view the operating status of the machine tool and realize remote monitoring, and has a high degree of intelligence and flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a block diagram of a machine tool adaptive control system based on digital twin according to an embodiment of the present invention;

[0058] Figure 2 It is a principle block diagram of the implementation of the machine tool adaptive control system based on digital twin according to an embodiment of the present invention;

[0059] Figure 3 It is a flow chart of a method for implementing a machine tool adaptive control system based on digital twins according to an embodiment of the present invention;

[0060] Figure 4 is a three-dimensional model diagram of a machine tool in an embodiment of the present invention;

[0061] Figure 5 is a three-dimensional rendering model diagram of a machine tool in an embodiment of the present invention;

[0062] Figure 6 is a flow chart of building an agent model in an embodiment of the present invention;

[0063] Figure 7 is an architecture diagram of a data acquisition system in an embodiment of the present invention;

[0064] Figure 8 This is an interactive interface diagram of a data acquisition system in an embodiment of the present invention;

[0065] Fig. 9 is a schematic diagram of the implementation principle of the fuzzy adaptive control algorithm in the embodiment of the present invention;

[0066] Fig.10 It is a visual monitoring interface diagram of the human-computer interaction system in an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0068] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0069] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0070] like Figure 1 As shown, this embodiment provides a machine tool adaptive control system based on digital twin, including a CNC machine tool, a data acquisition system, a digital twin model, a spindle load adaptive control module and a human-computer interaction system. By constructing a digital twin model of the CNC machine tool and a spindle load adaptive control module, real-time monitoring and adaptive regulation of the machining process are achieved.

[0071] (1) CNC machine tools

[0072] The CNC machine tool is the controlled object, including the basic structures such as the spindle, tool, workpiece and machine bed. The data such as spindle load change and processing parameters are all derived from the CNC machine tool and collected in real time through the data acquisition system.

[0073] (2) Data Acquisition System

[0074] The data acquisition system includes multiple sensors of various types, such as current sensors, voltage sensors, etc., as well as data processing units, databases, and data transmission units, which are used to collect real-time data of CNC machine tools (such as spindle load, temperature changes, etc.). The collected data is stored in the database, and the digital twin model realizes communication by reading the data in the database.

[0075] (3) Digital Twin Model

[0076] The digital twin model consists of a physical model and a data-driven model. The physical model combines the structure and dynamic characteristics of the machine tool to reflect the physical behavior of the machine tool during the processing. The data-driven model uses the support vector machine algorithm to establish the mapping relationship between the spindle load and the processing parameters based on the collected multi-source data. Combining the physical model and the data-driven model, a real-time updated digital twin model is formed to realize the monitoring and simulation of the processing process.

[0077] (4) Spindle load adaptive control module

[0078] The spindle load adaptive control module monitors the spindle load and processing status in real time and uses a fuzzy adaptive control algorithm to adjust the feed speed of the machine tool. When it is detected that the spindle load exceeds the set threshold or abnormal fluctuations occur, the spindle load adaptive control module uses a feedback mechanism to adaptively adjust the processing parameters to ensure the stability of the processing process and the processing quality.

[0079] (5) Human-computer interaction system

[0080] The human-machine interaction system includes a visual monitoring interface and a simulation panel. Operators can view the real-time status of the machine tool, the cutting force prediction results of the machining process, and the adjustment process of the adaptive control through the visual monitoring interface.

[0081] like Figure 2 As shown, the implementation principle of the machine tool adaptive control system in this embodiment is as follows:

[0082] The sensors of the data acquisition system are installed at different positions on the CNC machine tool to collect the spindle current and voltage, temperature changes and related operating data during the machine tool processing, and send the collected real-time data to the data processing unit. The data acquisition device uploads the collected data to the database.

[0083] The digital twin model communicates with the database through a data transmission unit to obtain the operation data of the machine tool in real time, and uses the physical model and the data-driven model to map the machining process in real time; the digital twin model sends the real-time data and the predicted results of the cutting force to the spindle load adaptive control module.

[0084] When the spindle load fluctuates abnormally or exceeds the set threshold, the spindle load adaptive control module uses a fuzzy adaptive control algorithm to adaptively adjust the feed speed of the machine tool based on the feedback information of the digital twin model; the spindle load adaptive control module dynamically adjusts the processing parameters of the CNC machine tool, including feed speed, spindle speed, and cutting depth, based on the feedback information of the digital twin model and real-time collected data. The adjusted processing parameters act on the machine tool through the actuator to achieve adaptive control of the processing process.

[0085] The human-machine interaction system performs a multi-directional three-dimensional visualization display of the machine tool through a visualization monitoring interface, and uses Unity's WebGL technology to realize remote monitoring of the machine tool's operating status.

[0086] like Figure 3 As shown, this embodiment also provides a method for implementing the above-mentioned machine tool adaptive control system based on digital twin, comprising the following steps:

[0087] Step S1: constructing a geometric model of the CNC machine tool based on the geometric features of the CNC machine tool and the constraint relationship between each component.

[0088] In this embodiment, the implementation method of step S1 is:

[0089] First, use 3D scanning technology or CAD model import to obtain the geometric data of key machine tool components (such as spindle, tool, worktable, etc.) and workpieces to ensure that the geometric accuracy of the model is consistent with the actual machine tool.

[0090] Secondly, use 3D modeling software (such as SolidWorks, UG, etc.) to build 3D geometric models of machine tools and workpieces. Figure 4 The model includes various components of the machine tool and can represent key points such as the contact point between the tool and the workpiece, the spindle position, and the workpiece position.

[0091] Finally, convert the 3D geometric model into STP format and import it into 3D rendering software (such as Blender, 3DMax, etc.). Figure 5 As shown, the 3D geometric model is lightweight, and materials are set for each component of the 3D geometric model to perform appearance rendering. At the same time, the parent-child relationship of each component is set according to the movement form of the machine tool and exported to FBX format.

[0092] Step S2: Acquire the data of the machine tool spindle during cutting, preprocess the acquired data and build a proxy model.

[0093] In this embodiment, the implementation method of step S2 is:

[0094] like Figure 6The figure shows the flow chart of agent model construction. By performing dynamic simulation on the cutting process of the spindle (such as Ansys, Deform and other software) or collecting machine tool operation data, the cutting force of the spindle at different cutting depths, feed speeds and spindle speeds is obtained, and the obtained data is divided into training data and verification data; in order to reduce the impact of different dimensions and speed up the training speed, the data is normalized:

[0095]

[0096] Then, a proxy model is constructed using a support vector machine model with RBF as the kernel function. The proxy model is trained using normalized data, and the hyperparameters of SVM are optimized using the K-means Optimizer (KO). After training, a proxy model for cutting force prediction is obtained.

[0097] Step S3: Build a data acquisition system for the machine tool operation status and current, voltage and temperature data.

[0098] In this embodiment, the implementation method of step S3 is:

[0099] First, select the corresponding sensor and development method according to the specific situation of the data machine tool and the signals to be collected.

[0100] Secondly, according to the working principle of the sensor and the data acquisition requirements, the sensor is installed at the key positions of the machine tool (such as the spindle, feed axis, etc.); after the installation is completed, the sensitivity, response speed and sampling frequency of the sensor are debugged to ensure the accuracy and real-time performance of data acquisition.

[0101] Finally, the data collected by the sensor is transmitted to the data processing unit; the data processing unit filters, denoises and extracts features of the collected data, and uploads the processed data to a database (such as MySQL, SQL Server, etc.), and the database communicates with the digital twin model through the data transmission unit.

[0102] In this embodiment, for the FANUC Series Oi-MF numerical control system, the architecture of the data acquisition system is as follows: Figure 7 As shown, the development steps are as follows:

[0103] First, secondary development is carried out through the Focas development kit provided by FANUC. The cnc_allclibhndl3 function is used to establish communication with the machine tool, and then the feed speed, spindle speed, running program and other operating data of the machine tool are obtained through cnc_acts, cnc_actf, pmc_rdpmcrng, cnc_rdspmeter and other functions.

[0104] Secondly, the OPC UA server is used to collect various sensor data to achieve unified collection of sensor data of different brands and types.

[0105] Finally, the collected machine tool data and sensor data are saved in the MySQL database, and the interactive interface of the data acquisition system is developed in VS2022, such as Figure 8 As shown, unified management and query of data can be achieved.

[0106] Step S4: Establish digital twin models of the spindle, tool, and workpiece.

[0107] In this embodiment, the implementation method of step S4 is:

[0108] Import the FBX file of the geometric model of the CNC machine tool generated in step S1 into the Unity3D software, develop the scripts for the linear motion of the machine tool x, y, and z axes and the rotary motion of the tool based on the DOTween animation plug-in, and optimize the tool cutting workpiece effect through dynamic mesh modification technology, and then add mesh colliders to the tools, workpieces and other parts of the twin machine tool, and the collision detection method is continuous detection. At the same time, read the database data in real time to drive the movement of the digital twin model to realize the real-time mapping of the digital twin model to the physical machine tool. And the script retains the development and editing interface to facilitate the promotion and secondary development of the script.

[0109] Secondly, by writing a script to parse the code running on the machine tool, the commands in the G code are converted into Unity coordinates, and the machine tool's feed speed, spindle speed, cutting depth and other data are obtained. Then, the agent model trained in step S2 can be used to simulate the machine tool processing process and predict the spindle cutting force in real time, providing a basis for optimizing the machine tool processing process.

[0110] Step S5: Establish a spindle load adaptive control module to adjust the machine tool feed speed in real time according to the real-time data and digital twin model of the machine tool.

[0111] According to the prediction results of the digital twin model and the real-time collected spindle load data, the target value and threshold range of the spindle load are set. When the spindle load deviates from the target value or exceeds the threshold range, the adaptive control strategy is triggered.

[0112] Select a suitable adaptive control algorithm (such as fuzzy adaptive control or PID control) to dynamically adjust machining parameters such as feed speed, cutting depth and spindle speed according to load changes to ensure that the spindle load is within a stable range.

[0113] In this embodiment, the implementation method of step S5 is:

[0114] Using a two-dimensional fuzzy controller, we can achieve Fig. 9The fuzzy adaptive machining control algorithm shown in the figure is constructed; a fuzzy controller is constructed in the Fuzzy toolbox of MATLAB, where the fuzzy sets of Ep, Cp and △u are (NB, NM, NS, 0, PS, PM, PB), the range of fuzzy subsets is (-3, -2, -1, 0, 1, 2, 3), and the membership functions are triangular function and trapezoidal function; after setting the fuzzy rule table, the Mamdani method is used for fuzzy reasoning, and the center of gravity method is used for anti-fuzzy operation; the predicted cutting torque is compared with the target torque, and the data of how much the feed speed can be changed to reach the target torque is obtained after the fuzzy controller, thereby achieving stability and efficiency of the machining process.

[0115] Step S6: Building a human-computer interaction system.

[0116] In this embodiment, the implementation method of step S6 is:

[0117] Using the UI tools and XCharts plug-in in Unity3D software, a human-computer interaction system is built. Its visual monitoring interface is as follows: Fig.10 As shown, 3D visualization of the machining process and real-time display of operating data are realized, and remote monitoring is achieved based on UnityWebGL technology.

[0118] Specifically, the UI tools in Unity3D software are used to build human-computer interaction functions, allowing operators to input processing parameters, adjust control strategies, set optimization goals through the interface, and transmit the input content to the adaptive control system to guide the adaptive regulation of the processing process.

[0119] Using the XCharts plug-in, the collected data and the prediction results of the twin model can be dynamically displayed in the form of charts, curves, dashboards, etc.

[0120] In order to facilitate the display and dynamic observation of the machine tool on the UI interface, dual cameras are set up in this embodiment. One camera is used to display the machine tool, and the camera can rotate or zoom the field of view around the machine tool, and then map the picture to the UI interface, so as to realize multi-angle viewing of the machine tool operation status; and the other camera is used to display the UI interface.

[0121] The use process of the machine tool adaptive control system based on digital twin provided in this embodiment is as follows:

[0122] 1. System initialization

[0123] (1) Check the hardware device connection

[0124] Check whether the connections of hardware devices such as machine tools, sensors, data acquisition cards and control modules are intact, and ensure that the connections of each device are stable and data transmission is normal.

[0125] (2) Start the data acquisition system

[0126] Turn on the power of the data acquisition system and start various sensors (such as vibration sensors, temperature sensors, etc.) to ensure that the sensors are in normal working condition and can collect real-time data.

[0127] (3) Start machine tool equipment and digital twin model

[0128] Start the actual machine tool equipment and check whether the machine tool's movement, spindle operation and other functions are normal. Start the digital twin model system and check whether the twin model can achieve data synchronization and real-time interaction with the physical machine tool.

[0129] (4) System parameter initialization and configuration

[0130] Configure the system initialization parameters in the human-computer interaction interface, including spindle speed, feed rate, cutting depth, tool type, workpiece material characteristics, etc. Set the target value and threshold range of the spindle load, and configure the target requirements and optimization goals of the machining task.

[0131] 2. Data collection and calibration

[0132] (1) Collecting initial state data

[0133] When the machine tool is unloaded and loaded, the spindle current, temperature and other status data are collected, and these data are preliminarily calibrated with the digital twin model to ensure that the model can accurately reflect the status of the actual machine tool.

[0134] (2) Calibrate twin model parameters

[0135] Based on the collected initial data, the parameters of the digital twin model are calibrated and adjusted.

[0136] (3) Verify data transmission

[0137] Check each data transmission line from the physical machine tool to the data acquisition system, and then to the twin model and control module to ensure the real-time and accuracy of data transmission.

[0138] 3. Start the digital twin model and spindle load adaptive control module

[0139] (1) Start the digital twin model

[0140] Start the digital twin model in the human-computer interaction interface and let the model enter the real-time update and state prediction mode. At this time, the digital twin model will continuously update its own state through the real-time data collected by the sensor and predict the spindle load during the processing.

[0141] (2) Start the spindle load adaptive control module

[0142] Start the spindle load adaptive control module and configure the control strategy according to the target value and threshold range of the spindle load and the processing workpiece.

[0143] 4. Setting and executing processing tasks

[0144] Start the actual machining task of the machine tool. At this time, the spindle load adaptive control module will dynamically adjust the machining parameters according to the real-time collected data.

[0145] 5. Real-time monitoring and human-computer interaction

[0146] (1) Real-time monitoring of processing status

[0147] On the human-computer interaction interface, various parameters of the processing status can be viewed in real time, including spindle load, temperature changes and other information.

[0148] (2) Abnormal alarm and processing

[0149] When the digital twin model predicts or the adaptive control system detects that key parameters such as spindle load exceed the set threshold, the system will issue an abnormal alarm signal.

[0150] 6. System maintenance and expansion

[0151] (1) Regular maintenance and calibration

[0152] Regularly maintain and calibrate hardware devices such as sensors, data acquisition cards, and control modules to ensure the accuracy and stability of system data acquisition. At the same time, check and update digital twin models and adaptive control modules to cope with the impact of processing technology changes and equipment aging.

[0153] (2) System expansion and function upgrade

[0154] Further expand and upgrade the system according to actual processing needs and technological development. For example, add new sensor types (such as stress sensors, thermal deformation monitoring sensors, etc.), optimize control algorithms (such as introducing adaptive neural network control, reinforcement learning, etc.), and integrate with other intelligent manufacturing systems (such as workshop management systems, manufacturing execution systems) to improve the overall intelligence level and application scenarios of the system.

[0155] The digital twin-based machine tool adaptive control system provided in this embodiment has a wide range of uses, including:

[0156] (1) Application in high-precision machining tasks

[0157] In high-precision machining tasks (such as aerospace parts, precision mold manufacturing, etc.), the workpiece materials are diverse, the machining accuracy requirements are high, and the spindle load and cutting force are prone to drastic fluctuations during the machining process. The system of the present invention can ensure the stability of the spindle load and cutting force through real-time monitoring and adaptive control, thereby achieving higher machining accuracy and surface quality.

[0158] (2) Application in intelligent manufacturing workshops and smart factories

[0159] The system of the present invention can be integrated with other equipment and systems (such as workshop management systems and manufacturing execution systems) of smart manufacturing workshops or smart factories to achieve global optimization control and management, providing effective support for the construction of smart manufacturing workshops and factories.

[0160] (3) Application in multi-material processing

[0161] In multi-material processing scenarios (such as aluminum alloy, titanium alloy, composite materials, etc.), due to the large differences in the mechanical properties of different materials, it is easy to cause drastic fluctuations in spindle load and cutting torque. Using the system of the present invention, the processing parameters can be adaptively adjusted according to the characteristics of different materials, realizing efficient control and management of the multi-material processing process.

[0162] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0163] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0164] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0166] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A machine tool adaptive control system based on digital twin, characterized in that: It includes CNC machine tools, data acquisition systems, digital twin models, spindle load adaptive control modules and human-computer interaction systems. By constructing the digital twin model of the CNC machine tools and the spindle load adaptive control module, real-time monitoring and adaptive regulation of the machining process can be achieved.

2. According to the digital twin-based machine tool adaptive control system of claim 1, it is characterized in that: The CNC machine tool is the controlled object, and its basic structure includes a spindle, a tool, a workpiece and a machine bed. The spindle load and processing parameter data are all derived from the CNC machine tool and collected in real time through a data acquisition system; The data acquisition system includes multiple sensors of various types, a data processing unit, a database and a data transmission unit, which are used to collect real-time data of the CNC machine tool and store the collected data in the database. The digital twin model realizes communication by reading the data in the database; The digital twin model consists of a physical model and a data-driven model; the physical model combines the structure and dynamic characteristics of the machine tool to reflect the physical behavior of the machine tool during the processing; the data-driven model uses a support vector machine algorithm based on the collected multi-source data to establish a mapping relationship between the spindle load and the processing parameters; combining the physical model and the data-driven model to form a real-time updated digital twin model to realize the monitoring and simulation of the processing process; The spindle load adaptive control module monitors the spindle load and processing status in real time and uses a fuzzy adaptive control algorithm to adjust the feed speed of the machine tool; when it is detected that the spindle load exceeds the set threshold or abnormal fluctuation occurs, the spindle load adaptive control module adaptively adjusts the processing parameters through a feedback mechanism to ensure the stability of the processing process and the processing quality; The human-machine interaction system includes a visual monitoring interface and a simulation panel, so as to view the real-time status of the machine tool and the cutting force prediction results of the machining process, as well as the adjustment process of the adaptive control through the visual monitoring interface.

3. The machine tool adaptive control system based on digital twin according to claim 2, characterized in that: The sensors of the data acquisition system are installed at different positions on the CNC machine tool to collect the spindle current voltage, temperature change and related operation data during the machining process of the machine tool, and send the collected real-time data to the data processing unit. The data acquisition device uploads the collected data to the database; The digital twin model communicates with the database through the data transmission unit to obtain the operation data of the machine tool in real time, and uses the physical model and the data-driven model to map the machining process in real time; the digital twin model sends the real-time data and the prediction results of the cutting force to the spindle load adaptive control module; When the spindle load fluctuates abnormally or exceeds the set threshold, the spindle load adaptive control module uses a fuzzy adaptive control algorithm to adaptively adjust the feed speed of the machine tool according to the feedback information of the digital twin model; the spindle load adaptive control module dynamically adjusts the processing parameters of the CNC machine tool, including feed speed, spindle speed, and cutting depth, according to the feedback information of the digital twin model and real-time collected data. The adjusted processing parameters act on the machine tool through the actuator to achieve adaptive control of the processing process; The human-machine interaction system performs a multi-directional three-dimensional visualization display of the machine tool through a visualization monitoring interface, and uses Unity's WebGL technology to realize remote monitoring of the machine tool's operating status.

4. A method for implementing a machine tool adaptive control system based on digital twin according to any one of claims 1 to 3, characterized in that: The following steps are involved: Step S1: constructing a geometric model of the CNC machine tool based on the geometric features of the CNC machine tool and the constraint relationship between each component; Step S2: obtaining data of the machine tool spindle during cutting, preprocessing the obtained data and building an agent model; Step S3: Building a data acquisition system for the machine tool operation status and current, voltage and temperature data; Step S4: Establishing digital twin models of the spindle, tool, and workpiece; Step S5: Establish a spindle load adaptive control module to adjust the machine tool feed speed in real time according to the real-time data of the machine tool and the digital twin model; Step S6: Building a human-computer interaction system.

5. The method for implementing a machine tool adaptive control system based on digital twin according to claim 4, characterized in that: The implementation method of step S1 is: First, use 3D scanning technology or CAD model import to obtain the geometric data of key machine tool components and workpieces to ensure that the geometric accuracy of the model is consistent with the actual machine tool; Secondly, use 3D modeling software to build a 3D geometric model of the machine tool and workpiece; the model includes various components of the machine tool and can represent the contact point between the tool and the workpiece, the spindle position, and the workpiece position; Finally, the 3D geometric model is converted into STP format and imported into the 3D rendering software. The 3D geometric model is lightweight, and materials are set for each component of the 3D geometric model for appearance rendering. At the same time, the parent-child relationship of each component is set according to the movement form of the machine tool and exported to FBX format.

6. The method for implementing a machine tool adaptive control system based on digital twin according to claim 4, characterized in that: The implementation method of step S2 is: By performing dynamic simulation on the cutting process of the spindle or collecting machine tool operation data, the cutting force of the spindle at different cutting depths, feed speeds and spindle speeds is obtained, and the obtained data is divided into training data and verification data; in order to reduce the impact of different dimensions and speed up the training speed, the data is normalized: Then, a surrogate model was constructed using a support vector machine model with RBF as the kernel function. The surrogate model was trained using normalized data, and the hyperparameters of the SVM were optimized using the K-means optimization algorithm. After training, a surrogate model for cutting force prediction was obtained.

7. The method for implementing a machine tool adaptive control system based on digital twin according to claim 4, characterized in that: The implementation method of step S3 is: First, select the appropriate sensor and development method based on the specific situation of the data machine tool and the signals to be collected; Secondly, the sensor is installed at the key position of the machine tool according to its working principle and data collection requirements; After installation, adjust the sensitivity, response speed and sampling frequency of the sensor to ensure the accuracy and real-time performance of data collection; Finally, the data collected by the sensor is transmitted to the data processing unit; the data processing unit filters, denoises and extracts features of the collected data, and uploads the processed data to the database, and the database communicates with the digital twin model through the data transmission unit.

8. The method for implementing a machine tool adaptive control system based on digital twin according to claim 4, characterized in that: The implementation method of step S4 is: Import the FBX file of the geometric model of the CNC machine tool generated in step S1 into the Unity3D software, develop scripts for the linear motion of the machine tool x, y, and z axes and the rotary motion of the tool based on the DOTween animation plug-in, optimize the tool cutting workpiece effect through dynamic mesh modification technology, and then add mesh colliders to the parts of the machine tool, and the collision detection method is continuous detection; at the same time, read the database data in real time to drive the movement of the digital twin model to realize the real-time mapping of the digital twin model to the physical machine tool; and the script retains the development and editing interface to facilitate the promotion and secondary development of the script; Secondly, the code running on the machine tool is parsed through a script, the commands in the G code are converted into Unity coordinates, and the feed speed, spindle speed, and cutting depth data of the machine tool are obtained. Then, the agent model trained in step S2 is used to simulate the machine tool processing process and predict the cutting force of the spindle in real time, providing a basis for optimizing the machine tool processing process.

9. The method for implementing a machine tool adaptive control system based on digital twin according to claim 4, characterized in that: The implementation method of step S5 is: A two-dimensional fuzzy controller is used to implement the fuzzy adaptive machining control algorithm. A fuzzy controller is constructed in the Fuzzy toolbox of MATLAB, where the fuzzy sets of Ep, Cp and △u are (NB, NM, NS, 0, PS, PM, PB), the range of fuzzy subsets is (-3, -2, -1, 0, 1, 2, 3), and the membership functions are triangular functions and trapezoidal functions. After setting the fuzzy rule table, the Mamdani method is used for fuzzy reasoning, and the center of gravity method is selected for anti-fuzzy operation; the predicted cutting torque is compared with the target torque, and the data of how much the feed speed should be changed to reach the target torque is obtained after the fuzzy controller, thereby achieving a stable and efficient machining process.

10. The method for implementing a machine tool adaptive control system based on digital twin according to claim 4, characterized in that: The implementation method of step S6 is: By using the UI tools and XCharts plug-ins in Unity3D software, a human-computer interaction system is built to achieve three-dimensional visualization of the processing process and real-time display of operating data. At the same time, remote monitoring is achieved based on Unity WebGL technology.

Images