A digital twin control system for product processing quality

Abstract

The present invention belongs to the field of quality control of processed products, and discloses a digital twin control system for product processing quality. According to the working principle of the numerical control system and the control strategy of the feedback loop of each moving part of the machine tool, a quasi-static digital twin system that fully matches the actual processing system is established, and the real-time following errors of each moving part of the machine tool under different load fluctuations during the processing are extracted to meet the control needs of the product processing quality. A dynamic digital twin control system for product processing quality is established through point-to-point real-time mapping, thereby realizing real-time monitoring of product quality. Based on the established digital twin control system, processing parameters under stable following errors can be obtained to realize real-time control of product processing quality. The present invention extracts real-time following errors in the processing process through the digital twin system, which can reduce the construction cost of the digital twin system, and effectively realize real-time control of product processing quality and processing accuracy without the need for any external sensors.

Description

Technical Field

[0001] The present invention belongs to the field of quality control of processed products, and in particular relates to a digital twin real-time control system for complex surface processing quality. Background Art

[0002] Real-time and effective monitoring of product processing status and processing quality during the manufacturing process is of great significance to ensuring the quality of parts processing and improving production efficiency. Changes in production status often mean instability in the processing process, which in turn leads to additional costs such as tool wear and reduced workpiece quality. In the past, many methods for monitoring processing status have been proposed, but due to the complexity and time-varying nature of the processing process, traditional methods are not easy to achieve real-time and accurate monitoring.

[0003] In the past few decades, the concept of digital twins has been widely recognized in various industries and has greatly changed the industrial field. Through the combination of machine tool CNC systems and digital twins, it is possible to dynamically predict machine tool performance and form decision-making and optimization instructions in combination with condition monitoring, fault diagnosis, performance prediction, parameter optimization, etc. Based on the instruction data of the machine tool CNC system, intelligent interpolation, servo tuning, error compensation and other technologies are used to autonomously control the operation process of intelligent machine tools during automatic execution. This enables intelligent machine tools to have historical and real-time data, database models, and decision and execution data, and can adapt to changes in processing conditions and machine tool status. At present, condition monitoring, simulation and visualization constitute the main uses of digital twins in intelligent manufacturing. Product quality can be analyzed and improved through the application of digital twin technology in parts manufacturing.

[0004] However, the comprehensive modeling cost of digital twins is high. Walker et al. (2014) outlined the main challenges of developing digital twins for computer numerical control (CNC) machining. The results showed that the realization of digital twin systems requires a large amount of sensor measurement data, which requires the introduction of a large number of sensors in the manufacturing system to monitor the effective data in real time. Wang et al. (2014) showed in the paper "A tool wear prediction and monitoring method based on machining power signals" that although adding sensors can ensure accuracy, the cost is high and it is difficult to apply in industry. In addition, in order to process massive data and provide timely feedback, the machine tool CNC system needs to have extremely strong computing power, which inevitably increases the production cost of the entire manufacturing system. Traditional digital twin systems usually require multiple physical sensors to monitor various phenomena related to the system, such as temperature and vibration. However, Shen et al. (2014) believe that many traditional production systems are not equipped with these multi-physical sensors by default, so once they are built, it is difficult to transform them with built-in sensors. In addition, the digital twin system has high requirements for real-time interaction of data fusion. Kreuzer et al.'s paper "Artificial intelligence in digital twins—A systematic literature review" believes that real-time mapping is a key function of the digital twin system. However, Böttjer et al.'s paper "A review of unit level digital twin applications in the manufacturing industry" shows that current digital twin applications rarely pay attention to real-time mapping of digital twins. Since the state and performance of machine tools are constantly changing during operation, digital twin systems face great challenges in real-time prediction and optimization without synchronous mapping with physical entities or processes. Summary of the invention

[0005] The present invention proposes a digital twin control system for product processing quality. By extracting real-time following errors in the processing process through the digital twin system, real-time control of product processing quality and processing accuracy can be effectively achieved.

[0006] The present invention aims to overcome the technical defects and shortcomings of the existing digital twin system for product processing quality, which requires a large number of sensors and has a high modeling cost. The accuracy of the machine tool will change under load, and the load acting on the motion system will cause a large deviation between the real-time position of the motion system and the command position, thereby affecting the processing accuracy of the machine tool and reducing the processing quality of the product. The present invention directly takes product processing quality as the control target and constructs a digital twin control system for product quality. By extracting the real-time following error in the processing process through the digital twin system, the construction cost of the digital twin system can be reduced, and real-time control of product processing quality and processing accuracy can be effectively achieved.

[0007] The technical solution of the present invention:

[0008] A digital twin control system for product processing quality, based on the working principle of the machine tool CNC system and the control strategy of the feedback loop of each moving part of the machine tool, establishes a quasi-static digital twin system that fully matches the actual processing system. On this basis, facing the control needs of product processing quality, the real-time following error of each moving part of the machine tool under different load fluctuations during the processing process is extracted, and a dynamic digital twin control system for product processing quality is established through point-to-point real-time mapping, thereby realizing real-time monitoring of product quality. Based on the established digital twin control system, the processing parameters under stable following error can be obtained to realize real-time control of product processing quality.

[0009] The specific steps are as follows:

[0010] Step 1: Build a digital twin control system for product processing quality

[0011] Taking product quality as the control target, considering that in the actual processing process, the workpiece material interferes with the grinding wheel tool system, which in turn causes material deformation and ultimately material removal, a digital twin control system is constructed based on the control principle of machine tool moving parts, including physical entities, digital twin data, virtual entities, services, and links between various components. The physical entities are the moving entities of the CNC grinder, and a digital twin control system for product quality is constructed based on this.

[0012] The physical entities of the corresponding moving parts of the grinding wheel and the workpiece are analyzed. In the CNC grinding machine, the grinding wheel is connected to the machine tool guide rail through the grinding wheel shaft, forming a "grinding wheel-grinding wheel shaft-CNC machine tool X-axis" connection component system; the grinding wheel is connected to the machine tool guide rail through the workpiece spindle or C-axis, forming a "workpiece-workpiece spindle / C-axis-CNC machine tool Z-axis" connection component system; the above two subsystems are connected together through the bed, distributed in the CNC machine tool coordinate system, and move and realize material removal through command signals in the machine tool CNC system, thereby forming the physical entity of the CNC grinding machine;

[0013] The virtual entity is the implementation mapping of the physical entity. Considering that the physical entity of the CNC grinder moves through the command signal in the CNC system of the machine tool: the CNC grinder sends the processing requirements of the product to the moving parts of the machine tool in the form of CNC code. The moving parts of the machine tool send control command signals according to the processing requirements. The corresponding servo driver converts the control command signals into power signals of the corresponding moving parts of the machine tool and drives the corresponding moving parts of the machine tool to move. After the moving parts of the machine tool move, the real-time position of the moving parts of the machine tool is recorded by the grating ruler and compared with the control command signal of the servo controller. The difference between the two is the corresponding motion error. The motion error is sent to the CNC grinder IPC as a feedback signal. The CNC grinder IPC adopts the corresponding PID control strategy according to the corresponding feedback signal, sends the corresponding control command signal to the servo driver again, and drives the corresponding moving parts of the machine tool to move until the motion error corresponding to the feedback signal reaches the control accuracy of the algorithm. The digital twin control system repeats the above control process to finally achieve the processing requirements of the corresponding product. Based on this, a virtual entity that completely corresponds to the physical entity is created, and real-time monitoring and optimization of the physical entity are realized through digital twin data.

[0014] The service contains key information related to the entire digital twin control system and provides real-time updates on changes in the status of physical and virtual entities;

[0015] Connection is a channel for transmitting information between physical entities and virtual entities. Using information transmission technology, physical entities provide real-time digital twin data for virtual entities.

[0016] Digital twin data includes data from the machine tool entity, status, performance, and control data from the machine tool motion status sensor;

[0017] The established digital twin control system realizes the real-time interaction between virtual entities and physical entities in the grinding process through digital twin data and mutual links, and can perform real-time simulation and timely optimization, thereby achieving precise control of product quality in the grinding process;

[0018] Step 2: Obtaining real-time spatial data of the workpiece based on the digital twin control system

[0019] With product quality as the control target, the relative movement between the workpiece and the grinding wheel is driven by the machine tool motion system to achieve precision machining of the workpiece; the cylindrical surface is machined by the rotational movement of the workpiece and the grinding axis in coordination with the linear feed movement of the X axis; the plane is machined by the rotational movement of the workpiece and the grinding axis in coordination with the linear feed movement of the Z axis; the rotational movement of the workpiece and the grinding axis in coordination with the two-axis interpolation movement of the X axis and the Z axis is used to machine the rotationally symmetrical surface, and the three-axis interpolation movement of the X axis, Z axis and C axis is used to machine the non-rotationally symmetrical complex surface;

[0020] Based on the fact that the surface formation of CNC grinding machine products is closely related to the point-to-point contact between the tool and the workpiece during the machining process, the theoretical coordinates of each machining point are extracted through the digital twin control system, and the spatial data distribution of the workpiece during the machining process is formed through coordinate transformation, as shown in formula (1);

[0021]

[0022] Among them, x actual 、z actual 、c actual They are the actual positions of the XZC coordinates in the machine tool coordinate system during the grinding process, and X(t), Y(t), and Z(t) represent the spatial data distribution of the workpiece coordinates;

[0023] Step 3: Digital twin system parameter calibration

[0024] In the actual machining process, the machining error consists of two parts, one is the machining principle error, and the other is the dynamic error caused by the machining load, which can be expressed as:

[0025]

[0026] Among them, f is the principle error of grinding, including the grinding wheel speed V s , workpiece speed V w , feed speed V f , Grinding depth a p The surface residual height caused by the surface wear marks; g is the principle error of the machine tool movement, including the machine tool XZC guide rail accuracy E X 、E Z 、E C , external interference u in the machining process; h is the elastic-plastic deformation caused by the material's own characteristics E under the action of the grinding force F, j is the elastic recovery, and w is the influencing parameter of the material's elastic recovery degree;

[0027] Under ideal conditions, the machining process is a smooth process. When machining a planar workpiece, the tool-workpiece contact point is on the ideal trajectory, the machined surface is in a flat state, and the error value should be 0. If only the dynamic characteristics of the machine tool are considered, the grinding point of the machine tool fluctuates near the ideal trajectory when machining a planar workpiece, and the machining error value fluctuates around 0. When machining difficult-to-machine materials, due to the difficulty of machining the materials, the machining process will introduce a large machining load, causing the machine tool tool contact point to fluctuate greatly near the ideal trajectory. Based on this, the machining error is expressed as:

[0028]

[0029] Among them, EX is the grinding force F, the material properties E, and the grinding wheel speed Vs , feed speed V f , Grinding depth a p , the influence of external interference u on the X-axis error; EZ is the grinding force F, the material properties E, and the workpiece speed V w , Grinding depth a p , the influence of external interference u on the Z-axis error; EC is the grinding force F, the material properties E, and the workpiece speed V w , the influence of external disturbance u on C-axis error;

[0030] By calibrating the parameters of the digital twin control system, the corresponding relationship between processing load and output voltage is established; when the moving parts of the machine tool are in a balanced position, the position error and output power are approximately zero; when loading, the load of the moving parts increases, generating a positive load Fe+, at which time the output power is less than the external load, and the position Xe of the moving parts deviates positively from the original command position; to ensure that the command position of the moving parts is relatively stable, the output power of the moving parts continues to increase, the motor power rises rapidly and maintains a new stable power output, and when the motor power increases to be consistent with the external load, the position Xe returns to the original command position;

[0031] When the external load is removed, the load on the moving part decreases. At this time, the motor output power is higher than the external load, and the moving part position Xe deviates from the original command position in the opposite direction. To ensure the stability of the command position of the moving part, the motor output power decreases, the power drops rapidly and maintains a new stable power output. When the motor power decreases to the same level as the external load, the guide rail position Xe returns to the original command position.

[0032] Based on this, the relationship between the load and the guide motor current and the guide position error is determined through load calibration, and then the real-time change of the load during the machining process is obtained according to the real-time change of the guide motor current and the guide position error during the grinding process; the real-time position signal of the machine tool is extracted through the digital twin control system, and the surface machining state of the workpiece is judged by the change of the X, Z, and C axis position signals under the machine tool coordinates, and the machining quality of the workpiece is obtained:

[0033]

[0034] Among them, x actual 、z actual 、c actual are the actual positions of the XZC coordinates in the machine tool coordinate system during grinding, instruction 、z instruction 、c instruction are the command positions of the XZC coordinates in the NC program after being discretized by the machine tool NC program, and x pe 、z pe 、c peThey are the influences of the three coordinates X, Z and C during the grinding process;

[0035] Based on this, the real-time position signal of the machine tool is converted into spatial coordinates:

[0036]

[0037] Based on this, the digital twin control system established is used to control the product processing quality in real time;

[0038] Step 4: Obtain real-time product processing quality through digital twin control system

[0039] Realize real-time monitoring and control of product processing quality based on the digital twin control system, and then adjust processing parameters and change processing load in real time during the processing, thereby improving product processing quality;

[0040] The difference between the theoretical position and the real-time position in the digital twin control system is regarded as the result of the deviation of the actual coordinate value from the command coordinate value under the influence of the machine tool dynamic characteristics, grinding parameters, and material response during the processing:

[0041]

[0042] Among them, X(F, E, V s , V f , a p , u) is the grinding force F, the material properties E, and the grinding wheel speed V s , feed speed V f , Grinding depth a p , the influence of external interference u on the deviation of the actual coordinate value of the X-axis from the command coordinate; Z(F, E, V w , a p , u) is the grinding force F, the material properties E, and the workpiece speed V w , Grinding depth a p , the influence of external interference u on Z-axis error; C(F, E,V w , u) is the grinding force F, the material properties E, and the workpiece speed V w , the influence of external disturbance u on C-axis error;

[0043] The error values ​​of the moving parts of the machine tool are extracted respectively to obtain the error distribution results of the product surface data:

[0044]

[0045] Based on this, real-time control of product processing quality can be achieved.

[0046] It should be noted that the cylindrical surface can be processed by the rotation of the workpiece and the grinding axis in conjunction with the linear feed motion of the X axis; the flat surface can be processed by the rotation of the workpiece and the grinding axis in conjunction with the linear feed motion of the Z axis; further, the rotation of the workpiece and the grinding axis in conjunction with the two-axis interpolation motion of the X axis and the Z axis can realize the processing of the rotationally symmetrical surface, and the three-axis interpolation motion of the X axis, the Z axis and the C axis can realize the processing of the non-rotationally symmetrical complex surface. Therefore, the present invention can realize the control of the processing quality of workpieces of any shape without any additional sensors.

[0047] Compared with the prior art, the beneficial effect of the present invention is that the traditional machine tool control method only realizes the precision control of the motion system through a closed-loop control strategy based on a grating ruler, which is a data-driven limited precision control method. When the actuator has deviations due to external conditions such as changes in process parameters and processing materials, it will directly affect the processing accuracy and control stability of the product. For this reason, the present invention takes product quality as the control target, directly faces the real-time following error of the moving parts in the product processing process as the control target, forms a digital twin control system for product processing quality, and realizes multi-closed-loop real-time control of product processing quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flowchart for constructing the digital twin control system of the present invention, and is also a main technical scheme diagram of the present invention.

[0049] Figure 2 It is the main moving part of the CNC machine tool in the embodiment of the present invention.

[0050] Figure 3 This is the product end face workpiece processing process in the embodiment of the present invention.

[0051] Figure 4 This is the process of processing a cylindrical workpiece in an embodiment of the present invention.

[0052] Figure 5 It is the movement path of the contact point between the tool and the workpiece during the product processing process in the embodiment of the present invention.

[0053] Figure 6 It is a corresponding distribution diagram of the contact points between the tool and the workpiece during the product processing process in the embodiment of the present invention.

[0054] Figure 7 It is the point-to-point correspondence between product processing error and processing load in the digital twin control system in the embodiment of the present invention.

[0055] Figure 8 It is a real-time monitoring signal of product surface morphology of the digital twin control system for cylindrical workpiece processing quality in the embodiment of the present invention.

[0056] Fig. 9 It is the processing site status at different times of the digital twin control system for cylindrical workpiece processing quality in the embodiment of the present invention.

[0057] Fig.10 It is a real-time control parameter for product processing errors based on the digital twin control system in an embodiment of the present invention.

[0058] Fig.11 The cylindrical product processing quality is obtained through the digital twin control system in the embodiment of the present invention.

[0059] Fig.12 It is the actual measured result of the surface accuracy of the cylindrical part obtained by the digital twin control system in the embodiment of the present invention.

[0060] Fig.13 The end surface product processing quality is obtained through the digital twin control system in the embodiment of the present invention.

[0061] Fig.14 It is the actual measured result of the surface accuracy of the end face component obtained by the digital twin control system in the embodiment of the present invention.

[0062] In the figure: 1 X-axis feed system, 2 workpiece spindle / C-axis, 3 workpiece, 4 grinding wheel, 5 grinding axis, 6 Z-axis feed system, 7 machining contact point, 101 contact point between tool and workpiece, 102 contact point movement process during machining, 103 contact point between tool and workpiece, 201 first control parameter change point, 202 second control parameter change point. DETAILED DESCRIPTION

[0063] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and technical solutions.

[0064] Since CNC grinders have an additional grinding axis moving part compared to lathes, CNC grinders are used as an example for illustration. It should be noted that the digital twin control system for product processing quality described in the present invention is not only applicable to grinders, but also to lathes.

[0065] The specific implementation of the present invention is described in detail below in conjunction with the technical scheme and the accompanying drawings.

[0066] Step 1: Build a digital twin control system for product processing quality

[0067] Based on the control principle of machine tool moving parts, a digital twin control system is constructed. Figure 1As shown in the figure, the CNC grinder IPC sends the product processing requirements to the moving parts in the form of CNC codes. The moving parts send control command signals according to the processing requirements. The corresponding servo drive converts the control commands into power signals (current and voltage values) of the corresponding moving parts and drives the corresponding moving parts to move. After the motion system moves, the real-time position of the moving parts is recorded by the grating ruler and compared with the controller control command signal. The difference between the two is the corresponding motion error. The motion error is sent to the IPC as a feedback signal. The IPC adopts a suitable PID control strategy according to the corresponding feedback signal, sends the corresponding control command signal to the servo drive again, and drives the corresponding moving parts to move until the motion error corresponding to the system feedback signal reaches the control accuracy of the algorithm. The control system repeats the above control process to finally achieve the processing requirements of the corresponding product.

[0068] The CNC grinder includes two linear motion systems, the X and Z axes, and two rotary motion systems, the grinding axis and the C axis. The physical entities are the various motion entities of the CNC grinder, which provide the mechanism and data basis for the digital twin model. The digital twin data consists of the following real-time data: data from the machine tool entity (such as position, vibration, workpiece material properties, geometric data, etc.); status, performance, and control data from decision control (such as machine tool displacement, speed, current, voltage data, and process parameter data, etc.). Based on this, a digital twin control system for product processing quality is constructed through the control signals of each motion system.

[0069] Using information transmission technology, the physical entity provides real-time processing data for the virtual entity, collects relevant physical, thermal, vibration, voltage, current and other physical process information during the processing, and maps the processing mechanism to the virtual entity. At the same time, the real-time status monitored in the physical entity is transmitted to the service system to guide the real-time monitoring of the grinding status. The virtual entity is monitored by the service system and continues to develop with the service system. In order to ensure the processing quality of the workpiece during the processing and reduce the corresponding control cost, it is necessary to reasonably control each link in the operation of the machine tool.

[0070] Existing CNC machine tool control systems mostly achieve precision control of motion systems through closed-loop control strategies, which is a data-driven limited precision control method. When the actuator deviates due to external conditions such as changes in process parameters and processing materials, it will directly affect the processing accuracy and control stability of the product. This is a semi-open-loop control strategy based on workpiece quality for machine tool operators. The present invention directly takes product quality as the control target, fully considers the relationship between the tool system in grinding, and forms a digital twin control system for product quality.

[0071] The main moving parts of the CNC grinding machine described in this embodiment include two linear feed systems of the guide rail X and Z axes and two rotation systems of the grinding axis and C axis. Figure 2 As shown, the C-axis is installed on the Z-axis feed system and the grinding axis is installed in the X-axis linear feed system.

[0072] Step 2: Obtaining real-time spatial data of the workpiece based on the digital twin control system

[0073] The accuracy of machine tools usually refers to the accuracy of motion that can be achieved by the motion system when the machine tool is unloaded, but the accuracy of machine tools will change when it is loaded. The motion system has a certain rigidity. The load acting on the motion system will cause a large deviation between the real-time position of the motion system and the command position, thereby affecting the machining accuracy of the machine tool and reducing the machining quality of the product. The present invention takes product quality as the control target, and the machining process on the ultra-precision CNC grinding machine is as follows: Figure 3 and Figure 4 As shown in the figure, the relative motion between the workpiece and the grinding wheel is driven by the machine tool motion system to achieve precision machining of the workpiece. The cylindrical surface can be machined by the rotational motion of the workpiece and the grinding axis in conjunction with the linear feed motion of the X axis; the plane can be machined by the rotational motion of the workpiece and the grinding axis in conjunction with the linear feed motion of the Z axis; further, the rotational motion of the workpiece and the grinding axis in conjunction with the two-axis interpolation motion of the X axis and the Z axis can achieve the machining of rotationally symmetrical surfaces, and the three-axis interpolation motion of the X axis, Z axis and C axis can achieve the machining of non-rotationally symmetrical complex surfaces.

[0074] The digital twin control system for product processing quality described in the present invention is specifically applied as follows: Figure 5 and Figure 6 As shown, the surface formation of CNC grinding machine products is closely related to the point-to-point contact between the tool and the workpiece during the machining process. Based on this, the theoretical coordinates of each machining point are extracted through the digital twin control system, and the spatial data distribution of the workpiece during the machining process is formed through coordinate transformation, as shown in Equation 8.

[0075]

[0076] Among them, x actual 、z actual 、c actual They are the actual positions of the XZC coordinates in the machine tool coordinate system during grinding, and X, Y, and Z represent the spatial data distribution of the workpiece coordinates.

[0077] Forming processing path such as Figure 5 As shown. Note that the processing paths are all theoretical contact points in the processing of digital twin products, such as Figure 6 shown.

[0078] Step 3: Digital twin system parameter calibration

[0079] In the actual machining process, the machining error consists of two parts. One part is the machining principle error, such as the machine tool guide rail accuracy, external interference in the machining process, etc.; the other part is the dynamic error caused by the machining load, such as the following error fluctuation caused by the machine tool moving parts under the change of dynamic load, which can be expressed as:

[0080]

[0081] Where, f is the principle error of grinding process, such as grinding parameters (grinding wheel speed V s , workpiece speed V w , feed speed V f , Grinding depth a p g is the principle error of machine tool motion, such as the XZC guide rail accuracy E of the machine tool. X 、E Z 、E C , external interference u in the machining process, etc.; h is the elastic-plastic deformation caused by the material's own characteristics E under the action of grinding force F, and the subsequent elastic recovery j, where w is the influencing parameter of the material's elastic recovery degree.

[0082] Under ideal conditions, the machining process is a smooth process. When machining a planar workpiece, the tool-workpiece contact point is on the ideal trajectory, the machined surface should be flat, and the error value should be 0. If only the dynamic characteristics of the machine tool are considered, the grinding point of the machine tool fluctuates around the ideal trajectory when machining a planar workpiece, and the machining error value should fluctuate around 0. When machining difficult-to-machine materials, due to the difficulty of machining the materials, the machining process will introduce a large machining load, causing the tool-workpiece contact point of the machine tool to fluctuate greatly around the ideal trajectory. Based on this, the machining error can be expressed as:

[0083]

[0084] Among them, EX is the grinding force F, the material properties E, and the grinding wheel speed V s , workpiece speed V w , feed speed V f , Grinding depth a p , the influence of external interference u on the X-axis error; EZ is the grinding force F, the material properties E, the grinding wheel speed V s , workpiece speed V w , feed speed V f , Grinding depth a p , the influence of external interference u on the Z-axis error; EC is the grinding force F, the material properties E, the grinding wheel speed V s , workpiece speed V w , feed speed V f , Grinding depth ap , the influence of external interference u, etc. on the C-axis error.

[0085] By calibrating the parameters of the digital twin system, the corresponding relationship between processing load and output voltage is established, such as Figure 7 As shown. From 0 to t0, the moving parts of the machine tool are in a balanced position, and the position error and output power are approximately zero. A constant load is applied at t0 and maintained until t2. During loading, the load on the moving parts of the machine tool increases, generating Fe+. At this time, the output power is less than the external load, and the position Xe of the moving parts deviates positively from the original command position. In order to ensure that the command position of the moving parts is relatively stable, the output power of the moving parts continues to increase, the motor power rises rapidly and maintains a new stable power output. When the motor power increases to be consistent with the external load, the position Xe returns to the original command position, and the moving parts reach a new equilibrium position at t1. The following error of the moving parts during loading is recorded as Xe+, and the change in machining load is recorded as Fe+.

[0086] During the load holding time t1-t2, the moving parts of the machine tool continue to maintain stable power output and relative stability of the command position at the equilibrium position. Until t2, the external load is removed and the load on the moving parts is reduced. At this time, the motor output power is higher than the external load, and the position Xe of the moving parts deviates from the original command position in the opposite direction. In order to ensure the stability of the command position of the moving parts, the motor output power is reduced, the power drops rapidly and maintains a new stable power output. When the motor power is reduced to the same as the external load, the guide rail position Xe returns to the original command position. At t3, the moving parts reach the equilibrium position before loading. The following error of the moving parts during unloading is recorded as Xe-, and the processing load change is Fe-.

[0087] Repeating the above process, the relationship between the following error and the machining load can be obtained as shown in the following formula:

[0088]

[0089] Based on this, combined with the extraction of real-time position signals of machine tools under the digital twin control system, the surface processing status of the workpiece can be judged by the changes of the X, Z, and C axis position signals under the machine tool coordinates, and then the processing quality of the workpiece can be obtained:

[0090]

[0091] Among them, x instruction , z instruction , c instruction are the command positions of the XZC coordinates in the NC program after being discretized by the machine tool NC program, and x actual , z actual , c actualThey are the actual positions of the XZC coordinates in the machine tool coordinate system during grinding, and the difference between them is x pe , z pe , c pe These are the influences on the three coordinates X, Z and C during the grinding process.

[0092] Based on this, the real-time position signal of the machine tool can be converted into spatial coordinates

[0093]

[0094] Based on this, the product processing quality can be controlled in real time through the established digital twin system.

[0095] Step 4: Obtain real-time product processing quality through digital twin control system

[0096] The digital twin control system established according to the present invention can realize real-time monitoring of product processing quality. Taking cylindrical parts processing as an example, Figure 8 and Fig. 9 The digital twin system established by the present invention can not only accurately identify the start and end times of workpiece processing, as shown at times 101, 102 and 103 in the figure, but also correspond one-to-one with the processing stages in reality.

[0097] The digital twin control system established according to the present invention can also realize real-time control of product processing load. Taking cylindrical parts processing as an example, Fig.10 When the processing parameter Vw is changed to 10 mm / min, 5 mm / min and 2 mm / min respectively, the corresponding average processing loads are 28 N, 22 N and 13 N respectively. According to formula 11, the processing error values ​​can be obtained to be 640 nm, 499 nm and 288 nm respectively. The processing parameters can be adjusted in real time during the processing, the processing load can be changed, and the product processing quality can be improved.

[0098] The difference between the theoretical position and the real-time position in the digital twin control system established by the present invention can be regarded as the result of the deviation of the actual coordinate value from the command coordinate value under the influence of the dynamic characteristics of the machine tool, grinding parameters, and material response during the processing:

[0099]

[0100] Among them, X(F, E, V s , V f , a p , u) is the grinding force F, the material properties E, and the grinding wheel speed V s , feed speed V f , Grinding depth a p, the influence of external interference u on the deviation of the actual coordinate value of the X-axis from the command coordinate; Z(F, E, V w , a p , u) is the grinding force F, the material properties E, and the workpiece speed V w , Grinding depth a p , the influence of external interference u on Z-axis error; C(F, E,V w , u) is the grinding force F, the material properties E, and the workpiece speed V w , the influence of external disturbance u on C-axis error;

[0101] By extracting the error values ​​of the moving parts of the machine tool respectively, the error distribution results of the product surface data can be obtained:

[0102]

[0103] like Figure 11-14 As shown, Fig.11 The processing quality of cylindrical products is obtained through the digital twin control system. Fig.12 It is the measured result of the surface accuracy of cylindrical parts. Fig.13 The end surface product processing quality is obtained through the digital twin control system, and the processing error is about 1.8 μm. Fig.14 This is the actual measurement result of the end face component surface accuracy, and the cylindricality is measured to be 1.4 μm. Based on this, real-time control of product processing quality can be achieved.

[0104] It should be noted that the cylindrical surface can be processed by the rotation of the workpiece and the grinding axis in conjunction with the linear feed motion of the X axis; the flat surface can be processed by the rotation of the workpiece and the grinding axis in conjunction with the linear feed motion of the Z axis; further, the rotation of the workpiece and the grinding axis in conjunction with the two-axis interpolation motion of the X axis and the Z axis can realize the processing of the rotationally symmetrical surface, and the three-axis interpolation motion of the X axis, the Z axis and the C axis can realize the processing of the non-rotationally symmetrical complex surface. Therefore, the present invention can realize the control of the processing quality of workpieces of any shape without any additional sensors.

Claims

1. A digital twin control system for product processing quality, characterized in that: Here are the steps: Step 1: Build a digital twin control system for product processing quality Taking product quality as the control target, considering that in the actual processing process, the workpiece material interferes with the grinding wheel tool system, which in turn causes material deformation and ultimately material removal, a digital twin control system is constructed based on the control principle of machine tool moving parts, including physical entities, digital twin data, virtual entities, services, and links between various components. The physical entities are the moving entities of the CNC grinder, and a digital twin control system for product quality is constructed based on this. The physical entities of the corresponding moving parts of the grinding wheel and the workpiece are analyzed. In the CNC grinding machine, the grinding wheel is connected to the machine tool guide rail through the grinding wheel shaft, forming a "grinding wheel-grinding wheel shaft-CNC machine tool X-axis" connection component system; the grinding wheel is connected to the machine tool guide rail through the workpiece spindle or C-axis, forming a "workpiece-workpiece spindle / C-axis-CNC machine tool Z-axis" connection component system; the above two subsystems are connected together through the bed, distributed in the CNC machine tool coordinate system, and move and realize material removal through command signals in the machine tool CNC system, thereby forming the physical entity of the CNC grinding machine; The virtual entity is the implementation mapping of the physical entity. Considering that the physical entity of the CNC grinder moves through the command signal in the CNC system of the machine tool: the CNC grinder sends the processing requirements of the product to the moving parts of the machine tool in the form of CNC code. The moving parts of the machine tool send control command signals according to the processing requirements. The corresponding servo driver converts the control command signals into power signals of the corresponding moving parts of the machine tool and drives the corresponding moving parts of the machine tool to move. After the moving parts of the machine tool move, the real-time position of the moving parts of the machine tool is recorded by the grating ruler and compared with the control command signal of the servo controller. The difference between the two is the corresponding motion error. The motion error is sent to the CNC grinder IPC as a feedback signal. The CNC grinder IPC adopts the corresponding PID control strategy according to the corresponding feedback signal, sends the corresponding control command signal to the servo driver again, and drives the corresponding moving parts of the machine tool to move until the motion error corresponding to the feedback signal reaches the control accuracy of the algorithm. The digital twin control system repeats the above control process to finally achieve the processing requirements of the corresponding product. Based on this, a virtual entity that completely corresponds to the physical entity is created, and real-time monitoring and optimization of the physical entity are realized through digital twin data. The service contains key information related to the entire digital twin control system and provides real-time updates on changes in the status of physical and virtual entities; Connection is a channel for transmitting information between physical entities and virtual entities. Using information transmission technology, physical entities provide real-time digital twin data for virtual entities. Digital twin data includes data from the machine tool entity, status, performance, and control data from the machine tool motion status sensor; The established digital twin control system realizes the real-time interaction between virtual entities and physical entities in the grinding process through digital twin data and mutual links, and can perform real-time simulation and timely optimization, thereby achieving precise control of product quality in the grinding process; Step 2: Obtaining real-time spatial data of the workpiece based on the digital twin control system With product quality as the control target, the relative movement between the workpiece and the grinding wheel is driven by the machine tool motion system to achieve precision machining of the workpiece; the cylindrical surface is machined by the rotational movement of the workpiece and the grinding axis in coordination with the linear feed movement of the X axis; the plane is machined by the rotational movement of the workpiece and the grinding axis in coordination with the linear feed movement of the Z axis; the rotational movement of the workpiece and the grinding axis in coordination with the two-axis interpolation movement of the X axis and the Z axis is used to machine the rotationally symmetrical surface, and the three-axis interpolation movement of the X axis, Z axis and C axis is used to machine the non-rotationally symmetrical complex surface; Based on the fact that the surface formation of CNC grinding machine products is closely related to the point-to-point contact between the tool and the workpiece during the machining process, the theoretical coordinates of each machining point are extracted through the digital twin control system, and the spatial data distribution of the workpiece during the machining process is formed through coordinate transformation, as shown in formula (1); Among them, x actual 、z actual 、c actual They are the actual positions of the XZC coordinates in the machine tool coordinate system during the grinding process, and X(t), Y(t), and Z(t) represent the spatial data distribution of the workpiece coordinates; Step 3: Digital twin system parameter calibration In the actual machining process, the machining error consists of two parts, one is the machining principle error, and the other is the dynamic error caused by the machining load, which can be expressed as: Among them, f is the principle error of grinding, including the grinding wheel speed V s , workpiece speed V w , feed speed V f , Grinding depth a p The surface residual height caused by the surface wear marks; g is the principle error of the machine tool movement, including the machine tool XZC guide rail accuracy E X 、E Z 、E C , external interference u in the machining process; h is the elastic-plastic deformation caused by the material's own characteristics E under the action of the grinding force F, j is the elastic recovery, and w is the influencing parameter of the material's elastic recovery degree; Under ideal conditions, the machining process is a smooth process. When machining a planar workpiece, the tool-workpiece contact point is on the ideal trajectory, the machined surface is in a flat state, and the error value should be 0. If only the dynamic characteristics of the machine tool are considered, the grinding point of the machine tool fluctuates near the ideal trajectory when machining a planar workpiece, and the machining error value fluctuates around 0. When machining difficult-to-machine materials, due to the difficulty of machining the materials, the machining process will introduce a large machining load, causing the tool-workpiece contact point of the machine tool to fluctuate greatly near the ideal trajectory. Based on this, the machining error is expressed as: Among them, EX is the grinding force F, the material properties E, and the grinding wheel speed V s , feed speed V f , Grinding depth a p , the influence of external interference u on the X-axis error; EZ is the grinding force F, the material properties E, and the workpiece speed V w , Grinding depth a p , the influence of external interference u on the Z-axis error; EC is the grinding force F, the material properties E, and the workpiece speed V w , the influence of external disturbance u on C-axis error; By calibrating the parameters of the digital twin control system, the corresponding relationship between processing load and output voltage is established; when the moving parts of the machine tool are in a balanced position, the position error and output power are approximately zero; when loading, the load of the moving parts increases, generating a positive load Fe+, at which time the output power is less than the external load, and the position Xe of the moving parts deviates positively from the original command position; to ensure that the command position of the moving parts is relatively stable, the output power of the moving parts continues to increase, the motor power rises rapidly and maintains a new stable power output, and when the motor power increases to be consistent with the external load, the position Xe returns to the original command position; When the external load is removed, the load on the moving part decreases. At this time, the motor output power is higher than the external load, and the moving part position Xe deviates from the original command position in the opposite direction. To ensure the stability of the command position of the moving part, the motor output power decreases, the power drops rapidly and maintains a new stable power output. When the motor power decreases to the same level as the external load, the guide rail position Xe returns to the original command position. Based on this, the relationship between the load and the guide motor current and the guide position error is determined through load calibration, and then the real-time change of the load during the machining process is obtained according to the real-time change of the guide motor current and the guide position error during the grinding process; the real-time position signal of the machine tool is extracted through the digital twin control system, and the surface machining state of the workpiece is judged by the change of the X, Z, and C axis position signals under the machine tool coordinates, and the machining quality of the workpiece is obtained: Among them, x actual 、z actual 、c actual are the actual positions of the XZC coordinates in the machine tool coordinate system during grinding, instruction 、z instruction 、c instruction are the command positions of the XZC coordinates in the NC program after being discretized by the machine tool NC program, and x pe 、z pe 、c pe They are the influences of the three coordinates X, Z and C during the grinding process; Based on this, the real-time position signal of the machine tool is converted into spatial coordinates: Based on this, the digital twin control system established is used to control the product processing quality in real time; Step 4: Obtain real-time product processing quality through digital twin control system Realize real-time monitoring and control of product processing quality based on the digital twin control system, and then adjust processing parameters and change processing load in real time during the processing, thereby improving product processing quality; The difference between the theoretical position and the real-time position in the digital twin control system is regarded as the result of the deviation of the actual coordinate value from the command coordinate value under the influence of the machine tool dynamic characteristics, grinding parameters, and material response during the processing: Among them, X(F, E, V s , V f , a p , u) is the grinding force F, the material properties E, and the grinding wheel speed V s , feed speed V f , Grinding depth a p , the influence of external interference u on the deviation of the actual coordinate value of the X-axis from the command coordinate; Z(F, E, V w , a p , u) is the grinding force F, the material properties E, and the workpiece speed V w , Grinding depth a p , the influence of external interference u on Z-axis error; C(F, E, V w , u) is the grinding force F, the material properties E, and the workpiece speed V w , the influence of external disturbance u on C-axis error; The error values ​​of the machine tool moving parts are extracted respectively to obtain the error distribution results of the product surface data: Based on this, real-time control of product processing quality can be achieved.

Images