Visualization method, device and equipment for monitoring data in machining process and medium

Abstract

The invention discloses a visualization method, device and equipment for monitoring data in a machining process and a medium, and relates to the technical field of precision manufacturing, and the method comprises the following steps: obtaining coordinate data of a cutter and machining monitoring data in a workpiece cutting process; performing data fusion of the coordinate data and the processing monitoring data based on real-time stream processing and real-time linear interpolation of historical data to obtain fused data; constructing a three-dimensional scene of the workpiece cutting process based on a finite element network, and putting the fusion data into the three-dimensional scene; and according to the three-dimensional scene, the fusion data in the view cone is rendered, and the rendering depth is determined according to the fusion data and the visual distance of the user. According to the method, the physical state of the cutter in the cutting process can be reproduced in real time, omnibearing visualization of the machining process in the three-dimensional angle and the stress aspect is achieved, then the real-time visualization monitoring effect of the numerical control machine tool machining process is achieved, and networking and informatization development of the numerical control technology is promoted.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of precision manufacturing technology, and in particular to a machining process monitoring data visualization method, device, equipment and medium. BACKGROUND

[0002] In the field of high-end precision manufacturing, real-time monitoring and analysis of the cutting process of numerical control machine tools is the key to ensuring machining quality, prolonging tool life and optimizing process parameters.

[0003] The application of advanced sensors such as smart tool holders makes it possible to obtain high-frequency dynamic data such as pressure, bending moment, torque, temperature and acceleration in the machining area. The existing cutting process monitoring method relies mainly on one-dimensional time series data curves (such as force-time curves), which cannot intuitively associate key physical quantities such as force, torque, temperature and vibration with the precise three-dimensional spatial position of the tool on the workpiece, lack spatial insight, and when processing high-density, high-dynamic spatiotemporal data, often have problems such as low rendering efficiency, high data processing delay and difficulty in effective interactive analysis.

[0004] Therefore, how to accurately synchronize these massive, multi-dimensional physical quantity data with the complex coordinate system of the machine tool in time and space, and present it in real time and intuitively, is an important issue that needs to be addressed in the current field of high-end precision manufacturing. SUMMARY

[0005] Therefore, the embodiments of the present application provide a machining process monitoring data visualization method, device, equipment and medium to solve the problem.

[0006] According to a first aspect, the embodiments of the present application provide a machining process monitoring data visualization method, which comprises:

[0007] Obtaining coordinate data of the tool and machining monitoring data of the workpiece cutting process; the coordinate data includes machine tool coordinate data, absolute coordinate data and relative coordinate data, the machine tool coordinate data is used to represent the absolute position of the tool relative to the machine tool origin, the absolute coordinate data is used to represent the position of the tool relative to the workpiece coordinate system origin, and the relative coordinate data is used to represent the position of the tool relative to the end point of the previous program segment; the machining monitoring data is a physical quantity parameter in the cutting process;

[0008] Based on real-time stream processing and real-time linear interpolation of historical data, data fusion of the coordinate data and the machining monitoring data is performed to obtain fusion data;

[0009] Based on a finite element network, a three-dimensional scene of the workpiece cutting process is constructed, and the fusion data is put into the three-dimensional scene;

[0010] According to the three-dimensional scene, the fusion data located in the view frustum is rendered, and a rendering depth is determined according to a visual distance of the user from the fusion data.

[0011] With reference to the first aspect, in a first implementation manner of the first aspect, the data fusion of the coordinate data and the machining monitoring data is performed based on real-time stream processing and real-time linear interpolation of historical data to obtain the fusion data, and the data fusion specifically includes:

[0012] A first buffer for receiving and storing the coordinate data and a second buffer for receiving and storing the machining monitoring data are respectively constructed;

[0013] A producer-consumer model is constructed by taking the acquired data as the producer and taking the data fusion as the consumer;

[0014] The average sampling frequencies of the first buffer and the second buffer are respectively determined, data corresponding to the buffer with the higher average sampling frequency is taken as the master data stream, and data corresponding to the buffer with the lower average sampling frequency is taken as the slave data stream;

[0015] The data fusion of the master data stream and the slave data stream is performed according to the absolute time stamp of each data in the master data stream and by using historical data in the slave data stream that does not exceed the absolute time stamp, to obtain master-slave fusion data;

[0016] Linear interpolation data is inserted into adjacent master-slave fusion data by using historical data in the slave data stream that does not exceed the absolute time stamp;

[0017] The fusion data is obtained according to the master-slave fusion data and the linear interpolation data.

[0018] With reference to the first implementation manner of the first aspect, in a second implementation manner of the first aspect, the data fusion of the master data stream and the slave data stream is performed according to the absolute time stamp of each data in the master data stream and by using historical data in the slave data stream that does not exceed the absolute time stamp, to obtain master-slave fusion data, and the data fusion specifically includes:

[0019] The absolute time stamp of each data in the master data stream is determined to obtain a first time stamp of each data in the master data stream, and each data in the master data stream is read in a time sequence according to the first time stamp to obtain first data for each data fusion;

[0020] In a case where the data in the master data stream is read, the absolute time stamp of each data in the slave data stream is determined to obtain a second time stamp of each data in the slave data stream, and data in the slave data stream that does not exceed the first time stamp and is closest to the first time stamp is read according to the second time stamp to obtain second data for each data fusion;

[0021] The first and second data for each data fusion are fused to obtain master-slave fusion data for each data fusion.

[0022] With reference to the first aspect and the first implementation manner of the first aspect, in a third implementation manner of the first aspect, the linear interpolation data is inserted into the adjacent two master-slave fusion data by using the historical data from the data stream that does not exceed the absolute timestamp, and specifically includes the following steps.

[0023] A time difference between the adjacent two master-slave fusion data is determined, and the number of data to be inserted between the adjacent two master-slave fusion data and the insertion time of each data to be inserted are determined according to the time difference;

[0024] According to each insertion time, data from the data stream that does not exceed the insertion time and is closest to the insertion time is read to obtain first interpolation data at each insertion time;

[0025] According to the absolute timestamp of each first interpolation data, data from the data stream that does not exceed the absolute timestamp and is closest to the absolute timestamp is read to obtain second interpolation data at each insertion time;

[0026] Linear interpolation is performed according to the insertion time, the first interpolation data and the second interpolation data to obtain third interpolation data at the insertion time;

[0027] Linear interpolation is performed according to the insertion time, data in the master data stream before and after the insertion time to obtain fourth interpolation data at the insertion time, and data fusion is performed on the third and fourth interpolation data to obtain linear interpolation data at each insertion time;

[0028] The linear interpolation data is inserted into the corresponding insertion time.

[0029] With reference to the first aspect, in a fourth implementation manner of the first aspect, the three-dimensional scene of the workpiece cutting process is constructed based on a finite element network, and the fusion data is placed in the three-dimensional scene, and specifically includes the following steps.

[0030] Model information of the workpiece and a machining range of the cutting process are determined, and a virtual bounding box of the workpiece in a three-dimensional space is constructed according to the model information and the machining range;

[0031] A minimum voxel resolution of a target voxel is determined;

[0032] The virtual bounding box is taken as a root node, and each fusion data is sequentially added to the root node in the order of absolute timestamps;

[0033] It is determined whether the root node reaches the target voxel, and in the case of not reaching, the root node is divided into eight first child nodes that are equal, and each fusion data is placed in the corresponding first child node according to the coordinate data;

[0034] determine whether the first child node reaches the target voxel, and in the case of determining not reaching, split the first child node in which the fusion data is put into eight second child nodes equally, put the fusion data into the corresponding second child node according to the coordinate data;

[0035] recursively go down along the tree until reaching the target voxel, put each fusion data into a leaf node, store the fusion data by using the leaf node, and obtain the three-dimensional scene in which the fusion data is put.

[0036] In combination with the first aspect, in a fifth implementation manner of the first aspect, the rendering the fusion data located in the view frustum according to the three-dimensional scene, and determining the rendering depth according to the visual distance of the user from the fusion data specifically comprises

[0037] start the recursive judgment from the root node, and determine the leaf node having the intersection with the view frustum;

[0038] determine the visual distance of the user, determine the rendering level according to the visual distance of the user, render the leaf node having the intersection with the view frustum according to the rendering level, and determine at least one visual attribute for the rendered leaf node.

[0039] In combination with the first aspect, in a sixth implementation manner of the first aspect, the obtaining the coordinate data of the tool and the machining monitoring data of the workpiece cutting process specifically comprises:

[0040] obtain the coordinate data, and generate the absolute timestamp of the coordinate data according to the clock information of the numerical control machine tool;

[0041] when the coordinate data is collected, no matter what coordinate system the coordinate data is in, the high-precision absolute timestamp of the coordinate data is generated according to the clock information of the master clock of the control system of the numerical control machine tool;

[0042] obtain the machining monitoring data, and generate the absolute timestamp of the machining monitoring data according to the clock information of the numerical control machine tool.

[0043] According to the second aspect, the embodiment of the present application further provides a machining process monitoring data visualization device, the device comprises:

[0044] a data acquisition module, configured to obtain the coordinate data of the tool and the machining monitoring data of the workpiece cutting process; the coordinate data comprises machine tool coordinate data, absolute coordinate data and relative coordinate data, the machine tool coordinate data is used to represent the absolute position of the tool relative to the machine tool origin, the absolute coordinate data is used to represent the position of the tool relative to the workpiece coordinate system origin, and the relative coordinate data is used to represent the position of the tool relative to the end point of the previous program segment; the machining monitoring data is a physical quantity parameter in the cutting process;

[0045] a data fusion module configured to fuse the coordinate data and the machining monitoring data based on real-time stream processing and real-time linear interpolation of historical data to obtain fused data;

[0046] a scene construction module configured to construct a three-dimensional scene of the workpiece cutting process based on a finite element network and place the fused data in the three-dimensional scene;

[0047] a data rendering module configured to render the fused data located in a view frustum according to the three-dimensional scene and determine a rendering depth according to a visual distance between the fused data and a user.

[0048] According to a third aspect, an embodiment of the present application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the machining process monitoring data visualization method according to any one of the above aspects when executing the program.

[0049] According to a fourth aspect, an embodiment of the present application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is executable on a processor to implement the steps of the machining process monitoring data visualization method according to any one of the above aspects.

[0050] The machining process monitoring data visualization method, device, equipment and medium provided by the application are based on real-time stream processing and real-time linear interpolation of historical data, coordinate data and machining monitoring data are fused, absolute time stamp is used as the only reference information for data fusion in the data fusion process, the accurate association of coordinate data and machining monitoring data is ensured, and a high-quality data basis is provided for fine three-dimensional reconstruction. A three-dimensional scene of the workpiece cutting process is constructed based on a finite element network, and the fused data is placed in the three-dimensional scene, the finite element network generation method can only store the space region actually passed by the tool, compared with the traditional dense voxel grid, the data volume is greatly compressed, and the memory occupation can be reduced by several orders of magnitude. According to the three-dimensional scene, the fused data located in the view frustum is rendered, and the rendering depth is determined according to the visual distance between the fused data and the user, and the three-dimensional scene constructed has a natural hierarchical structure, so that the three-dimensional scene is a multi-level data structure and the data points in the three-dimensional scene can contain unit information required for rendering, thereby realizing the visualization rendering of multi-dimensional unit information, that is, the physical quantity field, and the newly acquired fused data can be updated to the three-dimensional scene in real time through the data stream, realizing the quasi-real-time three-dimensional visualization monitoring of the dynamic process, and the whole rendering process can ensure high real-time performance. The application can reproduce the physical state of the tool in the cutting process in real time, realize the all-around visualization of the machining process in the three-dimensional angle and the stress, and further realize the real-time visualization monitoring effect of the machining process of the numerical control machine tool, and promote the network and informatization development of numerical control technology. BRIEF DESCRIPTION OF DRAWINGS

[0051] The features and advantages of the application will be more clearly understood through reference to the following drawings, which are presented for illustrative purposes only and are not to be construed as limiting the application, in which:

[0052] Figure 1 One of the flowcharts of the machining process monitoring data visualization method provided by the application is shown;

[0053] Figure 2 The schematic diagram of data fusion in the machining process monitoring data visualization method provided by the application is shown;

[0054] Figure 3 The schematic diagram of the workpiece machining path in the machining process monitoring data visualization method provided by the application is shown;

[0055] Figure 4 The visualization effect diagram of the workpiece machining in the machining process monitoring data visualization method provided by the application is shown;

[0056] Figure 5A schematic diagram of constructing a three-dimensional scene based on a sparse voxel octree in a method for visualizing machining process monitoring data provided by the present application is shown.

[0057] Figure 6 A structural schematic diagram of a device for visualizing machining process monitoring data provided by the present application is shown.

[0058] Figure 7 A hardware structure schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0059] To make the objectives, technical solutions, and advantages of embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0060] In the field of high-end precision manufacturing, real-time monitoring and analysis of the cutting process of a numerical control machine tool is the key to ensuring machining quality, prolonging tool life, and optimizing process parameters.

[0061] The application of advanced sensors such as smart tool holders makes it possible to obtain high-frequency dynamic data such as pressure, bending moment, torque, temperature, and acceleration in the machining area. Existing cutting process monitoring methods mostly rely on one-dimensional time series data curves (such as force-time curves), which cannot intuitively associate key physical quantities such as force, torque, temperature, and vibration with the accurate three-dimensional spatial position of the tool on the workpiece, lack spatial insight, and result in low rendering efficiency, high data processing delay, and difficulty in effective interactive analysis when processing high-density, high-dynamic spatiotemporal data.

[0062] Specifically, cutting geometry simulation is currently mainly offline simulation, which mainly verifies the correctness of the tool path to prevent interference, overcutting, and undercutting, and does not interact with the numerical control machine tool in real time. The monitoring of the machine tool machining process is mainly based on video acquisition electronic board monitoring, the data presentation is single, the user's perception efficiency of the monitoring is low, and the realism of real-time visual virtual monitoring is insufficient. To reduce occupation, most numerical control simulation systems are relatively rough in modeling, and the simulation of the precision machining process is not detailed and intuitive enough.

[0063] In summary, how to accurately synchronize these massive, multi-dimensional physical quantity data with the complex coordinate system of the machine tool in time and space, and present them in real time and intuitively, is an important issue that needs to be addressed in the current field of high-end precision manufacturing.

[0064] In view of the above technical problems, the machining process monitoring data visualization method provided in the embodiments of the present application aims to solve the whole process from data acquisition, data processing to final three-dimensional visualization presentation, reproduce the physical state of the tool in the cutting process in real time, realize the all-round visualization of the machining process in three-dimensional angle and stress, and further realize the real-time visualization monitoring effect of the machining process of the numerical control machine tool, and promote the network and informatization development of the numerical control technology. The machining process monitoring data visualization method can be used in electronic equipment, including but not limited to computers, mobile terminals and the like, Figure 1 is a flowchart of the machining process monitoring data visualization method according to the embodiments of the present application, as shown in Figure 1 , the method can include the following steps:

[0065] S101, coordinate data of the tool and machining monitoring data of the workpiece cutting process are acquired. By acquiring the above-mentioned coordinate data and machining monitoring data, the multi-source heterogeneous data of the numerical control machine tool and the smart tool holder module carried on the numerical control machine tool are obtained.

[0066] In the present embodiment, the coordinate data includes machine coordinates data, absolute coordinates / workpiece coordinates data and relative coordinates data, wherein the machine coordinates data is used to represent the absolute position of the tool relative to the machine origin, the absolute coordinates / workpiece coordinates data is used to represent the position of the tool relative to the workpiece coordinate system origin, and the relative coordinates data is used to represent the position of the tool relative to the end point of the previous program segment (machining segment).

[0067] The machining monitoring data is various physical quantity parameters in the cutting process, including but not limited to pressure, bending moment, torque, temperature, acceleration, sound and angular displacement, etc.

[0068] The coordinate data and the machining monitoring data can be stored in the electronic equipment in advance, or can be acquired by the electronic equipment from the outside world. For example, by accessing the internal bus, special API interface or standard industrial communication protocol (such as OPC UA, MTConnect) of the control system (CNC Controller) of the numerical control machine tool, the three-dimensional coordinate data of the tool center point (Tool Center Point, TCP) is read in real time, and then the coordinate data of the tool is obtained; the machining monitoring data of the cutting process is collected by the sensor and the wireless data transmission module (such as Zigbee, Bluetooth, Wi-Fi) built in the smart tool holder module.

[0069] There are no restrictions on the specific form of obtaining coordinate data and processing monitoring data, as long as the electronic equipment can acquire the coordinate data and processing monitoring data.

[0070] S102. Based on real-time stream processing and real-time linear interpolation of historical data, coordinate data and processing monitoring data are fused to obtain fused data. The fused data includes three types of information: absolute timestamp, coordinate data, and processing monitoring data.

[0071] Please see Figure 2 In this embodiment, a low-latency, high-throughput data fusion pipeline is constructed based on real-time stream processing and real-time linearity of historical data. This data fusion pipeline can merge multi-source heterogeneous data streams from CNC machine tools and smart tool holder modules into a unified, time-precisely aligned structured data stream. Furthermore, the data fusion process does not introduce future data and uses historical data, providing an immediate data source for subsequent 3D data visualization.

[0072] S103. Construct a three-dimensional scene of the workpiece cutting process based on the finite element network, and put the fused data into the three-dimensional scene.

[0073] In this embodiment, a three-dimensional scene of the workpiece cutting process is constructed based on the finite element network generation method. The generation of the finite element network refers to discretizing a continuous geometric model into the element information required for calculation. The quality of the finite element network directly affects the speed and accuracy of the solution. The generation of the finite element network can employ methods such as mapping, quad (octet) tree method, and Delaunay triangulation method, without specific limitations here.

[0074] S104. Based on the 3D scene, render the fused data located within the view frustum, and determine the rendering depth based on the visual distance between the fused data and the user.

[0075] Since 3D scenes are constructed based on finite element networks, they are multi-level data structures, and the data points within a 3D scene can contain the unit information required for rendering. This allows for the visualization and rendering of multi-dimensional unit information, i.e., physical quantity fields. Taking the generation of the finite element network using a sparse voxel octree (SVO) as an example, users can customize the display settings for visualization rendering according to their needs. For instance, for color, the color of a voxel can represent the temperature or the magnitude of other required signals (such as a blue-red gradient); for opacity, the brightness or transparency of a voxel can represent the magnitude of the required signal; for geometric symbols, an arrow or vector can be rendered at the voxel position, its direction and length representing the direction and amplitude of the required signal; for dynamic effects, blinking or pulsation effects can be used to represent the exceeding limits or violent fluctuations of the required signal. Users can interactively switch between the currently visualized physical quantities or overlay multiple physical quantities, thereby achieving multi-dimensional insights into complex processing procedures.

[0076] For example, the intelligent tool holder module collects four physical parameters in real time during machining at an average sampling frequency of 1024 Hz: pressure, torque, lateral bending moment, and longitudinal bending moment, obtaining machining monitoring data. Simultaneously, the coordinate information of various parameters during machining is acquired by the intelligent tool holder module within the T-600SE drilling and milling machining center (Taiqun Precision Machine Tool), obtaining coordinate data. The workpiece material is aluminum alloy, and the cutting path for machining the workpiece is as follows... Figure 3 As shown, the 3D display of forces during the processing is achieved through visualization rendering, and the visualization effect is as follows: Figure 4 As shown, ch1 represents the transverse bending moment, ch4 represents the total longitudinal bending moment, ch3 represents the pressure, and ch2 represents the torque. Based on this visualization, users can intuitively observe the force and deformation behavior of the tool during the machining process, and even abnormal situations such as overcutting, wear, and vibration.

[0077] The visualization method for machining process monitoring data of this invention is based on real-time stream processing and real-time linear interpolation of historical data to fuse coordinate data and machining monitoring data. During the data fusion process, an absolute timestamp is used as the sole reference information for data fusion, ensuring the accurate correlation between coordinate data and machining monitoring data, providing a high-quality data foundation for refined 3D reconstruction. A 3D scene of the workpiece cutting process is constructed based on a finite element network (FEM), and the fused data is placed into this 3D scene. The FEM generation method can store only the spatial area actually traversed by the tool. For sparse scenes with complex tool paths and machining only in localized areas of the workpiece, compared to traditional dense voxel meshes, it significantly compresses the data volume, reducing memory usage by several orders of magnitude. Based on the 3D scene, fused data located within the view frustum is rendered, and the rendering depth is determined according to the visual distance between the fused data and the user. The constructed 3D scene has a natural hierarchical structure, thus it is a multi-level data structure, and the data points in the 3D scene can contain the unit information required for rendering. This enables the visualization rendering of multi-dimensional unit information, i.e., physical quantity fields. Simultaneously, newly acquired fused data can be updated to the 3D scene in real time via data streams, achieving near real-time 3D visualization monitoring of dynamic processes. The entire rendering process ensures a high degree of real-time performance. This invention can reproduce the physical state of the tool during the cutting process in real time, achieving comprehensive visualization of the machining process in terms of 3D angles and forces. This enables real-time visualization monitoring of CNC machine tool machining processes, promoting the networking and informatization of CNC technology.

[0078] In this embodiment, step S101 specifically includes:

[0079] S1011. Obtain coordinate data and generate an absolute timestamp of the coordinate data based on the clock information of the CNC machine tool.

[0080] When coordinate data is acquired, regardless of the coordinate system, a high-precision absolute timestamp will be generated based on the clock information of the master clock of the CNC machine tool's control system.

[0081] S1012. Acquire machining monitoring data and generate an absolute timestamp of the machining monitoring data based on the clock information of the CNC machine tool.

[0082] After collecting the machining monitoring data, it can be synchronized with the clock information of the CNC machine tool to generate an absolute timestamp of the machining monitoring data. Alternatively, the clock information of each sensor in the smart tool holder module can be used to synchronize with the clock information of the CNC machine tool to generate an absolute timestamp of the machining monitoring data.

[0083] In this embodiment, step S102 specifically includes:

[0084] S1021. Construct a first buffer for receiving and storing coordinate data, and a second buffer for receiving and storing processing monitoring data.

[0085] To account for data loss or congestion due to network jitter and system load, two independent and thread-safe data buffers will be constructed: a first buffer and a second buffer. These buffers allow the real-time streaming framework to be used only during data aggregation, thus achieving data decoupling and non-blocking processing.

[0086] In this embodiment, the data buffer is implemented using a ring buffer or a message queue.

[0087] S1022. The acquired data is treated as the producer and the data fusion is treated as the consumer. A production-consumption model is constructed. The producer will push the acquired coordinate data to the first buffer and the acquired processing monitoring data to the second buffer. The consumer will read the data asynchronously from the first and second buffers.

[0088] Because the acquired coordinate data and processing monitoring data may have inherent differences in sampling frequency, asynchronous data streams are first aggregated and buffered in real time before data fusion. The first and second buffers, along with the producer-consumer model, effectively smooth out the jitter problem caused by network fluctuations or system load, ensuring that the data fusion module can always process the currently arriving data at a stable pace. This avoids the waiting-blocking process and is the foundation for achieving real-time data fusion processing.

[0089] S1023. Determine the average sampling frequency of the first and second buffers respectively, that is, determine the average sampling frequency of the coordinate data and the processing monitoring data respectively. Use the data corresponding to the buffer with the higher average sampling frequency as the main data stream and the data corresponding to the buffer with the lower average sampling frequency as the slave data stream. For example, if the average sampling frequency of the coordinate data pushed to the first buffer is higher, then the coordinate data is used as the main data stream and the processing monitoring data is used as the slave data stream. The main data stream will be used as the reference in subsequent data fusion.

[0090] It should be noted that if the average sampling frequencies of the two buffers are equal, one type of data will be randomly selected as the main data stream and the other as the slave data stream.

[0091] S1024. Based on the absolute timestamp of each data in the master data stream and using historical data in the slave data stream that does not exceed the absolute timestamp, perform data fusion between the master and slave data streams to obtain master-slave fused data.

[0092] In this embodiment, a fusion data is determined for each data in the main data stream, and the fusion data is taken from the buffer corresponding to the data stream. The data in the main data stream is used as the reference, and based on the absolute timestamp of the data in the main data stream used as the reference, historical data no more than the absolute timestamp is determined from the data stream. Then, the data in the historical data that is closest to the absolute timestamp is used as the fusion data, and then data fusion is performed.

[0093] More specifically, step S1024 includes:

[0094] S10241. Determine the absolute timestamp of each data in the main data stream to obtain the first timestamp of each data in the main data stream. Read each data in the main data stream in sequence according to the chronological order of the first timestamps to obtain the first data for each data fusion.

[0095] In this embodiment, whenever a piece of data is retrieved from the buffer corresponding to the main data stream and read by the consumer, the data will immediately trigger time alignment and fusion calculation. That is, each piece of data in the main data stream that is read is the first piece of data for a data fusion.

[0096] S10242. When data in the main data stream is read, determine the absolute timestamp of each data in the data stream to obtain the second timestamp of each data in the data stream. Based on the second timestamp, read the data in the data stream that does not exceed the first timestamp and is closest to the first timestamp in sequence to obtain the second data for each data fusion.

[0097] To avoid high latency, data fusion is performed on time alignment without introducing future data. Therefore, it reads data from the data stream that is no more than the first time stamp and closest to the first time stamp in sequence according to the second time stamp, and uses the data from the data stream to predict the data at the corresponding time node.

[0098] Assume the absolute timestamp of a certain data in the main data stream is When the data is retrieved from the first buffer, it will be searched in the second buffer for an absolute timestamp that does not exceed [a certain value]. The data will be in the second buffer at the distance The most recent historical data at that moment is used as the second data point, and the second data point is directly used as... The data values ​​at each time point. This process employs a zero-order hold strategy, which assumes that the signal from the data stream remains unchanged between two adjacent sampling points, thereby achieving low-latency, high-real-time data prediction.

[0099] S10243. Perform data fusion on the first and second data of each data fusion to obtain the master-slave fused data of each data fusion.

[0100] The master-slave fused data consists of structured multi-attribute spatiotemporal data points. The data format of the master-slave fused data is: {Timestamp: T_master, Coordinate: (X, Y, Z), Attributes: [Force, Torque,...]}.

[0101] S1025. Using historical data from the data stream that does not exceed the absolute timestamp, insert linear interpolation data into two adjacent master-slave fused data.

[0102] Considering that the data point density of the master-slave fused data is still relatively sparse, in this embodiment, historical data will also be used to perform real-time linear interpolation to generate at least one data point in two adjacent master-slave fused data, thus obtaining linear interpolated data. Linear interpolation data can increase the point density of the fused data.

[0103] More specifically, step S1025 includes:

[0104] S10251. Determine the time difference between two adjacent master-slave fused data sets, and determine the number of data to be inserted between the two adjacent master-slave fused data sets and the insertion time of each data to be inserted based on the time difference. By reasonably setting the number of data to be inserted, the point density of the fused data can be reasonably controlled. A linear interpolation data will be inserted at each data to be inserted, and each data to be inserted will also have a corresponding insertion time.

[0105] S10252. Based on each insertion time, read the data from the data stream that is no more than the insertion time and is closest to the insertion time, and obtain the first interpolated data for each insertion time.

[0106] S10253. Based on the absolute timestamp of each first interpolation data, read the data from the data stream that does not exceed the absolute timestamp and is closest to the absolute timestamp, and obtain the second interpolation data for each insertion time.

[0107] S10254. Perform linear interpolation based on the insertion time, the first interpolation data, and the second interpolation data to obtain the third interpolation data at the insertion time. The third interpolation data is a portion of the data used for subsequent data fusion.

[0108] Assume the absolute timestamp of a certain insertion time is During real-time linear interpolation, the system searches the second buffer for timestamps that do not exceed a certain threshold. The data will be in the second buffer at the distance The most recent historical data at this moment is used as the first interpolation data, and the absolute timestamp corresponding to this first interpolation data is: Then place in Before and with The most recent historical data in the second buffer is used as the second interpolation data, and the absolute timestamp corresponding to this first interpolation data is... In terms of time Then, linear interpolation is performed based on the first and second interpolation data to obtain the linear interpolation data at that insertion time. In this way, real-time linear interpolation processing is performed for each insertion time to obtain the linear interpolation data at each insertion time. The data format of the linear interpolation data is consistent with the data format of the master-slave data fusion, which will not be elaborated here.

[0109] S10255. Based on the data in the main data stream before and after the insertion time, perform linear interpolation to obtain the fourth interpolated data at the insertion time. Then, fuse the third and fourth interpolated data to obtain the linear interpolated data for each insertion time. Similarly, use linear interpolation to obtain the fourth interpolated data for each insertion time. This fourth interpolated data is a portion of the data used for subsequent data fusion.

[0110] S10256. Insert the linear interpolation data to the corresponding insertion time.

[0111] S1026. Obtain fused data based on master-slave fused data and linear interpolation data. After obtaining the linear interpolation data, each linear interpolation data point is inserted to the corresponding insertion time, and then combined with the master-slave fused data to obtain the final fused data.

[0112] Please see Figure 5 Taking the generation of finite element networks using SVO as an example, in this embodiment, step S103 specifically includes:

[0113] S1031. Determine the model information of the workpiece and the machining range of the cutting process, and construct a virtual bounding box of the workpiece in three-dimensional space based on the model information and the machining range.

[0114] The model information can be the CAD model of the workpiece, combined with the machining range of the workpiece cutting process, to determine a virtual bounding box in three-dimensional space. This virtual bounding box will become the root node of the subsequent SVO recursion. SVO is an efficient 3D spatial data structure, especially suitable for representing sparse voxel data. SVO is a spatial data structure for efficiently organizing voxels, and SVO significantly reduces memory usage by storing only the actual voxels, providing efficient query and dynamic update capabilities.

[0115] S1032. Determine the minimum voxel resolution of the target voxel. The target voxel can be understood as the maximum depth of the SVO recursion. The minimum voxel resolution determines the level of detail of the final visualization model, which can reach the micrometer level to match the processing accuracy requirements.

[0116] S1033. Use the virtual bounding box as the root node, and add each fused data to the root node in the order of absolute timestamps, thereby locating each fused data to the root node.

[0117] S1034. Determine whether the root node has reached the target voxel resolution, i.e., whether the voxel resolution of the root node has reached the minimum voxel resolution. If it has not, divide the root node into eight equal first child nodes, and put each fused data into the corresponding first child node according to the coordinate data. This process is the first recursive split, where the fused data is put into one of the eight first child nodes according to the coordinate data, i.e., the coordinate information.

[0118] S1035. Determine whether the first child node has reached the target voxel resolution, i.e., whether the voxel resolution of the first child node has reached the minimum voxel resolution. If it has not, divide the first child node containing the fused data into eight equal second child nodes, and place the fused data into the corresponding second child node according to the coordinate data. This process is the second recursive split, which redistributes each piece of fused data according to the coordinate data, placing the fused data into one of the eight second child nodes.

[0119] S1036. Recursively follow the tree downwards until the target voxel is reached. Place each fused data into a leaf node and use the leaf node to store various types of information contained in the fused data to obtain the 3D scene containing the fused data.

[0120] In this embodiment, fused data points are streamed and inserted into the SVO in real time. The insertion process follows the recursive splitting rule of the SVO, starting from the root node and dynamically and on-demand creating child nodes along the path based on the spatial location of the SVO until the target voxel is reached. This process ensures that only the spatial region containing data is subdivided and stored. The SVO dynamically generates tree and leaf nodes (target voxels) based on the spatial location of the data points, creating voxels only along the actual trajectory path traversed by the tool. For vast unprocessed areas, no storage space is occupied, greatly compressing the data volume. Each activated leaf node stores the coordinate data corresponding to the fused data and the machining monitoring data, i.e., spatial location and sensor data vectors.

[0121] In this embodiment, step S104 specifically includes:

[0122] S1041. Starting from the root node, perform recursive judgment to determine the leaf nodes that intersect with the view frustum.

[0123] Because SVO is a tree structure, the system can quickly determine the location from the root node. If a large node is completely outside the view frustum, the system can skip that node and all its descendant nodes without processing them. Only nodes that intersect with the view frustum are checked and rendered, recursively checking until the leaf nodes are reached. This way, regardless of the size of the entire 3D scene dataset (e.g., recording several hours of processing paths), only the currently visible portion is processed in each frame. This greatly reduces the GPU load and ensures a high frame rate.

[0124] S1042. Determine the user's visual distance, determine the rendering level based on the user's visual distance, render the leaf nodes that intersect with the view frustum based on the rendering level, and determine at least one visual attribute for the rendered leaf nodes. That is, in visualization, multiple physical quantities are presented simultaneously through different visual attributes (such as color, shape, dynamic effects, etc.).

[0125] By leveraging the hierarchical data structure of the finite element method (FEM) network generation, frustum culling and detail rendering can be implemented efficiently. Specifically, only nodes within the frustum are rendered, and the rendering depth is dynamically selected based on the node's visual distance from the user, thus significantly improving the rendering frame rate while maintaining visual quality. When the user zooms out to observe the overall path, the system does not need to render every tiny voxel; instead, it can directly render higher-level, coarser nodes. When the user zooms in to observe local details, the system then progressively delves deeper into the frustum, ultimately rendering voxels with precise data. This ensures that whether the user is browsing macroscopically or exploring microscopically, and even when zooming in for microscopic details or zooming out to observe the overall processing path, the system renders with the most appropriate load, maintaining a consistently smooth frame rate.

[0126] In this embodiment, users can perform interactive operations such as slicing, cross-sectional analysis, and numerical query on the generated three-dimensional "machining trajectory physical field" to intuitively locate the precise spatial location of machining defects (such as built-up edge, entanglement, etc.) and their corresponding physical parameters.

[0127] Specifically, the system can efficiently traverse the SVO (Sectional Value Entity), collecting only all voxels located on this "section" and displaying their physical properties (force, temperature, etc.) as a two-dimensional color image. This helps identify stress concentration zones or overheating areas inside the workpiece. Users can also directly click on any voxel on the 3D model with the mouse, and the system can quickly locate the corresponding leaf node in the SVO using spatial coordinates, popping up an information box displaying the precise spatial coordinates, timestamp, and all sensor readings of that point.

[0128] The visualization device for monitoring machining process data provided in the embodiments of the present invention is described below. The visualization device for monitoring machining process data described below and the visualization method for monitoring machining process data described above can be referred to in correspondence with each other.

[0129] Due to the aforementioned technical problems, this invention also provides a visualization device for monitoring machining process data. This device aims to solve the entire process from data acquisition and processing to final three-dimensional visualization, reproduce the physical state of the tool during the cutting process in real time, and achieve comprehensive visualization of the machining process in terms of three-dimensional angles and forces. This enables real-time visualization monitoring of the CNC machine tool machining process and promotes the networking and informatization of CNC technology. Figure 6 This is a schematic diagram of the structure of a method for visualizing machining process monitoring data according to an embodiment of the present invention, such as... Figure 6 As shown, the device may include:

[0130] The data acquisition module is used to acquire the coordinate data of the cutting tool and the machining monitoring data of the workpiece cutting process. By acquiring the aforementioned coordinate data and machining monitoring data, multi-source heterogeneous data of the CNC machine tool and the smart tool holder module mounted on the CNC machine tool can be obtained.

[0131] In this embodiment, the coordinate data includes machine tool coordinate data, absolute workpiece coordinate data, and relative coordinate data. The machine tool coordinate data is used to characterize the absolute position of the tool relative to the machine tool origin, the absolute coordinate / workpiece coordinate data is used to characterize the position of the tool relative to the workpiece coordinate system origin, and the relative coordinate data is used to characterize the position of the tool relative to the end point of the previous program segment (machining segment).

[0132] The machining monitoring data includes various physical parameters during the cutting process, including but not limited to pressure, bending moment, torque, temperature, acceleration, sound, and angular displacement.

[0133] Coordinate data and machining monitoring data can be pre-stored in electronic devices or acquired by the electronic devices from external sources. For example, by accessing the internal bus, dedicated API interface, or standard industrial communication protocol of the CNC machine tool's control system, the three-dimensional coordinate data of the tool's center point can be read in real time, thereby obtaining the tool's coordinate data; machining monitoring data of the cutting process can be collected through sensors and wireless data transmission modules (such as Zigbee, Bluetooth, and Wi-Fi) built into the smart tool holder module.

[0134] There are no restrictions on the specific form of obtaining coordinate data and processing monitoring data, as long as the electronic equipment can acquire the coordinate data and processing monitoring data.

[0135] The data fusion module is used to fuse coordinate data and processing monitoring data based on real-time stream processing and real-time linear interpolation of historical data, resulting in fused data. The fused data includes three types of information: absolute timestamps, coordinate data, and processing monitoring data.

[0136] In this embodiment, a low-latency, high-throughput data fusion pipeline is constructed based on real-time stream processing and real-time linearity of historical data. This data fusion pipeline can merge multi-source heterogeneous data streams from CNC machine tools and smart tool holder modules into a unified, time-precisely aligned structured data stream. Furthermore, the data fusion process does not introduce future data and uses historical data, providing an immediate data source for subsequent 3D data visualization.

[0137] The scene construction module is used to construct a 3D scene of the workpiece cutting process based on the finite element network and put the fused data into the 3D scene.

[0138] In this embodiment, a three-dimensional scene of the workpiece cutting process is constructed based on the finite element network generation method. The generation of the finite element network refers to discretizing a continuous geometric model into the element information required for calculation. The quality of the finite element network directly affects the speed and accuracy of the solution. The generation of the finite element network can employ methods such as mapping, quad (octet) tree method, and Delaunay triangulation method, without specific limitations here.

[0139] The data rendering module is used to render the fused data located within the view frustum based on the 3D scene, and to determine the rendering depth based on the visual distance between the fused data and the user.

[0140] Since the 3D scene is constructed based on the finite element network, the 3D scene is a multi-level data structure and the data points in the 3D scene can contain the unit information required for rendering, thus enabling the visualization rendering of multi-dimensional unit information, that is, physical quantity fields.

[0141] The visualization device for machining process monitoring data of the present invention performs data fusion of coordinate data and machining monitoring data based on real-time stream processing and real-time linear interpolation of historical data. During the data fusion process, an absolute timestamp is used as the sole reference information for data fusion, ensuring the accurate correlation between coordinate data and machining monitoring data, providing a high-quality data foundation for refined 3D reconstruction. A 3D scene of the workpiece cutting process is constructed based on a finite element network, and the fused data is placed into the 3D scene. The finite element network generation method can store only the spatial area actually traversed by the tool. For sparse scenes with complex tool paths and machining only in localized areas of the workpiece, compared to traditional dense voxel meshes, it greatly compresses the data volume, reducing memory usage by several orders of magnitude. Based on the 3D scene, fused data located within the view frustum is rendered, and the rendering depth is determined according to the visual distance between the fused data and the user. The constructed 3D scene has a natural hierarchical structure, thus it is a multi-level data structure, and the data points in the 3D scene can contain the unit information required for rendering. This enables the visualization rendering of multi-dimensional unit information, i.e., physical quantity fields. Simultaneously, newly acquired fused data can be updated to the 3D scene in real time via data streams, achieving near real-time 3D visualization monitoring of dynamic processes. The entire rendering process ensures a high degree of real-time performance. This invention can reproduce the physical state of the tool during the cutting process in real time, achieving comprehensive visualization of the machining process in terms of 3D angles and forces. This enables real-time visualization monitoring of CNC machine tool machining processes, promoting the networking and informatization of CNC technology.

[0142] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other via the communication bus 740. The processor 710 can call logical commands in the memory 730 to execute a visualization method for machining process monitoring data, the method including:

[0143] Acquire tool coordinate data and workpiece cutting process monitoring data; coordinate data includes machine tool coordinate data, absolute coordinate data and relative coordinate data. Machine tool coordinate data is used to characterize the absolute position of the tool relative to the machine tool origin, absolute coordinate data is used to characterize the position of the tool relative to the workpiece coordinate system origin, and relative coordinate data is used to characterize the position of the tool relative to the end point of the previous program segment. The machining monitoring data consists of physical quantity parameters during the cutting process.

[0144] Based on real-time stream processing and real-time linear interpolation of historical data, coordinate data and processing monitoring data are fused to obtain fused data.

[0145] A 3D scene of the workpiece cutting process is constructed based on the finite element network, and the fused data is put into the 3D scene;

[0146] Based on the 3D scene, render the fused data located within the view frustum, and determine the rendering depth based on the visual distance between the fused data and the user.

[0147] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0148] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the visualization method for machining process monitoring data provided by the above methods, the method comprising:

[0149] Acquire tool coordinate data and workpiece cutting process monitoring data; coordinate data includes machine tool coordinate data, absolute coordinate data and relative coordinate data. Machine tool coordinate data is used to characterize the absolute position of the tool relative to the machine tool origin, absolute coordinate data is used to characterize the position of the tool relative to the workpiece coordinate system origin, and relative coordinate data is used to characterize the position of the tool relative to the end point of the previous program segment. The machining monitoring data consists of physical quantity parameters during the cutting process.

[0150] Based on real-time stream processing and real-time linear interpolation of historical data, coordinate data and processing monitoring data are fused to obtain fused data.

[0151] A 3D scene of the workpiece cutting process is constructed based on the finite element network, and the fused data is put into the 3D scene;

[0152] Based on the 3D scene, render the fused data located within the view frustum, and determine the rendering depth based on the visual distance between the fused data and the user.

[0153] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the visualization methods for performing machining process monitoring data provided above, the method comprising:

[0154] Acquire tool coordinate data and workpiece cutting process monitoring data; coordinate data includes machine tool coordinate data, absolute coordinate data and relative coordinate data. Machine tool coordinate data is used to characterize the absolute position of the tool relative to the machine tool origin, absolute coordinate data is used to characterize the position of the tool relative to the workpiece coordinate system origin, and relative coordinate data is used to characterize the position of the tool relative to the end point of the previous program segment. The machining monitoring data consists of physical quantity parameters during the cutting process.

[0155] Based on real-time stream processing and real-time linear interpolation of historical data, coordinate data and processing monitoring data are fused to obtain fused data.

[0156] A 3D scene of the workpiece cutting process is constructed based on the finite element network, and the fused data is put into the 3D scene;

[0157] Based on the 3D scene, render the fused data located within the view frustum, and determine the rendering depth based on the visual distance between the fused data and the user.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of visualizing machining process monitoring data, characterized by, The method comprises: acquiring coordinate data of a tool and machining monitoring data of a workpiece cutting process; the coordinate data comprises machine tool coordinate data, absolute coordinate data and relative coordinate data, the machine tool coordinate data is used to represent an absolute position of the tool relative to a machine tool origin, the absolute coordinate data is used to represent a position of the tool relative to a workpiece coordinate system origin, and the relative coordinate data is used to represent a position of the tool relative to an end point of a previous program segment; the machining monitoring data is a physical quantity parameter in a cutting process; based on real-time stream processing and real-time linear interpolation of historical data, data fusion of the coordinate data and the machining monitoring data is performed to obtain fused data; a three-dimensional scene of the workpiece cutting process is constructed based on a finite element network, and the fused data is placed in the three-dimensional scene; based on the three-dimensional scene, the fused data located in a view frustum is rendered, and a rendering depth is determined according to a visual distance between the fused data and a user.

2. The method of visualizing machining process monitoring data of claim 1, wherein, The data fusion of the coordinate data and the machining monitoring data based on real-time stream processing and real-time linear interpolation of historical data to obtain fused data specifically comprises: a first buffer for receiving and storing the coordinate data and a second buffer for receiving and storing the machining monitoring data are respectively constructed; a production-consumption model is constructed by taking the acquired data as a producer and taking the data fusion as a consumer; average sampling frequencies of the first and second buffers are respectively determined, data corresponding to the buffer with a higher average sampling frequency is taken as a master data stream, and data corresponding to the buffer with a lower average sampling frequency is taken as a slave data stream; data fusion of the master and slave data streams is performed according to an absolute timestamp of each data in the master data stream and by using historical data in the slave data stream that does not exceed the absolute timestamp, to obtain master-slave fused data; linear interpolation data is inserted into adjacent two master-slave fused data by using historical data in the slave data stream that does not exceed the absolute timestamp; the fused data is obtained according to the master-slave fused data and the linear interpolation data.

3. The method of visualizing machining process monitoring data of claim 2, wherein, The data fusion of the master and slave data streams according to an absolute timestamp of each data in the master data stream and by using historical data in the slave data stream that does not exceed the absolute timestamp to obtain master-slave fused data specifically comprises: the absolute timestamp of each data in the master data stream is determined to obtain a first timestamp of each data in the master data stream, and each data in the master data stream is read in order according to the first timestamp on the time axis to obtain first data for each data fusion; in the case that the data in the master data stream is read, the absolute timestamp of each data in the slave data stream is determined to obtain a second timestamp of each data in the slave data stream, and data in the slave data stream that does not exceed the first timestamp and is closest to the first timestamp is read in order according to the second timestamp to obtain second data for each data fusion; the first and second data for each data fusion are fused to obtain master-slave fused data for each data fusion.

4. The method of visualizing machining process monitoring data of claim 2, wherein, The linear interpolation data is inserted into adjacent two master-slave fused data by using historical data in the slave data stream that does not exceed the absolute timestamp, specifically comprising: determining a time difference between two adjacent master-slave fusion data, determining a number of data to be inserted between the two adjacent master-slave fusion data according to the time difference, and determining an insertion time of each data to be inserted; reading, according to each insertion time, data from the data stream that is closest to the insertion time and is not later than the insertion time, to obtain first interpolation data of each insertion time; reading, according to an absolute timestamp of each first interpolation data, data from the data stream that is closest to the absolute timestamp and is not later than the absolute timestamp, to obtain second interpolation data of each insertion time; performing linear interpolation according to the insertion time, the first interpolation data and the second interpolation data to obtain third interpolation data of the insertion time; performing linear interpolation according to the insertion time and data in the master data stream before and after the insertion time to obtain fourth interpolation data of the insertion time, and performing data fusion on the third and fourth interpolation data to obtain linear interpolation data of each insertion time; inserting the linear interpolation data into the corresponding insertion time.

5. The method of visualizing machining process monitoring data of claim 1, wherein, The method comprises the following steps: determining model information of the workpiece and a machining range of the cutting process, and constructing a virtual bounding box of the workpiece in a three-dimensional space according to the model information and the machining range; determining a minimum voxel resolution of a target voxel; taking the virtual bounding box as a root node, and sequentially adding each fusion data to the root node according to the order of absolute timestamps; determining whether the root node reaches the target voxel, and in the case of not reaching the target voxel, dividing the root node into eight first child nodes, and placing each fusion data into a corresponding first child node according to coordinate data; determining whether the first child node reaches the target voxel, and in the case of not reaching the target voxel, dividing the first child node into eight second child nodes, and placing the fusion data into a corresponding second child node according to the coordinate data; recursively descending along the tree until the target voxel is reached, placing each fusion data into a leaf node, and storing the fusion data in the leaf node to obtain a three-dimensional scene with the fusion data.

6. The method of visualizing machining process monitoring data of claim 1, wherein, The method comprises the following steps: starting from the root node to recursively determine a leaf node that has an intersection with the view frustum; determining a visual distance of the user, determining a rendering level according to the visual distance of the user, rendering the leaf node that has the intersection with the view frustum according to the rendering level, and determining at least one visual attribute for the rendered leaf node.

7. The method of visualizing machining process monitoring data of claim 1, wherein, The method comprises the following steps: obtaining coordinate data, and generating an absolute timestamp of the coordinate data according to clock information of the numerical control machine tool; when the coordinate data is collected, no matter what coordinate system the coordinate data is in, an absolute timestamp of the coordinate data with high precision is generated according to clock information of a master clock of a control system of the numerical control machine tool; obtaining machining monitoring data, and generating an absolute timestamp of the machining monitoring data according to clock information of the numerical control machine tool.

8. A device for visualizing machining process monitoring data, characterized by The device comprises: The data acquisition module is configured to acquire coordinate data of the tool and machining monitoring data of the workpiece cutting process. The coordinate data includes machine tool coordinate data, absolute coordinate data, and relative coordinate data. The machine tool coordinate data is used to represent the absolute position of the tool relative to the machine tool origin. The absolute coordinate data is used to represent the position of the tool relative to the origin of the workpiece coordinate system. The relative coordinate data is used to represent the position of the tool relative to the end point of the previous program segment. The machining monitoring data is a physical quantity parameter in the cutting process. The data fusion module is configured to perform data fusion of the coordinate data and the machining monitoring data based on real-time stream processing and real-time linear interpolation of historical data to obtain fused data. The scene construction module is configured to construct a three-dimensional scene of the workpiece cutting process based on a finite element network and place the fused data in the three-dimensional scene. The data rendering module is configured to render the fused data located in a view frustum according to the three-dimensional scene and determine a rendering depth according to the visual distance between the fused data and a user.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps of the machining process monitoring data visualization method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the machining process monitoring data visualization method according to any one of claims 1 to 7.

Images