A multi-blade large-size cutter disc wear prediction and early warning method, device and medium
By using multi-sensor data fusion and machine learning algorithms, a multi-blade large-size tool turret wear prediction and early warning system was constructed. This system solved the problems of difficulty in quantifying tool wear status and predicting tool life, and enabled real-time monitoring of tool status and accurate prediction of wear, thereby improving machining quality and efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- DONGFANG ELECTRIC MACHINERY
- Filing Date
- 2025-09-02
- Publication Date
- 2026-06-26
AI Technical Summary
In the machining of large-size multi-blade cutter heads, the tool wear status is difficult to quantify intuitively, the tool life is difficult to predict, and the uneven wear of multiple blades leads to unstable machining quality. Traditional methods are difficult to monitor and accurately predict the tool wear status in real time.
By employing multi-sensor data fusion, machine learning algorithms, and physical models, a multi-channel data synchronous acquisition system is constructed by installing vibration and current sensors. Combining convolutional neural networks, bidirectional long short-term memory networks, and graph neural networks, a wear prediction model is built, a wear distribution heat map is generated, and online adaptive learning is performed to achieve accurate prediction of tool wear and life.
It improves tool utilization, reduces production costs, enhances machining quality and efficiency, strengthens the controllability of the machining process, reduces unplanned maintenance, and ensures machining safety and quality stability.
Smart Images
Figure CN121018277B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of mechanical manufacturing and control technology, and in particular to the process control of discrete manufacturing of large machine tools, specifically to a method, equipment and medium for predicting and warning the wear of multi-blade large-size tool turrets. Background Technology
[0002] The statements in this section are provided only as background information in connection with this disclosure and may not constitute prior art.
[0003] A multi-insert large-size cutter head typically has multiple inserts (20-32) to accommodate the machining needs of workpieces of different sizes and shapes. For example, indexable face milling cutters (such as the 850, 890, and 920 series) are widely used in rotor milling machines for roughing, semi-finishing, and finishing complex parts such as the lower grooves and crescent grooves on the shafts of large steam turbine generators. However, the following technical challenges exist in the practical application of multi-insert large-size cutter heads:
[0004] 1. Difficulty in intuitively quantifying tool wear: In the machining process of large-size multi-blade cutter heads, the tool wear status cannot be monitored in real time through intuitive or quantitative methods, making it difficult for operators to judge the health status of the tools in a timely manner, thus affecting machining quality and efficiency.
[0005] 2. Difficulty in predicting tool life: Due to the complexity and randomness of tool wear, traditional empirical prediction methods are difficult to accurately predict the cutting life limit of tools, resulting in tools failing prematurely before reaching their theoretical life or continuing to be used after reaching their life limit, which increases machining risks and production costs.
[0006] 3. Uneven wear of multi-insert tools: During the machining process, due to factors such as the uneven distribution of cutting forces and the layout of the inserts, the wear state of each insert often varies significantly, leading to a decline in the overall performance of the tool. In some cases, some inserts may be damaged while others can still be used, further increasing the difficulty of machining quality control.
[0007] 4. Unstable machining quality due to high randomness: The wear of multi-blade tools is characterized by high randomness and non-linearity. Traditional methods are difficult to effectively capture its wear pattern, resulting in large fluctuations in quality during the machining process, making it difficult to meet the requirements of high-precision machining.
[0008] To address these issues, existing technologies lack a systematic method capable of real-time monitoring of tool wear, accurate prediction of tool life, and effective resolution of multi-insert wear imbalance. Therefore, developing a method for predicting wear and lifespan of large-size multi-insert toolheads based on multi-sensor data fusion, machine learning algorithms, and physical models has significant theoretical and practical value. Summary of the Invention
[0009] The purpose of this invention is to address the problems of uneven and uncertain wear of multi-blade large-size cutterheads, which leads to excessive wear of some blades and inaccurate tool wear identification and poor adaptability. This invention provides a method, device, and medium for predicting and warning the wear of multi-blade large-size cutterheads, which can effectively improve tool utilization, ensure machining safety and quality stability, reduce and avoid machining quality loss, thereby solving the above-mentioned problems.
[0010] The technical solution of the present invention is as follows:
[0011] A method for predicting and warning of wear on a large-size multi-blade cutter head, comprising:
[0012] Step S1: Synchronous Acquisition and Feature Fusion of Multi-Source Heterogeneous Data
[0013] A vibration sensor is installed at the isomorphic part of the machine tool cutter head, and a current sensor is configured at the spindle end to build a multi-channel data synchronous acquisition system. The vibration signal and the current signal are synchronized at the millisecond level through timestamp alignment technology to generate a multimodal dataset that integrates spatiotemporal features.
[0014] Step S2: Construction of a single-insert wear-life co-prediction model:
[0015] Based on a multimodal dataset, a cascaded deep learning architecture is designed. The front-end uses a convolutional neural network (CNN) to extract the frequency domain spatial features of the signal, while the back-end uses a bidirectional long short-term memory network (Bi-LSTM) to model the temporal dependence of the signal. The model outputs a wear prediction value for the wear amount of a single cutting tool. The wear prediction value is combined with the tool physical degradation model, and a mapping relationship between wear amount and remaining life is established through a fully connected network (Dense) to generate a probabilistic life prediction interval.
[0016] Step S3: Multi-blade collaborative wear distribution modeling and dynamic compensation:
[0017] Based on the cutterhead layout topology, a wear correlation model between blades based on graph neural networks is constructed. The wear influence weight of adjacent blades is quantified through an attention mechanism. Combined with the prediction results of step S2, a heat map of multi-blade wear distribution is generated.
[0018] Step S4: Online Adaptive Learning and Prediction Robustness Enhancement
[0019] Based on the offline-trained CNN-Bi-LSTM model, an online learning module is deployed. An incremental random forest algorithm and a Bayesian optimization framework are used to fuse newly collected processed data in real time and dynamically update the model parameters to achieve adaptive calibration of the prediction model. At the same time, Monte Carlo sampling is used to simulate the uncertainty boundary of the prediction results and output the lifetime prediction value with confidence interval.
[0020] Furthermore, in step S1, the multi-channel data synchronous acquisition system uses Beckhoff data processing module and EtherCAT bus module to realize signal acquisition and preprocessing, and ensures that the data delay error is less than 1ms through hardware time synchronization protocol.
[0021] Furthermore, the probabilistic lifetime prediction interval mentioned in step S2 is implemented in the following way:
[0022] A quantile regression layer is introduced at the end of the CNN-Bi-LSTM network, which outputs several quantile values of wear amount. The probability density function of remaining lifetime is then generated by fitting the Weibull distribution.
[0023] Furthermore, the P10, P50, and P90 quantiles of the wear amount are output.
[0024] Furthermore, the graph neural network described in step S3 employs a graph attention network.
[0025] Furthermore, the graph attention network dynamically calculates the association weights between blade nodes in the following way:
[0026] Step a: Perform a linear transformation on the feature vector of each blade to obtain the mapped feature vector;
[0027] Step b: Concatenate the mapped feature vectors into a joint feature vector, and calculate its attention score using the trainable parameter vector;
[0028] Step c: After applying the LeakyReLU nonlinear activation function to the attention score, calculate the unnormalized attention coefficients of the current blade node i and its neighboring node j;
[0029] Step d: Repeat steps a-c for all neighboring nodes k∈N(i), and generate the final attention weights between node i and each neighboring node through exponential function and normalization processing. The values reflect the contribution of neighboring blades to the current blade wear state.
[0030] Furthermore, the incremental random forest algorithm described in step S4 uses a Hoeffding tree structure.
[0031] Furthermore, the data distribution characteristics of the latest N samples are preserved through a sliding window mechanism, and the model update frequency is synchronized with the machine tool processing cycle.
[0032] The present invention also proposes an electronic device, comprising:
[0033] At least one processor; and a memory communicatively connected to said at least one processor;
[0034] The memory stores instructions that can be executed by the at least one processor, and the at least one processor executes the instructions stored in the memory to perform the method described above.
[0035] The present invention also proposes a computer-readable storage medium for storing instructions that, when executed, cause the method described above to be implemented.
[0036] Compared with existing technologies, the advantages of this invention are:
[0037] 1. Improve tool utilization: Through accurate wear and life prediction, the health status of tools can be monitored in real time, avoiding premature tool replacement or overuse, significantly improving tool utilization and reducing production costs.
[0038] 2. Improve machining quality and efficiency: Real-time early warning and multi-insert wear balance management functions can effectively reduce machining quality fluctuations caused by tool wear, and improve machining efficiency and product consistency.
[0039] 3. Reduce maintenance costs: By predicting tool life and wear condition, preventive tool maintenance can be achieved, reducing unexpected downtime and unplanned maintenance, and lowering equipment maintenance costs.
[0040] 4. Enhanced controllability of the machining process: The application of multi-sensor data fusion and multi-blade wear distribution model makes the tool status during the machining process more transparent and controllable, providing reliable technical support for intelligent manufacturing. Attached Figure Description
[0041] Figure 1 Flowchart of a method for predicting and warning wear of multi-blade large-size tool turrets;
[0042] Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0043] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0044] The features and performance of the present invention will be further described in detail below with reference to embodiments.
[0045] Example 1
[0046] A multi-insert large-size cutter head typically has multiple inserts (20-32) to accommodate the machining needs of workpieces of different sizes and shapes. For example, indexable face milling cutters (such as the 850, 890, and 920 series) are widely used in rotor milling machines for roughing, semi-finishing, and finishing complex parts such as the lower grooves and crescent grooves on the shafts of large steam turbine generators. However, the following technical challenges exist in the practical application of multi-insert large-size cutter heads:
[0047] 1. Difficulty in intuitively quantifying tool wear: In the machining process of large-size multi-blade cutter heads, the tool wear status cannot be monitored in real time through intuitive or quantitative methods, making it difficult for operators to judge the health status of the tools in a timely manner, thus affecting machining quality and efficiency.
[0048] 2. Difficulty in predicting tool life: Due to the complexity and randomness of tool wear, traditional empirical prediction methods are difficult to accurately predict the cutting life limit of tools, resulting in tools failing prematurely before reaching their theoretical life or continuing to be used after reaching their life limit, which increases machining risks and production costs.
[0049] 3. Uneven wear of multi-insert tools: During the machining process, due to factors such as the uneven distribution of cutting forces and the layout of the inserts, the wear state of each insert often varies significantly, leading to a decline in the overall performance of the tool. In some cases, some inserts may be damaged while others can still be used, further increasing the difficulty of machining quality control.
[0050] 4. Unstable machining quality due to high randomness: The wear of multi-blade tools is characterized by high randomness and non-linearity. Traditional methods are difficult to effectively capture its wear pattern, resulting in large fluctuations in quality during the machining process, making it difficult to meet the requirements of high-precision machining.
[0051] To address these issues, existing technologies lack a systematic method capable of real-time monitoring of tool wear, accurate prediction of tool life, and effective resolution of multi-insert wear imbalance. Therefore, developing a method for predicting wear and lifespan of large-size multi-insert toolheads based on multi-sensor data fusion, machine learning algorithms, and physical models has significant theoretical and practical value.
[0052] This embodiment addresses the aforementioned problems by providing a method, device, and medium for predicting and warning of wear on large-size multi-blade cutterheads, which can effectively improve tool utilization, ensure machining safety and quality stability, and reduce and avoid machining quality losses.
[0053] In this embodiment, for details, please refer to... Figure 1 A method for predicting and warning of wear on a multi-blade, large-size cutterhead, comprising:
[0054] Step S1: Synchronous Acquisition and Feature Fusion of Multi-Source Heterogeneous Data
[0055] A vibration sensor is installed at the isomorphic part of the machine tool cutter head, and a current sensor is configured at the spindle end to build a multi-channel data synchronous acquisition system. The vibration signal and the current signal are synchronized at the millisecond level through timestamp alignment technology to generate a multimodal dataset that integrates spatiotemporal features.
[0056] Step S2: Construction of a single-insert wear-life co-prediction model:
[0057] Based on a multimodal dataset, a cascaded deep learning architecture is designed. The front-end uses a convolutional neural network (CNN) to extract the frequency domain spatial features of the signal, while the back-end uses a bidirectional long short-term memory network (Bi-LSTM) to model the temporal dependence of the signal. The model outputs a wear prediction value for the wear amount of a single cutting tool. The wear prediction value is combined with the tool physical degradation model, and a mapping relationship between wear amount and remaining life is established through a fully connected network (Dense) to generate a probabilistic life prediction interval.
[0058] Step S3: Multi-blade collaborative wear distribution modeling and dynamic compensation:
[0059] Based on the cutterhead layout topology, a wear correlation model between blades based on graph neural networks is constructed. The wear influence weight of adjacent blades is quantified through an attention mechanism. Combined with the prediction results of step S2, a heat map of multi-blade wear distribution is generated.
[0060] Step S4: Online Adaptive Learning and Prediction Robustness Enhancement
[0061] Based on the offline-trained CNN-Bi-LSTM model, an online learning module is deployed. An incremental random forest algorithm and a Bayesian optimization framework are used to fuse newly collected processed data in real time and dynamically update the model parameters to achieve adaptive calibration of the prediction model. At the same time, Monte Carlo sampling is used to simulate the uncertainty boundary of the prediction results and output the lifetime prediction value with confidence interval.
[0062] In this embodiment, specifically, the multi-channel data synchronous acquisition system in step S1 uses Beckhoff data processing module and EtherCAT bus module to realize signal acquisition and preprocessing, and ensures that the data delay error is less than 1ms through hardware time synchronization protocol.
[0063] In this embodiment, specifically, the probabilistic lifetime prediction interval in step S2 is implemented in the following way:
[0064] A quantile regression layer is introduced at the end of the CNN-Bi-LSTM network, which outputs the P10, P50, and P90 quantile values of wear amount. The probability density function of remaining lifetime is generated by fitting the Weibull distribution.
[0065] In this embodiment, specifically, the graph neural network described in step S3 is a graph attention network.
[0066] In this embodiment, the graph attention network dynamically calculates the association weights between blade nodes in the following way:
[0067] Step a: Perform a linear transformation on the feature vector of each blade to obtain the mapped feature vector;
[0068] Step b: Concatenate the mapped feature vectors into a joint feature vector, and calculate its attention score using the trainable parameter vector;
[0069] Step c: After applying the LeakyReLU nonlinear activation function to the attention score, calculate the unnormalized attention coefficients of the current blade node i and its neighboring node j;
[0070] Step d: Repeat steps a-c for all neighboring nodes k∈N(i), and generate the final attention weights between node i and each neighboring node through exponential function and normalization processing. The values reflect the contribution of neighboring blades to the current blade wear state.
[0071] In this embodiment, specifically, the incremental random forest algorithm described in step S4 adopts a Hoeffding tree structure and retains the data distribution characteristics of the latest N samples through a sliding window mechanism. The model update frequency is synchronized with the machine tool processing cycle.
[0072] Based on the same technical concept, embodiments of the present invention also provide an electronic device that can implement the multi-blade large-size cutterhead wear prediction and early warning method provided in the above embodiments of the present invention. In one embodiment, the electronic device can be a server, a terminal device, or other electronic equipment. Figure 2 As shown, the electronic device may include:
[0073] At least one processor and a memory connected to the at least one processor. In this embodiment of the invention, the specific connection medium between the processor and the memory is not limited. Figure 2 The example used is the connection between the processor and memory via a bus. The bus... Figure 2 The connections between other components are indicated by thick lines and are for illustrative purposes only, not as limiting information. Buses can be divided into address buses, data buses, control buses, etc., but for ease of representation, [the specific bus type is not shown here]. Figure 2 The processor is represented by a single thick line, but this does not imply that there is only one bus or one type of bus. Alternatively, a processor can also be called a controller; there are no restrictions on the name.
[0074] In this embodiment of the invention, the memory stores instructions executable by at least one processor. By executing the instructions stored in the memory, the at least one processor can perform the aforementioned method for predicting and warning the wear of a multi-blade, large-size tool turret. The processor can implement... Figure 2 The functions of each module in the device shown.
[0075] The processor is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory and calling data stored in memory, it can monitor the device's various functions and process data, thereby enabling overall monitoring of the device.
[0076] In an alternative design, the processor may include one or more processing units. The processor may integrate an application processor and a modem processor, wherein the application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may also not be integrated into the processor. In some embodiments, the processor and memory may be implemented on the same chip; in some embodiments, they may also be implemented separately on separate chips.
[0077] The processor can be a general-purpose processor, such as a CPU, digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the multi-blade large-size tool disk wear prediction and early warning method disclosed in the embodiments of this invention can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.
[0078] Memory, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory can include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic memory, magnetic disk, optical disk, etc. Memory is any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. In embodiments of the present invention, memory can also be a circuit or any other device capable of implementing storage functions, used to store program instructions and / or data.
[0079] By designing and programming the processor, the code corresponding to the multi-blade large-size tool disk wear prediction and early warning method described in the foregoing embodiments can be embedded into the chip, thereby enabling the chip to execute the steps of the method described in the foregoing embodiments during operation. How to design and program the processor is a technique well known to those skilled in the art, and will not be elaborated here.
[0080] Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing computer instructions, which, when executed on a computer, cause the computer to perform the aforementioned method for predicting and warning the wear of a multi-blade large-size cutterhead.
[0081] In some alternative embodiments, the present invention also provides a method for predicting and warning wear of a multi-blade large-size cutterhead, which can also be implemented as a program product including program code. When the program product is run on a device, the program code is used to cause the control device to perform the steps in the method for predicting and warning wear of a multi-blade large-size cutterhead according to various exemplary embodiments of the present invention as described above.
[0082] It should be noted that although several units or sub-units of the apparatus have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the invention, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units. Furthermore, although the operation of the method of the invention is described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.
[0083] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0084] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a server, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0085] Program code for performing the operations of this invention can be written using any combination of one or more programming languages, including object-oriented programming languages such as Java and C++, as well as conventional procedural programming languages such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0086] In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).
[0087] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0088] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0089] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.
[0090] This background section is provided to generally present the context of the invention. The work of the currently named inventors, the work to the extent described in this background section, and aspects of this section that did not constitute prior art at the time of application are neither expressly nor impliedly acknowledged as prior art to the invention.
Claims
1. A method for predicting and warning of wear on a large-size multi-blade cutter head, characterized in that, include: Step S1: Synchronous Acquisition and Feature Fusion of Multi-Source Heterogeneous Data A vibration sensor is installed at the isomorphic part of the machine tool cutter head, and a current sensor is configured at the spindle end to build a multi-channel data synchronous acquisition system. The vibration signal and the current signal are synchronized at the millisecond level through timestamp alignment technology to generate a multimodal dataset that integrates spatiotemporal features. Step S2: Construction of a single-insert wear-life co-prediction model: Based on a multimodal dataset, a cascaded deep learning architecture is designed. The front-end uses a convolutional neural network (CNN) to extract the frequency domain spatial features of the signal, while the back-end uses a bidirectional long short-term memory network (Bi-LSTM) to model the temporal dependence of the signal. The model outputs a wear prediction value for the wear amount of a single cutting tool. The wear prediction value is combined with the tool physical degradation model, and a mapping relationship between wear amount and remaining life is established through a fully connected network (Dense) to generate a probabilistic life prediction interval. Step S3: Multi-blade collaborative wear distribution modeling and dynamic compensation: Based on the cutterhead layout topology, a wear correlation model between blades based on graph neural networks is constructed. The wear influence weight of adjacent blades is quantified through an attention mechanism. Combined with the prediction results of step S2, a heat map of multi-blade wear distribution is generated. Step S4: Online Adaptive Learning and Prediction Robustness Enhancement Based on the offline-trained CNN-Bi-LSTM model, an online learning module is deployed. An incremental random forest algorithm and a Bayesian optimization framework are used to fuse newly collected processed data in real time and dynamically update the model parameters to achieve adaptive calibration of the prediction model. At the same time, Monte Carlo sampling is used to simulate the uncertainty boundary of the prediction results and output the lifetime prediction value with confidence interval.
2. The method for predicting and warning wear of a large-size multi-blade cutter head according to claim 1, characterized in that, In step S1, the multi-channel data synchronous acquisition system uses Beckhoff data processing module and EtherCAT bus module to realize signal acquisition and preprocessing, and ensures that the data delay error is less than 1ms through hardware time synchronization protocol.
3. The method for predicting and warning wear of a large-size multi-blade cutter head according to claim 1, characterized in that, The probabilistic lifetime prediction interval mentioned in step S2 is achieved in the following way: A quantile regression layer is introduced at the end of the CNN-Bi-LSTM network, which outputs several quantile values of wear amount. The probability density function of remaining lifetime is then generated by fitting the Weibull distribution.
4. The method for predicting and warning wear of a large-size multi-blade cutter head according to claim 1, characterized in that, Output the P10, P50, and P90 percentile values of wear amount.
5. The method for predicting and warning wear of a large-size multi-blade cutter head according to claim 1, characterized in that, The graph neural network described in step S3 employs a graph attention network.
6. The method for predicting and warning wear of a large-size multi-blade cutter head according to claim 5, characterized in that, Graph attention networks dynamically calculate the association weights between blade nodes in the following way: Step a: Perform a linear transformation on the feature vector of each blade to obtain the mapped feature vector; Step b: Concatenate the mapped feature vectors into a joint feature vector, and calculate its attention score using the trainable parameter vector; Step c: After applying the LeakyReLU nonlinear activation function to the attention score, calculate the unnormalized attention coefficients of the current blade node i and its neighboring node j; Step d: Repeat steps a-c for all neighboring nodes k∈N(i), and generate the final attention weights between node i and each of its neighboring nodes through exponential function and normalization. Its value reflects the contribution of adjacent blades to the current blade wear state.
7. The method for predicting and warning wear of a large-size multi-blade cutter head according to claim 1, characterized in that, The incremental random forest algorithm described in step S4 uses a Hoeffding tree structure.
8. The method for predicting and warning wear of a large-size multi-blade cutter head according to claim 7, characterized in that, The sliding window mechanism preserves the data distribution characteristics of the latest N samples, and the model update frequency is synchronized with the machine tool processing cycle.
9. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which executes the instructions stored in the memory to perform the method as described in any one of claims 1-8.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store instructions that, when executed, cause the method as described in any one of claims 1-8 to be implemented.