Tool health state dynamic evaluation method and system based on digital twinning

Abstract

The present application relates to the technical field of tool management, and particularly relates to a tool magazine tool health state dynamic evaluation method and system based on digital twinning. When the tool is returned to the warehouse, the tool service data of the tool magazine tool cabin is received from the tool warehouse request end; the tool life cycle data is traced back; the health factor threshold list is obtained by searching and statistically analyzing a plurality of healthy tool samples, taking the tool service task list, the tool service time length list, the tool working parameter and the tool execution task as constraints; the tool damage evaluation is carried out based on the tool health factor list and the health factor threshold list, and the tool predicted damage distribution information is generated; and the tool digital twinning model is constructed according to the tool predicted damage distribution information to execute the tool health state evaluation. The accuracy and reliability of the tool health state evaluation are improved.

Description

Technical Field

[0001] This invention relates to the field of tool management technology, and in particular to a method and system for dynamic assessment of the health status of tools in a tool magazine based on digital twins. Background Technology

[0002] In the field of tool health assessment, traditional methods primarily monitor parameters characterizing tool health, such as temperature, vibration, force, and sound signals, and compare these parameters with preset threshold values ​​to determine the tool's health status based on these differences. However, the threshold values ​​used in traditional methods are static data. In actual operation, the tool's condition fluctuates continuously. Because static data cannot adapt to these fluctuations in real time, accurate assessment of tool health is difficult, potentially leading to misjudgments. This can affect the proper use and maintenance of tools, increasing uncertainty and costs in the production process, resulting in technical problems related to the accuracy and reliability of tool health assessment. Summary of the Invention

[0003] This invention addresses the technical problem of low accuracy and reliability in tool health status assessment in existing technologies by providing a method and system for dynamic assessment of tool health status based on digital twins.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0005] In a first aspect, the present invention provides a method for dynamic assessment of the health status of tools in a tool magazine based on digital twins, comprising: when a tool is returned to the magazine, receiving tool service data from the tool magazine's tool compartment from a tool removal request end, wherein the tool service data includes tool health factors, tool operating parameters, and tool execution tasks; tracing back tool lifecycle data, wherein the tool lifecycle data includes a tool service task list and a tool service duration list; using the tool service task list, the tool service duration list, the tool operating parameters, and the tool execution tasks as constraints, retrieving several healthy tool samples for statistical analysis to obtain a health factor threshold list; performing tool damage assessment based on the tool health factor list and the health factor threshold list to generate tool predicted damage distribution information; and constructing a tool digital twin model based on the tool predicted damage distribution information to perform tool health status assessment.

[0006] Optionally, the tool service data may also include tool operating time, including updating the tool lifecycle data based on the task performed by the tool and the tool operating time.

[0007] Optionally, using the tool service task list, the tool service duration list, the tool operating parameters, and the tool execution tasks as constraints, a number of healthy tool samples are retrieved for statistical analysis to obtain a health factor threshold list. This includes: constructing primary constraints based on the tool service task list and the tool service duration list; constructing secondary constraints based on the tool operating parameters and the tool execution tasks; collecting a number of primary healthy tool samples that satisfy the primary constraints; collecting a number of healthy tool samples that satisfy the secondary constraints from the primary healthy tool samples; performing a common-attribute central tendency analysis on the health factors of the number of healthy tool samples to obtain the distribution range of representative values ​​of multiple health factor attributes, and constructing the health factor threshold list.

[0008] Specifically, the process involves performing a common-attribute central tendency analysis on the health factors of the several healthy tool samples to obtain the distribution intervals of representative values ​​for multiple health factor attributes, and constructing the health factor threshold list. This includes: obtaining the nominal regions for multiple health factor attributes; performing common-attribute intersection interval statistics on the distribution intervals of representative values ​​for multiple health factor attributes and the nominal regions for multiple health factor attributes to obtain the fitting intervals for multiple health factor attributes, and constructing the health factor threshold list.

[0009] Specifically, based on the tool service task list and the tool service duration list, a first-level constraint condition is constructed, including: extracting a first tool service task type and a first task type service duration from the tool service task list and the tool service duration list; using the first tool service task type as a constraint, collecting a number of pre-service tool damage states, a number of post-service tool damage states, and a number of tool service durations that correspond one-to-one; performing a consistency comparison on the number of pre-service tool damage states and the number of post-service tool damage states to obtain a number of consistency comparison results; and based on the number of consistency comparison results, extracting selected tools whose comparison results are inconsistent from the number of tool service durations. Service duration set; calculate the set value of the selected tool service duration set, and set it as the first tool service task type sensitive duration; when the first task type service duration is less than the first tool service task type sensitive duration, delete the first tool service task type and the first task type service duration from the tool service task list and the tool service duration list; when the first task type service duration is greater than or equal to the first tool service task type sensitive duration, do not make any changes; when the tool service task list and the tool service duration list have been traversed, obtain all updated tool service task lists and updated tool service duration lists, and construct the first-level constraint conditions.

[0010] Optionally, tool damage assessment is performed based on the tool health factor list and the health factor threshold list to generate tool predicted damage distribution information. This includes: matching a tool damage assessment model from a tool damage assessment model library based on the tool's task and tool model, wherein the tool damage assessment model is trained using machine learning with multiple sets of data, and each set of data includes tool health factor record data, health factor threshold record data, and labels identifying tool damage distribution information for a preset task type; inputting the tool health factor list and the health factor threshold list into the tool damage assessment model, and outputting the tool predicted damage distribution information.

[0011] The tool damage assessment model is trained using machine learning with multiple sets of data. Each set of data includes tool health factor record data, health factor threshold record data, and labels identifying tool damage distribution information for a preset task type. The process includes: randomly configuring tool health factor record data and health factor threshold record data based on the tool health factor's rated range; collecting several tool damage distribution information for a preset tool model, constrained by the preset task type, the tool health factor record data, and the health factor threshold record data; calculating the concentrated value of damage with the same location and attribute based on the several tool damage distribution information to obtain labels identifying tool damage distribution information; and storing the tool damage assessment model, the preset tool model, and the preset task type in association, adding them to the tool damage assessment model library.

[0012] Optionally, based on the predicted tool damage distribution information, a tool digital twin model is constructed to perform a tool health status assessment, including: obtaining a first execution task, wherein the first execution task has a predefined tool baseline digital twin model; when the tool baseline digital twin model is inconsistent with the tool digital twin model, the first execution task is marked as abnormal; when the tool baseline digital twin model is consistent with the tool digital twin model, the first execution task is marked as normal; the abnormal mark or the normal mark, combined with the first execution task, is added to the tool health status assessment result and sent to the tool management terminal.

[0013] Secondly, the present invention provides a dynamic assessment system for the health status of tool magazine tools based on digital twins, comprising:

[0014] The service data acquisition module is used to receive tool service data from the tool magazine tool compartment from the tool release request terminal when the tool is returned to the magazine. The tool service data includes tool health factors, tool working parameters and tool execution tasks.

[0015] The tool life cycle tracking module is used to track tool life cycle data, wherein the tool life cycle data includes a tool service task list and a tool service duration list;

[0016] The healthy cutting tool analysis module is used to retrieve several healthy cutting tool samples for statistical analysis based on the cutting tool service task list, the cutting tool service duration list, the cutting tool working parameters, and the cutting tool execution tasks, and to obtain a list of health factor thresholds.

[0017] The tool damage assessment module is used to assess tool damage based on the tool health factor list and the health factor threshold list, and generate tool predicted damage distribution information.

[0018] The health status assessment module is used to construct a digital twin model of the tool based on the tool's predicted damage distribution information to perform tool health status assessment.

[0019] By implementing this invention, when a tool is returned to its storage compartment, the tool service data of the tool magazine tool compartment can be received from the tool release request terminal. The tool service data includes tool health factors, tool working parameters, and tool execution tasks, ensuring that the acquired data is real-time information directly related to the actual service status of the tool, and providing accurate and effective data support for subsequent steps.

[0020] By implementing this invention, it is possible to trace back tool lifecycle data, wherein the tool lifecycle data includes a tool service task list and a tool service duration list. By tracing back the tool lifecycle data, the usage history of the tool can be grasped from a macro perspective, providing a basis for the subsequent construction and analysis of constraints based on the tool service history, making the evaluation more comprehensive and reliable.

[0021] By implementing this invention, it is possible to retrieve several healthy tool samples for statistical analysis, constrained by the tool service task list, the tool service duration list, the tool working parameters, and the tool execution tasks, to obtain a health factor threshold list. By retrieving and analyzing healthy tool samples through multi-dimensional constraints, the resulting health factor threshold list can better fit the actual service conditions of the tool. It is no longer a static fixed threshold, but a threshold range that dynamically adapts to different service conditions of the tool, thus improving the accuracy and applicability of the thresholds.

[0022] By implementing this invention, tool damage assessment can be performed based on the tool health factor list and the health factor threshold list, generating tool predicted damage distribution information. By comparing and analyzing the dynamic health factor threshold list with the actual tool health factor list, the tool damage can be assessed more accurately. The generated predicted damage distribution information provides a specific reference for subsequent in-depth assessment of the tool health status.

[0023] By implementing this invention, a digital twin model of the tool can be constructed based on the predicted damage distribution information of the tool to perform tool health status assessment, thereby achieving accurate grasp and real-time monitoring of the tool health status and providing scientific and effective decision support for tool management and maintenance.

[0024] In summary, implementing this invention can improve the accuracy and reliability of tool health status assessment. Attached Figure Description

[0025] Figure 1 A flowchart illustrating the dynamic assessment method for tool health status based on digital twin provided by this invention;

[0026] Figure 2 A schematic diagram of the structure of the tool magazine tool health status dynamic evaluation system based on digital twin provided by the present invention.

[0027] In the attached diagram, the components represented by each number are as follows:

[0028] Service data acquisition module 11, tool life traceability module 12, healthy tool analysis module 13, tool damage assessment module 14, and health status assessment module 15. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0031] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0032] Example 1, as Figure 1 As shown, this embodiment of the invention provides a method for dynamic evaluation of the health status of tools in a tool magazine based on digital twins, including:

[0033] S100: When the tool is returned to the storage compartment, the tool service data of the tool magazine tool compartment is received from the tool release request terminal. The tool service data includes tool health factor, tool working parameters and tool execution task.

[0034] S200: Backtrack tool lifecycle data, wherein the tool lifecycle data includes a tool service task list and a tool service duration list;

[0035] S300: Using the tool service task list, the tool service duration list, the tool working parameters, and the tool execution task as constraints, retrieve several healthy tool samples for statistical analysis to obtain a health factor threshold list;

[0036] S400: Based on the tool health factor list and the health factor threshold list, perform tool damage assessment and generate tool predicted damage distribution information;

[0037] S500: Based on the predicted damage distribution information of the tool, construct a digital twin model of the tool to perform a tool health status assessment.

[0038] In step S100 of this embodiment, when the tool is returned to the magazine, tool service data from the tool magazine's tool compartment is received from the tool release request terminal. This tool service data includes tool health factors, tool operating parameters, and the task being performed by the tool.

[0039] When a tool completes its work and is returned to its storage compartment, the system retrieves tool service data from the tool magazine's tool compartment from the tool release request end (such as the machine tool control system or production management system). Specifically, the physical return of the tool to its storage compartment can be used as a signal source, with sensors (such as position sensors or RFID readers) detecting the tool's storage status and triggering the data reception process. The data is then sent from the tool release request end to the tool magazine management system via industrial Ethernet, fieldbus (such as Profinet or EtherCAT), or IoT communication protocols (such as MQTT). Within the tool magazine, each tool corresponds one-to-one with its tool compartment, and each tool has a unique identifier, such as an RFID tag ID or barcode number.

[0040] Tool health factors include parameters that characterize the tool's health status, such as the maximum temperature signal (e.g., 82℃), the maximum vibration amplitude signal (12.5m / s²), and the maximum X-axis cutting force signal (e.g., 350N) during the cutting process. These parameters can be collected in real time by built-in sensors or external monitoring devices.

[0041] The tool operating parameters are machining process parameters such as cutting speed (m / min), feed rate (mm / r), and depth of cut (mm), which are derived from the operation records of the machine tool control system.

[0042] The tool execution task records the type of machining task undertaken by the tool, such as external turning, end milling, drilling, etc., which is generated by the production management system when assigning tasks.

[0043] For example, a tool with model number "M-500" performed a face milling task A. Its service data can be of the following types: tool health factor [maximum vibration amplitude 15m / s², maximum temperature 75℃]; tool working parameters [cutting speed: 150m / min, feed rate: 0.1mm / r, depth of cut: 3mm]; tool task [face milling].

[0044] In step S100 of this application embodiment, the tool service data further includes tool working time, including: updating the tool life cycle data based on the task performed by the tool and the tool working time.

[0045] Tool working time is a key indicator for measuring tool physical wear and is directly related to tool life. By recording the actual cutting time, the shortcomings of traditional static thresholds (such as fixed number of machining operations) can be avoided, as they cannot reflect the actual load.

[0046] The tool lifecycle data (tool task list + duration list) is dynamically updated with each tool return to storage, forming a "digital history" for the tool. For example, this lifecycle data could be in the form of: Task B: External turning, tool working time 2 hours; Task C: External turning, tool working time 1.5 hours; Task D: External turning, tool working time 0.5 hours, and so on. The tool lifecycle data is updated based on the tool's executed tasks and working times.

[0047] In step S200 of this application embodiment, the tool lifecycle data is traced back, wherein the tool lifecycle data includes a tool service task list and a tool service duration list.

[0048] The step S300 of this embodiment, which involves retrieving several healthy tool samples for statistical analysis, is based on tool lifecycle data. Therefore, before executing step S300, it is necessary to first trace back the tool lifecycle data, that is, to extract the complete list of tool service tasks and tool service duration recorded after the target tool enters the tool magazine.

[0049] In step S300 of this application embodiment, several healthy tool samples are retrieved and statistically analyzed based on constraints such as the tool service task list, the tool service duration list, the tool working parameters, and the tool execution tasks to obtain a health factor threshold list, including:

[0050] Based on the tool service task list and the tool service duration list, first-level constraints are constructed, and based on the tool working parameters and the tool execution tasks, second-level constraints are constructed.

[0051] Collect several first-level healthy cutting tool samples that meet the first-level constraint conditions;

[0052] From the plurality of first-level healthy tool samples, collect the plurality of healthy tool samples that satisfy the second-level constraint conditions;

[0053] A central tendency analysis of the health factors of the several healthy cutting tool samples is performed to obtain the distribution range of representative values ​​of multiple health factor attributes, and a health factor threshold list is constructed.

[0054] In this embodiment of the application, this step is to filter matching healthy tool samples based on the tool's historical service data (i.e., the tool service task list and the tool service duration list) and the current task parameters (i.e., the tool working parameters and the tool execution task), and generate a dynamic health factor threshold list through statistical analysis to solve the problem that traditional static thresholds cannot adapt to changes in working conditions.

[0055] In step S300 of this application embodiment, based on the tool service task list and the tool service duration list, a first-level constraint condition is constructed, including:

[0056] Extract the first tool service task type and the first task type service duration from the tool service task list and the tool service duration list;

[0057] Based on the first tool service task type, collect a number of tool damage states before service, a number of tool damage states after service, and a number of tool service durations that correspond one-to-one.

[0058] A consistency comparison is performed on the pre-service tool damage states and the post-service tool damage states to obtain several consistency comparison results.

[0059] Based on the aforementioned consistency comparison results, a set of selected tool service durations whose comparison results are inconsistent is extracted from the aforementioned tool service durations.

[0060] Calculate the set of concentrated values ​​of the selected tool service durations and set them as the first tool service task type sensitive duration;

[0061] When the service duration of the first task type is less than the sensitive service duration of the first tool service task type, the first tool service task type and the service duration of the first task type are deleted from the tool service task list and the tool service duration list. When the service duration of the first task type is greater than or equal to the sensitive service duration of the first tool service task type, no changes are made.

[0062] Once the tool service task list and the tool service duration list have been traversed, the complete updated tool service task list and updated tool service duration list are obtained, and the first-level constraint conditions are constructed.

[0063] Suppose that a certain tool T-600 needs to perform an "external turning" task with a service time of 1.5 hours, and a first-level constraint needs to be constructed.

[0064] First, using "external turning" as a constraint, healthy tool samples (such as 10 cutting tool samples of the same model) for the same type of task are retrieved from the historical database. Several tool damage states before service, several tool damage states after service, and several tool service durations are collected for each group of samples.

[0065] The aforementioned service durations are obtained from the aforementioned list of tool service durations, such as 1.0h, 1.5h, 2.0h, 2.5h, etc.

[0066] The tool damage status can be measured by the amount of rake face wear. Specifically, before and after each use of the tool, the amount of rake face wear VB (unit: mm) is measured to an accuracy of 0.01 mm using an optical microscope or a 3D scanner to obtain tool damage status information.

[0067] The pre-service damage condition can be defined as follows: rake face wear VB = 0.05 mm; the post-service damage condition can be defined as VB = 0.1 mm. If the change in damage condition before and after service is less than or equal to the threshold (e.g., VB change ≤ 0.05 mm), it is considered "consistent"; otherwise, it is considered "inconsistent". These are the results of the aforementioned consistency comparisons.

[0068] Next, from the aforementioned tool service durations (e.g., 1.0h, 1.5h, 2.0h, 2.5h, etc.), a set of selected tool service durations (e.g., 2.0h, 2.5h) whose comparison results are inconsistent is extracted. The set of concentrated values ​​for these selected tool service durations is then calculated. This concentrated value can be the mean, such as (2.0+2.5) / 2=2.25h. This concentrated value, 2.25h, can then be defined as the sensitive duration for the "external turning" task (i.e., the first tool service task type).

[0069] Then, if the service duration of the first task type (e.g., external turning) is less than the sensitive service duration of the first tool task type (e.g., 2.25 hours), the first tool task type and its service duration are removed from the tool service task list and the tool service duration list. If the service duration of the first task type (e.g., 3.0 hours) is greater than or equal to the sensitive service duration of the first tool task type (e.g., 2.25 hours), no changes are made. If the tool reaches the scrap standard after performing a certain task, the service task corresponding to that task is removed.

[0070] Once the tool service task list and the tool service duration list have been traversed, and all updated tool service task lists and updated tool service duration lists are obtained, the first-level constraint condition is constructed. In the example above, this means removing all "external turning" tasks with a duration <2.25h from the tool service task list, leaving only valid data for tasks with a duration ≥2.25h. Using this method, several first-level healthy tool samples that satisfy the first-level constraint condition can be obtained.

[0071] Furthermore, secondary constraints need to be constructed based on the tool's operating parameters, the task it performs, and the primary constraints. These secondary constraints can be the operating parameters. For example, suppose a certain tool T-600 needs to perform an "external turning" task for 1.5 hours, requiring secondary constraints. If the tool's operating parameters for this task are a cutting speed of 150 m / min and a feed rate of 0.2 mm / r, then the secondary constraints could be: cutting speed 150 m / min ± 10% (135–165 m / min), feed rate 0.2 mm / r ± 20% (0.16–0.24 mm / r). Then, samples that simultaneously satisfy the secondary constraints are selected from several primary healthy tool samples. For example:

[0072] For example, if the working parameters of 3 out of 5 sets of first-level healthy tool samples are within the above range, then they become the several healthy tool samples that satisfy the second-level constraint conditions.

[0073] In step S300 of this application embodiment, a central tendency analysis of the health factors of the plurality of healthy tool samples is performed to obtain the distribution range of representative values ​​of multiple health factor attributes, and the health factor threshold list is constructed, including:

[0074] Obtain multiple health factor attribute rating regions;

[0075] By performing statistical analysis on the intersection intervals of the representative values ​​of the multiple health factor attributes and the nominal regions of the multiple health factor attributes, fitting intervals of multiple health factor attributes are obtained, and the health factor threshold list is constructed.

[0076] In this embodiment of the application, to perform a central tendency analysis of the health factors of the several healthy tool samples, obtain the distribution range of representative values ​​of multiple health factor attributes, and construct the health factor threshold list, it is first necessary to obtain the health factor data of several (e.g., 3 groups) healthy tool samples (e.g., maximum vibration amplitude m / s² during cutting, maximum temperature ℃, maximum cutting force N along the X-axis).

[0077] For example, in the above example, it is necessary to collect health factor data from three healthy knife samples, such as sample A (12.5 m / s², 78℃, 280 N), sample B (11.8 m / s², 82℃, 270 N), and sample C (13.2 m / s², 75℃, 290 N).

[0078] For example, the process of constructing threshold ranges for health factors is introduced using the maximum vibration amplitude as an example.

[0079] First, the tool material and design standards need to be determined, and the rated range of vibration amplitude needs to be set: [10~15m / s²] (i.e., normal working range).

[0080] Then, the maximum vibration amplitude of the three groups of healthy cutting tool samples was statistically analyzed. The mean was (12.5+11.8+13.2) / 3≈12.5m / s², and the standard deviation was σ≈0.68m / s². Therefore, the distribution range can be the mean ±2σ, which is [12.5-1.36,12.5+1.36]=[11.14,13.86]m / s².

[0081] Next, the intersection of the rated area and the distribution interval needs to be calculated. In this example, the distribution interval [11.14, 13.86] ∩ the rated area [10, 15] = [11.14, 13.86] m / s². However, if the distribution interval exceeds the rated area (e.g., [8, 16]), the intersection will be [10, 15].

[0082] The intersection of the aforementioned rated area and the distribution interval, i.e., the fitting interval, is used as the maximum vibration amplitude threshold.

[0083] Using the same method, thresholds for health factors such as maximum temperature and maximum cutting force can be calculated, generating a list of these health factor thresholds. For example, when the rated range for the maximum temperature is [60, 90]℃, its fitting range is [75, 82]℃. When the rated range for the maximum cutting force on the X-axis is [200, 350]N, its fitting range is [270, 290]N.

[0084] The health factor threshold list can be constructed using the methods described above.

[0085] In step S400 of this application embodiment, tool damage assessment is performed based on the tool health factor list and the health factor threshold list to generate tool predicted damage distribution information, including:

[0086] Based on the tool's execution task and tool model, a tool damage assessment model is matched from a tool damage assessment model library. The tool damage assessment model is trained using machine learning through multiple sets of data. Each set of data includes tool health factor record data, health factor threshold record data, and labels that identify tool damage distribution information for a preset task type.

[0087] Input the list of tool health factors and the list of health factor thresholds into the tool damage assessment model, and output the tool predicted damage distribution information.

[0088] In step S400 of this embodiment, a machine learning model is used to analyze the tool health factors and threshold data to predict the damage distribution of various parts of the tool (such as the amount of tool surface wear), providing damage state input for digital twin modeling.

[0089] In step S400 of this application embodiment, the tool damage assessment model is trained using machine learning through multiple sets of data. Each set of data includes tool health factor record data of a preset task type, health factor threshold record data, and labels identifying tool damage distribution information, including:

[0090] Based on the rated range of tool health factors, randomly configure tool health factor recording data and health factor threshold recording data;

[0091] Using the preset task type, the tool health factor recorded data, and the health factor threshold recorded data as constraints, several tool damage distribution information are collected for a preset tool model;

[0092] Based on the aforementioned tool damage distribution information, calculate the concentrated value of damage with the same location and attribute to obtain a label that identifies the tool damage distribution information.

[0093] The tool damage assessment model, preset tool model, and preset task type are associated and stored, and added to the tool damage assessment model library.

[0094] The input data for the tool damage assessment model includes a list of health factors and a list of health factor thresholds. The output data is the predicted tool damage distribution information. The health factor list contains the specific health factor data for the tool currently performing the task, such as 12.5 m / s², 78℃, and 280 N; the health factor threshold list is obtained through the aforementioned steps. The predicted tool damage distribution information can be the tool wear location and wear degree. Through machine learning model analysis, a set of predicted damage distribution information can be generated as follows: predicted wear amount for the rake face (VB) is 0.25 mm; and the predicted wear amount for the flank face (VC) is 0.18 mm. The sample data used to train the tool damage assessment model should be no less than 5000 sets (covering different tool models, machining materials, and cutting parameters).

[0095] The basis for this prediction is that if the maximum vibration amplitude is close to the threshold, it indicates a decrease in tool stability; while if the maximum tool temperature is too high, it will accelerate the softening of the tool material and increase the wear rate.

[0096] Among them, collecting several tool damage distribution information can be achieved by using an optical microscope or a 3D scanner to measure the tool rake face wear VB (unit: mm) and flank face wear VC (unit: mm) before and after each use of the tool, with an accuracy of 0.01 mm, under the constraints of preset task type, tool health factor recorded data and the health factor threshold.

[0097] Based on the aforementioned tool damage distribution information, a cluster value calculation for damage of the same location and attribute is performed to obtain labels identifying the tool damage distribution information. These labels are then grouped by wear location (rake face, flank face) and attribute (VB, VC wear amount), for example, all rake face VB data are grouped together. The mean value is then calculated for each group; for example, the mean VB wear amount on the rake face is 0.2 mm, and the mean VC wear amount on the flank face is 0.15 mm. The dispersed damage data is transformed into labels suitable for model training, ensuring that the labels reflect the clustering trend and distribution range of damage of the same location and attribute. The labeled data is then used to train the tool damage assessment model.

[0098] Alternatively, the tool damage assessment model can be constructed using a multilayer perceptron (MLP) neural network. The number of neurons in the input layer is the same as the number of input features; there are two hidden layers with 16 neurons each, using ReLU as the activation function; the output layer has two neurons (VB wear prediction value and VC wear prediction value); the loss function is mean squared error (MSE); and the optimizer is Adam (with a learning rate of 0.001, β1=0.9, and β2=0.999).

[0099] In training the tool damage assessment model, the batch size was set to 32; the number of training rounds was 200; the validation set ratio was 20%; the training was stopped if the validation loss did not decrease for 10 consecutive rounds to prevent overfitting; when the validation set MSE decreased by less than 0.001 for 10 consecutive rounds, the model was considered to have converged and the tool damage assessment model was obtained.

[0100] In this embodiment, the wear mechanisms and health factor thresholds of different tool types, materials (e.g., cemented carbide / ceramic), and task types (e.g., turning / milling) vary significantly. Therefore, it is necessary to train multiple independent tool damage assessment models according to the training method described above, based on different health factor record data, health factor threshold record data, and labels identifying tool damage distribution information corresponding to different tool types and task types, to correspond to different tool types and task types.

[0101] Due to the complexity of tool conditions in practical applications, associating tool damage assessment models with tool type and task type allows the models to adapt to the unique characteristics of different tools and tasks. Therefore, it is necessary to associate and store tool damage assessment models, preset tool types, and preset task types, adding them to the tool damage assessment model library. This involves first extracting these three types of information, then establishing their relationships, for example, using tool type and task type as index keys to store the corresponding models as values. Finally, according to this association structure, the data is added to the tool damage assessment model library for subsequent quick retrieval of matching assessment models based on tool type and task type.

[0102] Finally, the tool health factor list and the health factor threshold list are input into the tool damage assessment model to output the tool predicted damage distribution information (e.g., the rake face wear VB is 0.2mm and the flank face wear VC is 0.15mm).

[0103] In step S500 of this application embodiment, a tool digital twin model is constructed based on the tool predicted damage distribution information to perform a tool health status assessment, including:

[0104] Obtain a first execution task, wherein the first execution task has a predefined tool reference digital twin model;

[0105] When the tool reference digital twin model is inconsistent with the tool digital twin model, an anomaly is marked for the first execution task;

[0106] When the tool reference digital twin model is consistent with the tool digital twin model, the first execution task is normally identified;

[0107] The abnormal or normal identifier, combined with the first executed task, is added to the tool health status assessment result and sent to the tool management terminal.

[0108] In this embodiment of the application, a reference digital twin model is obtained, for example, based on the current first execution task (such as "external turning"), a predefined tool reference digital twin model (such as the standard three-dimensional model of a T-700 turning tool, which includes material properties, geometric parameters and ideal wear threshold) is retrieved from the model library.

[0109] The predicted damage distribution information generated by S400 (e.g., the wear amount VB on the rake face is 0.2 mm and the wear amount VC on the flank face is 0.15 mm) is input into the baseline model to drive the model parameter update and generate a digital twin model reflecting the current tool state (e.g., the wear area on the tool face is color-marked).

[0110] If the wear and damage location of the actual model exceed the ideal threshold of the baseline model (e.g., VB > 0.3mm), it is judged as "inconsistent" and the task is marked as "abnormal"; if all parameters are within the range of the baseline model, it is judged as "consistent" and the task is marked as "normal".

[0111] Finally, the identification result (abnormal / normal) and the first executed task information (such as tool model, task type, etc.) are packaged and sent to the tool management terminal for operators to monitor or make decisions (such as triggering a tool change warning).

[0112] The method for obtaining the tool reference digital twin model is existing technology and well-known to those skilled in the art, and will not be elaborated here. Example 2, as... Figure 2 As shown, based on the same inventive concept as the method for dynamic evaluation of tool health status of a tool magazine based on digital twin provided in Embodiment 1, this embodiment of the invention also provides a system for dynamic evaluation of tool health status of a tool magazine based on digital twin, comprising:

[0113] The service data acquisition module 11 is used to receive tool service data from the tool magazine tool compartment from the tool release request terminal when the tool is returned to the magazine. The tool service data includes tool health factors, tool working parameters and tool execution tasks.

[0114] The tool life traceability module 12 is used to trace tool life cycle data, wherein the tool life cycle data includes a tool service task list and a tool service duration list;

[0115] The healthy cutting tool analysis module 13 is used to retrieve several healthy cutting tool samples for statistical analysis based on the cutting tool service task list, the cutting tool service duration list, the cutting tool working parameters and the cutting tool execution tasks, and to obtain a list of health factor thresholds.

[0116] Tool damage assessment module 14 is used to assess tool damage based on the tool health factor list and the health factor threshold list, and generate tool predicted damage distribution information;

[0117] The health status assessment module 15 is used to construct a digital twin model of the tool based on the tool's predicted damage distribution information to perform a tool health status assessment.

[0118] Furthermore, the service data acquisition module 11 includes the following execution steps:

[0119] The tool lifecycle data is updated based on the task performed by the tool and the tool's operating time.

[0120] Furthermore, the healthy tool analysis module 13 includes the following execution steps:

[0121] Based on the tool service task list and the tool service duration list, first-level constraints are constructed, and based on the tool working parameters and the tool execution tasks, second-level constraints are constructed.

[0122] Collect several first-level healthy cutting tool samples that meet the first-level constraint conditions;

[0123] From the plurality of first-level healthy tool samples, collect the plurality of healthy tool samples that satisfy the second-level constraint conditions;

[0124] A central tendency analysis of the health factors of the several healthy cutting tool samples is performed to obtain the distribution range of representative values ​​of multiple health factor attributes, and a health factor threshold list is constructed.

[0125] Specifically, a central tendency analysis of the health factors of the several healthy tool samples is performed to obtain the distribution range of representative values ​​of multiple health factor attributes, and a health factor threshold list is constructed, including:

[0126] Obtain multiple health factor attribute rating regions;

[0127] By performing statistical analysis on the intersection intervals of the representative values ​​of the multiple health factor attributes and the nominal regions of the multiple health factor attributes, fitting intervals of multiple health factor attributes are obtained, and the health factor threshold list is constructed.

[0128] Based on the tool service task list and the tool service duration list, first-level constraints are constructed, including:

[0129] Extract the first tool service task type and the first task type service duration from the tool service task list and the tool service duration list;

[0130] Based on the first tool service task type, collect a number of tool damage states before service, a number of tool damage states after service, and a number of tool service durations that correspond one-to-one.

[0131] A consistency comparison is performed on the pre-service tool damage states and the post-service tool damage states to obtain several consistency comparison results.

[0132] Based on the aforementioned consistency comparison results, a set of selected tool service durations whose comparison results are inconsistent is extracted from the aforementioned tool service durations.

[0133] Calculate the set of concentrated values ​​of the selected tool service durations and set them as the first tool service task type sensitive duration;

[0134] When the service duration of the first task type is less than the sensitive service duration of the first tool service task type, the first tool service task type and the service duration of the first task type are deleted from the tool service task list and the tool service duration list. When the service duration of the first task type is greater than or equal to the sensitive service duration of the first tool service task type, no changes are made.

[0135] Once the tool service task list and the tool service duration list have been traversed, the complete updated tool service task list and updated tool service duration list are obtained, and the first-level constraint conditions are constructed.

[0136] Furthermore, the tool damage assessment module 14 includes the following execution steps:

[0137] Based on the tool health factor list and the health factor threshold list, tool damage is assessed to generate predicted tool damage distribution information, including:

[0138] Based on the tool's execution task and tool model, a tool damage assessment model is matched from a tool damage assessment model library. The tool damage assessment model is trained using machine learning through multiple sets of data. Each set of data includes tool health factor record data, health factor threshold record data, and labels that identify tool damage distribution information for a preset task type.

[0139] Input the list of tool health factors and the list of health factor thresholds into the tool damage assessment model, and output the tool predicted damage distribution information.

[0140] The tool damage assessment model is trained using machine learning with multiple sets of data. Each set of data includes tool health factor record data for a preset task type, health factor threshold record data, and labels identifying tool damage distribution information, including:

[0141] Based on the rated range of tool health factors, randomly configure tool health factor recording data and health factor threshold recording data;

[0142] Using the preset task type, the tool health factor recorded data, and the health factor threshold recorded data as constraints, several tool damage distribution information are collected for a preset tool model;

[0143] Based on the aforementioned tool damage distribution information, calculate the concentrated value of damage with the same location and attribute to obtain a label that identifies the tool damage distribution information.

[0144] The tool damage assessment model, preset tool model, and preset task type are associated and stored, and added to the tool damage assessment model library.

[0145] Furthermore, the health status assessment module 15 includes the following execution steps:

[0146] Obtain a first execution task, wherein the first execution task has a predefined tool reference digital twin model;

[0147] When the tool reference digital twin model is inconsistent with the tool digital twin model, an anomaly is marked for the first execution task;

[0148] When the tool reference digital twin model is consistent with the tool digital twin model, the first execution task is normally identified;

[0149] The abnormal or normal identifier, combined with the first executed task, is added to the tool health status assessment result and sent to the tool management terminal.

[0150] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0151] Those skilled in the art will understand that embodiments of the present invention can provide 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.

[0152] 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, as well as 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 computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0153] 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.

[0154] 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.

[0155] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0156] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dynamic assessment of tool health status in a tool magazine based on digital twins, characterized in that, include: When the tool is returned to the storage compartment, the tool service data of the tool magazine tool compartment is received from the tool release request terminal. The tool service data includes tool health factors, tool working parameters and tool execution tasks. The tool lifecycle data is traced back, including a list of tool service tasks and a list of tool service durations. Using the tool service task list, the tool service duration list, the tool working parameters, and the tool execution tasks as constraints, several healthy tool samples are retrieved and statistically analyzed to obtain a health factor threshold list. Based on the tool health factor list and the health factor threshold list, tool damage is assessed, and tool predicted damage distribution information is generated. Based on the predicted damage distribution information of the cutting tool, a digital twin model of the cutting tool is constructed to perform a cutting tool health status assessment; Specifically, using the tool service task list, the tool service duration list, the tool operating parameters, and the tool execution tasks as constraints, several healthy tool samples are retrieved for statistical analysis to obtain a health factor threshold list, including: Based on the tool service task list and the tool service duration list, first-level constraints are constructed, and based on the tool working parameters and the tool execution tasks, second-level constraints are constructed. Collect several first-level healthy cutting tool samples that meet the first-level constraint conditions; From the plurality of first-level healthy tool samples, collect the plurality of healthy tool samples that satisfy the second-level constraint conditions; A central tendency analysis of the same attribute is performed on the health factors of the several healthy tool samples to obtain the distribution range of representative values ​​of multiple health factor attributes, and the health factor threshold list is constructed. Specifically, a central tendency analysis of the health factors of the several healthy tool samples is performed to obtain the distribution intervals of representative values ​​of multiple health factor attributes, and a health factor threshold list is constructed, including: Obtain multiple health factor attribute rating regions; By performing statistical analysis on the intersection intervals of the representative values ​​of the multiple health factor attributes and the nominal regions of the multiple health factor attributes, fitting intervals of multiple health factor attributes are obtained, and the health factor threshold list is constructed.

2. The method as described in claim 1, characterized in that, The tool service data also includes tool operating time, including: The tool lifecycle data is updated based on the task performed by the tool and the tool's operating time.

3. The method as described in claim 1, characterized in that, Based on the tool service task list and the tool service duration list, first-level constraints are constructed, including: Extract the first tool service task type and the first task type service duration from the tool service task list and the tool service duration list; Based on the first tool service task type, collect a number of tool damage states before service, a number of tool damage states after service, and a number of tool service durations that correspond one-to-one. A consistency comparison is performed on the pre-service tool damage states and the post-service tool damage states to obtain several consistency comparison results. Based on the aforementioned consistency comparison results, a set of selected tool service durations whose comparison results are inconsistent is extracted from the aforementioned tool service durations. Calculate the set of concentrated values ​​of the selected tool service durations and set them as the first tool service task type sensitive duration; When the service duration of the first task type is less than the sensitive service duration of the first tool service task type, the first tool service task type and the service duration of the first task type are deleted from the tool service task list and the tool service duration list. When the service duration of the first task type is greater than or equal to the sensitive service duration of the first tool service task type, no changes are made. Once the tool service task list and the tool service duration list have been traversed, the complete updated tool service task list and updated tool service duration list are obtained, and the first-level constraint conditions are constructed.

4. The method as described in claim 1, characterized in that, Based on the tool health factor list and the health factor threshold list, tool damage is assessed to generate predicted tool damage distribution information, including: Based on the tool's execution task and tool model, a tool damage assessment model is matched from a tool damage assessment model library. The tool damage assessment model is trained using machine learning through multiple sets of data. Each set of data includes tool health factor record data, health factor threshold record data, and labels that identify tool damage distribution information for a preset task type. Input the list of tool health factors and the list of health factor thresholds into the tool damage assessment model, and output the tool predicted damage distribution information.

5. The method as described in claim 4, characterized in that, The tool damage assessment model is trained using machine learning with multiple sets of data. Each set of data includes tool health factor record data for a preset task type, health factor threshold record data, and labels identifying tool damage distribution information, including: Based on the rated range of tool health factors, randomly configure tool health factor recording data and health factor threshold recording data; Using the preset task type, the tool health factor recorded data, and the health factor threshold recorded data as constraints, several tool damage distribution information are collected for a preset tool model; Based on the aforementioned tool damage distribution information, calculate the concentrated value of damage with the same location and attribute to obtain a label that identifies the tool damage distribution information. The tool damage assessment model, preset tool model, and preset task type are associated and stored, and added to the tool damage assessment model library.

6. The method as described in claim 1, characterized in that, Based on the predicted tool damage distribution information, a digital twin model of the tool is constructed to perform a tool health status assessment, including: Obtain a first execution task, wherein the first execution task has a predefined tool reference digital twin model; When the tool reference digital twin model is inconsistent with the tool digital twin model, an anomaly is marked for the first execution task; When the tool reference digital twin model is consistent with the tool digital twin model, the first execution task is normally identified; The abnormal or normal identifier, combined with the first executed task, is added to the tool health status assessment result and sent to the tool management terminal.

7. A dynamic assessment system for the health status of tool magazines based on digital twins, characterized in that, include: The service data acquisition module is used to receive tool service data from the tool magazine tool compartment from the tool release request terminal when the tool is returned to the magazine. The tool service data includes tool health factors, tool working parameters and tool execution tasks. The tool life cycle tracking module is used to track tool life cycle data, wherein the tool life cycle data includes a tool service task list and a tool service duration list; The healthy cutting tool analysis module is used to retrieve several healthy cutting tool samples for statistical analysis based on the cutting tool service task list, the cutting tool service duration list, the cutting tool working parameters, and the cutting tool execution tasks, and to obtain a list of health factor thresholds. The tool damage assessment module is used to assess tool damage based on the tool health factor list and the health factor threshold list, and generate tool predicted damage distribution information. The health status assessment module is used to construct a digital twin model of the tool based on the tool's predicted damage distribution information to perform tool health status assessment. Specifically, using the tool service task list, the tool service duration list, the tool operating parameters, and the tool execution tasks as constraints, several healthy tool samples are retrieved for statistical analysis to obtain a health factor threshold list, including: Based on the tool service task list and the tool service duration list, first-level constraints are constructed, and based on the tool working parameters and the tool execution tasks, second-level constraints are constructed. Collect several first-level healthy cutting tool samples that meet the first-level constraint conditions; From the plurality of first-level healthy tool samples, collect the plurality of healthy tool samples that satisfy the second-level constraint conditions; A central tendency analysis of the same attribute is performed on the health factors of the several healthy tool samples to obtain the distribution range of representative values ​​of multiple health factor attributes, and the health factor threshold list is constructed. Specifically, a central tendency analysis of the health factors of the several healthy tool samples is performed to obtain the distribution intervals of representative values ​​of multiple health factor attributes, and a health factor threshold list is constructed, including: Obtain multiple health factor attribute rating regions; By performing statistical analysis on the intersection intervals of the representative values ​​of the multiple health factor attributes and the nominal regions of the multiple health factor attributes, fitting intervals of multiple health factor attributes are obtained, and the health factor threshold list is constructed.

Images