Numerical control machine tool cutter life prediction system and method and medium

Abstract

The invention relates to the technical field of cutter management, in particular to a numerical control machine tool cutter service life prediction system and method and a medium, and the system comprises a cutter information obtaining module which is used for obtaining the machining duration, the number of machined parts and the machining information of machining state imaging of a cutter; the cutter state analysis module is used for performing state analysis on the cutter in a staged manner on the basis of machining information acquisition of the cutter; the tool service life prediction module analyzes the machining state of the tool through the tool state analysis module, predicts the service life of the tool of which the machining state is detected at present through a tool service life prediction model established in advance, and sets a service life use threshold value based on the prediction of the service life of the tool; and judging the continuation of the use of the tool by comparing the predicted value of the service life of the tool with a service life use threshold value. The method can predict the service life of the tool in stages, and has the advantage of reducing the difficulty of automatic analysis and judgment.

Description

Technical Field

[0001] The present invention relates to the technical field of tool management, and more specifically, it relates to a tool life prediction system, method and medium for a numerically controlled machine tool. Background Art

[0002] Numerical control tools are tools used for cutting in machining, also known as cutting tools. In a broad sense, cutting tools include both tools and abrasives; at the same time, "numerical control tools" include accessories such as tool shanks and tool holders in addition to the cutting blades. After common numerical control tools are damaged in use, they are usually replaced immediately by manual labor. Therefore, manual assistance must be carried out to avoid operations such as idle machining or incorrect machining.

[0003] For example, the Chinese invention patent with the patent name of an intelligent tool control method and system based on a machine tool and the patent number of 202011508269.2 discloses the following: obtaining the state information of the machine tool in real time and analyzing in real time whether there is a machine tool in an operating state. If so, controlling a monitoring camera to capture the working image in real time and obtaining the operating information of the machine tool in real time, analyzing in real time whether the machine tool needs to replace a tool. If so, extracting the tool replacement information included in the operating information and comparing the tool replacement information with the stored tool information of the tool disc located in the machining area to analyze whether there is a matching tool. If so, analyzing in real time whether there is a problem with the tool matched by the tool disc. If so, controlling the moving mechanism to drive the tool disc to move back to the tool replacement area and controlling the moving mechanism matched by the tool replacement area to move to the machining area, and controlling the tool disc moved to the machining area to rotate the matched tool by using the first rotating shaft to correspond to the machining part and controlling the second rotating shaft of the tool to drive the tool to rotate perpendicular to the machining surface of the machining part.

[0004] Based on automated analysis to replace manual operation, analyzing in real time whether the machine tool needs to replace a tool, but analyzing in real time whether there is damage to the tools on the tool disc existing in the machining area requires constantly paying attention to the damage state of the tools. Taking pictures of them through a camera, and based on the analysis of the images, it takes an analysis time process to determine the degree of damage. Coupled with the requirements of real-time analysis, it is necessary to maintain a high-running state, which further increases the difficulty of the automated analysis determination. Summary of the Invention

[0005] Aiming at the above deficiencies of the prior art, the purpose of the present invention is to provide a tool life prediction system for a numerically controlled machine tool, which predicts the tool service life in stages and has the advantage of reducing the difficulty of automated analysis determination.

[0006] The above technical purpose of the present invention is achieved through the following technical solutions: A tool life prediction system for a numerically controlled machine tool includes

[0007] A tool information acquisition module, which is used to acquire the processing duration, the number of processed parts, and the processing information of the processing state imaging of the tool;

[0008] A tool status analysis module, which performs staged status analysis on the tool based on the acquisition of the tool's processing information;

[0009] A tool life prediction module, through the analysis of the tool processing state by the tool status analysis module, and using a pre-established tool life prediction model to predict the life of the tool in the current detected processing state. Based on the prediction of the tool life, a life usage threshold is set, and the continuation of the tool usage is determined by comparing the predicted value of the tool service life with the life usage threshold.

[0010] Preferably, the processing duration includes the actual total processing time of the processing tool and the actual processing section duration of the processing tool for the same product. Based on different time periods of the total processing time, the tool status analysis module is triggered in stages to analyze the tool processing state, and the actual processing duration of the same product processed by the tool is detected.

[0011] Preferably, based on the identification of the processing section duration, a duration threshold for triggering the tool status analysis module to analyze is set in stages. The longer the processing section duration, the more intensive the trigger times for triggering the tool status analysis module to analyze.

[0012] Preferably, the tool status analysis module includes, based on the understanding of the initial processing tool, by analyzing the difference between the processing tool and the initial tool, setting a scrap threshold depth for the processing tool, and comparing the current length of the processing tool with the scrap threshold depth of the processing tool to determine the tool scrap status.

[0013] Preferably, the tool scrap status includes mild damage, normal damage, and severe damage. The normal damage, moderate damage, and severe damage are divided according to different depths of the processing dimensions of the processing tool.

[0014] Preferably, the tool status analysis module uses a laser displacement sensor to measure the tool length in real time and further accurately measures the wear depth of the tool through visual inspection.

[0015] Preferably, the tool life prediction module includes a data transmission unit connected to the tool status analysis module and a decision-making unit for determining the available status of the tool based on the tool life prediction value. After using the data transmission unit to transmit the data analyzed by the tool status analysis module to the tool life prediction module, the tool life prediction model is used to predict the life of the tool in the current detected processing state, and then the decision-making unit determines the subsequent use of the tool.

[0016] On the one hand, a method for predicting the tool life of a numerically controlled machine tool is provided, including the following steps:

[0017] S1: Obtain the processing information of the tool, including the processing duration, the number of processed parts, and the imaging of the processing status, through the tool information acquisition module;

[0018] S2: Based on the acquisition of the tool's processing information, perform staged status analysis on the tool through the tool status analysis module;

[0019] S3: Through the analysis of the tool's processing status by the tool status analysis module, predict the life of the tool currently under inspection using a pre-established tool life prediction model. Based on the prediction of the tool life, set a life usage threshold in the tool life prediction module, and determine the continuation of the tool's use by comparing the predicted value of the tool's service life with the life usage threshold.

[0020] On the one hand, a computer-readable storage medium is provided, including instructions stored in the computer-readable storage medium. When the instructions are executed, the above-mentioned method for predicting the tool life of a numerically controlled machine tool is executed.

[0021] In summary, the beneficial effects of the present invention are as follows: The tool information acquisition module obtains the processing information of the tool, including the processing duration, the number of processed parts, and the imaging of the processing status. Based on the acquisition of the tool's processing information, perform staged status analysis on the tool through the tool status analysis module, and predict the life of the tool currently under inspection using a pre-established tool life prediction model. Based on the prediction of the tool life, set a life usage threshold in the tool life prediction module, and determine the continuation of the tool's use by comparing the predicted value of the tool's service life with the life usage threshold. By performing staged status analysis on the tool, the difficulty requirement of real-time analysis is reduced, and the high operating maintenance state of life prediction can be alleviated, further reducing the determination difficulty of automated analysis. Description of the Drawings

[0022] Figure 1 is the structural block diagram of the tool life prediction system for a numerically controlled machine tool according to an embodiment of the present invention;

[0023] Figure 2 is the step schematic diagram of the method for predicting the tool life of a numerically controlled machine tool according to an embodiment of the present invention.

[0024] Reference numerals: 1. Tool information acquisition module; 2. Tool status analysis module; 3. Tool life prediction module. Detailed Embodiments

[0025] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0026] It should be noted that when a component is referred to as being "fixed to" or "disposed on" another component, it can be directly on the other component or indirectly on the other component. When a component is referred to as being "connected to" another component, it can be directly or indirectly connected to the other component.

[0027] It should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0028] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0029] A tool life prediction system for a numerically controlled machine tool, see Figure 1 , including:

[0030] A tool information acquisition module 1 for acquiring the processing duration, the number of processed parts and the processing information of the processed state imaging of the tool;

[0031] A tool state analysis module 2 for stagewise analyzing the state of the tool based on the acquired processing information of the tool;

[0032] A tool life prediction module 3, through the analysis of the tool processing state by the tool state analysis module, and using a pre-established tool life prediction model to predict the life of the tool in the current detected processing state. Based on the prediction of the tool life, a life usage threshold is set, and the continuation of the tool usage is determined by comparing the predicted value of the tool service life with the life usage threshold.

[0033] In this embodiment, the tool information acquisition module 1 acquires the processing duration, the number of processed parts, and the processing information of the processing state imaging of the tool. Based on the acquisition of the tool's processing information, the tool state analysis module 2 performs phased state analysis on the tool, and predicts the life of the tool currently being detected for the processing state based on the previously established tool life prediction model. Based on the prediction of the tool life, in the tool life prediction module 3, a life usage threshold is set, and the continuation of the tool usage is determined by comparing the predicted value of the tool service life with the life usage threshold. By performing phased state analysis on the tool, the difficulty requirement of real-time analysis is reduced, and by being able to relieve the high operating maintenance state of life prediction, the determination difficulty of automated analysis is further reduced.

[0034] The processing duration includes the actual total processing time of the processing tool and the actual processing segment duration of the processing tool for the same product. Based on different time periods of the total processing time, the tool state analysis module 2 is triggered in stages to analyze the processing state of the tool, and the actual processing duration of the same product processed by the tool is detected.

[0035] Phased triggering strategies, for example:

[0036] Primary trigger: Trigger a regular analysis once every total processing duration exceeds 30 minutes or after processing 100 products.

[0037] Secondary trigger: When the processing segment duration suddenly exceeds the historical normal, that is, outside the range of ±2 times the standard deviation of the historical mean, immediately start an emergency analysis;

[0038] Heavy trigger: Set the detection duration distribution of the processing segment duration. For example, usually the actual processing segment duration for processing one product is 2 minutes. Based on the processing segment duration, set the duration thresholds of the processing segment duration, which are 2 minutes 30 seconds; 2 minutes 40 seconds; 2 minutes 45 seconds, etc.

[0039] The longer the processing segment duration, it may indirectly indicate tool damage. Therefore, based on the identification of the processing segment duration, set the duration thresholds for triggering the tool state analysis module 2 to analyze in stages. The longer the processing segment duration, the more intensive the trigger times for triggering the tool state analysis module 2 to analyze. The processing segment duration is the actual duration used for processing one product.

[0040] Similarly, in terms of the number of processed parts, for example, trigger a regular analysis once after processing 100 products. When processing the first 500 products, trigger a regular analysis once every 100 products processed; when processing another 200 products, trigger a regular analysis once every 50 products processed; when processing another 100 products, trigger a regular analysis once every 20 products processed. The more processed parts, the more intensive the trigger times for triggering the tool state analysis module to analyze.

[0041] Since the scrapped state of the tool is not only directly determined based on the processing duration, but the state of the tool is jointly determined by various influencing factors. Therefore, the tool state analysis module 2 includes setting a scrapped threshold depth for the processing tool by analyzing the difference between the processing tool and the initial tool based on the understanding of the initial processing tool, and comparing the current length of the processing tool with the scrapped threshold depth of the processing tool to determine the scrapped state of the tool.

[0042] By setting the tool scrapped state to include three states: mild damage, normal damage, and severe damage, the normal damage, moderate loss, and severe damage are divided according to different depths of the processing dimensions of the processing tool.

[0043] Quantify the damage level:

[0044] Mild damage: The tool length wear ≤ 0.1mm, or the wear is less than 1% of the initial length of the new tool;

[0045] Normal damage: The wear is 0.1mm - 0.3mm or the wear is greater than 1% - 3% of the initial length of the new tool;

[0046] Severe damage: The wear ≥ 0.3mm or the wear exceeds 3% of the initial length of the new tool.

[0047] The tool state analysis module 2 in this embodiment uses a laser displacement sensor to measure the tool length in real time, and further accurately measures the wear depth of the tool through visual detection, so as to accurately detect the wear degree of the tool.

[0048] In addition, the tool life prediction module 3 includes a data transmission unit connected to the tool state analysis module 2 and a decision-making unit for determining the available state of the tool based on the tool life prediction value. After using the data transmission unit to transmit the data analyzed by the tool state analysis module to the tool life prediction module, the tool life prediction model is used to predict the life of the tool in the current detected processing state, and then the decision-making unit determines the subsequent use of the tool.

[0049] After receiving the data of the tool state analysis module analyzing the state of the tool by the data transmission unit, the tool life prediction model established in advance is used to predict the life of the tool in the current detected processing state, and then a life usage threshold is set. By comparing the predicted value of the tool service life with the life usage threshold, the continuation of the tool use is determined, that is, the predicted value of the tool service life and the life usage threshold.

[0050] First set the basic threshold:

[0051] According to the theoretical life provided by the tool manufacturer, the average life of the tool is 500 minutes. Combining historical data, a life usage threshold is set. For example, the life usage threshold is 50 minutes;

[0052] When the predicted value of the tool life is less than or equal to 50 minutes, the tool life prediction module determines that the tool is scrapped. If the predicted value of the tool life is greater than 50 minutes, the tool life prediction module determines that the tool can continue to be used.

[0053] Regarding the establishment of the tool life prediction model,

[0054] 1. Data Definition and Input

[0055] According to the above requirements, the model only uses the following three types of data:

[0056] Processing duration: The cumulative usage time of the tool, calculated in minutes of installation.

[0057] Number of processed parts: The cumulative number of parts processed by the tool.

[0058] Processing status imaging: The image of the tool edge wear.

[0059] 2. Model Architecture Design

[0060] According to the data characteristics, a "time series - image fusion model" is constructed, and the specific structure is as follows:

[0061] The input layer includes a time series data branch and an image data branch. The time series data branch includes the processing duration and the number of processed parts and is transmitted through a fully connected network; the image data branch includes the processing status imaging and is transmitted through a convolutional neural network;

[0062] The feature fusion layer splices the time series and image features and conducts feature fusion analysis;

[0063] The prediction layer outputs the remaining tool life based on the feature fusion analysis

[0064] 3. Detailed Explanation of Model Components

[0065] 3.1 Time Series Data Branch

[0066] Input: Processing duration; Number of processed parts.

[0067] Processing flow:

[0068] Feature engineering:

[0069] Calculate the "number of parts processed per unit time": Number of parts / minute = Number of processed parts / Processing duration.

[0070] Generate the "processing stage label": Divide the cumulative duration into three stages: initial stage (0 - 30%), middle stage (30 - 70%), and late stage (70 - 100%).

[0071] Fully connected network:

[0072] Input: Processing duration, number of processed parts, parts per minute, processing stage.

[0073] Structure: Two-layer fully connected; for example, 64 neurons, ReLU activation.

[0074] Output: 128-dimensional time series feature vector.

[0075] 3.2 Image Data Branch

[0076] Input: Imaging of processing status, such as 512×512 pixel RGB image.

[0077] Processing flow:

[0078] Improved lightweight CNN:

[0079] Backbone network: MobileNetV3, pre-trained weights, adapted to industrial images.

[0080] Modification: The last layer is replaced with adaptive pooling + fully connected layer, outputting a 256-dimensional image feature vector.

[0081] Wear feature extraction:

[0082] Visualized by Grad-CAM to ensure that the model focuses on the tool edge wear area.

[0083] 3.3 Feature Fusion and Prediction Feature Concatenation: Concatenate the time series feature and the image feature into a 384-dimensional vector.

[0084] Prediction layer:

[0085] Structure: Two-layer fully connected, 128 neurons + 1 neuron, ReLU + linear activation. Output: Remaining life.

[0086] 4. Model Training and Optimization

[0087] 4.1 Data Preparation Example

[0088]

[0089] 4.2 Training Strategy

[0090] Loss function: Mean squared error + L2 regularization.

[0091] Optimizer: Adam.

[0092] Federated learning adaptation:

[0093] Local training models for each machine tool, only upload encrypted model parameters.

[0094] After aggregating the parameters in the cloud, they are sent down for update to adapt to different workshop conditions.

[0095] 4.3 Performance Verification

[0096] Dataset division: 70% for training, 15% for validation, and 15% for testing.

[0097] Expected metrics:

[0098] Mean Absolute Error: ±25 minutes.

[0099] Accuracy rate of key tool warning, that is, correct warning when remaining life ≤ threshold: ≥90%.

[0100] In addition, based on the above-mentioned numerical control machine tool tool life prediction system, a numerical control machine tool tool life prediction method is provided. See Figure 2 , including the following steps:

[0101] S1: Through the tool information acquisition module 1, acquire the processing duration, number of processed parts, and processing information of the imaging of the processing state of the tool;

[0102] S2: Based on the acquisition of the tool's processing information, stagewise analyze the tool's state through the tool state analysis module 2;

[0103] S3: Through the analysis of the tool's processing state by the tool state analysis module 2, and predict the life of the tool currently being detected for processing with the pre-established tool life prediction model. Based on the prediction of the tool life, in the tool life prediction module 3, set the life usage threshold, and determine the continuation of the tool usage by comparing the predicted value of the tool service life with the life usage threshold.

[0104] Based on the above-mentioned numerical control machine tool tool life prediction method, the beneficial effects of the corresponding method embodiments are realized accordingly, which will not be elaborated here.

[0105] It should be noted that in the embodiments of the present application Figure 1 The division of the modules of the numerical control machine tool tool life prediction system shown is schematic, only a logical function division. In actual implementation, there may be other division methods. In addition, in each embodiment of the present application, each functional unit can be integrated in a processing unit, can also exist separately physically, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware, can also be implemented in the form of software functional units, or can be implemented in the form of a combination of software and hardware.

[0106] It should be noted that in the embodiments of the present application, if the above-mentioned method is implemented in the form of software function modules and sold or used as an independent product, it can also be stored in a computer storage medium. Based on such an understanding, the technical solution of the embodiments of the present application, in essence or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing an electronic device to execute all or part of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), magnetic disks, or optical discs that can store program codes.

[0107] The embodiments of the present application provide a computer device, which may be a server. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above-mentioned method is implemented.

[0108] The embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the method provided in the above-mentioned embodiments are implemented.

[0109] The embodiments of the present application provide a computer program product containing instructions. When it runs on a computer, it causes the computer to execute the steps in the method provided in the above-mentioned method embodiments.

[0110] Those skilled in the art can understand that in one embodiment, the numerical control machine tool tool life prediction system provided by the present application can be implemented in the form of a computer program, and the computer program can run on a computer device. Each program module constituting the above-mentioned device can be stored in the memory of the computer device. The computer program constituted by each program module causes the processor to execute the steps in the methods described in the various embodiments of the present application in this specification.

[0111] It should be pointed out here that: the descriptions of the above storage medium and device embodiments are similar to the descriptions of the above method embodiments and have beneficial effects similar to those of the method embodiments. For the technical details not disclosed in the storage medium, storage medium, and device embodiments of the present application, please refer to the descriptions of the method embodiments of the present application for understanding.

[0112] The above embodiments are merely explanations of the present invention and are not limitations thereof. After reading this specification, those skilled in the art may make modifications to these embodiments that do not contribute creatively, but as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A tool life prediction system for a numerical control machine tool, characterized in that: Including: A tool information acquisition module, configured to acquire machining information such as the machining duration, the number of machined parts, and the machining state imaging of the tool; A tool state analysis module, which performs staged state analysis on the tool based on the acquired machining information of the tool; A tool life prediction module, through the analysis of the tool machining state by the tool state analysis module, and uses a pre-established tool life prediction model to predict the life of the tool with the current detected machining state. Based on the prediction of the tool life, a life usage threshold is set, and the continuation of the tool usage is determined by comparing the predicted value of the tool service life with the life usage threshold.

2. The tool life prediction system for a numerically controlled machine tool according to claim 1, characterized in that: The machining duration includes the actual total machining time of the machining tool and the actual machining section duration of the machining tool for the same product. Based on different time periods of the total machining time, the tool state analysis module is triggered in stages to analyze the machining state of the tool, and the actual machining duration of the same product machined by the tool is detected.

3. A tool life prediction system for a numerically controlled machine tool according to claim 2, characterized in that: Based on the identification of the machining section duration, a duration threshold for triggering the tool state analysis module to analyze is set in stages. The longer the machining section duration, the more intensive the triggering times for the tool state analysis module to analyze.

4. A tool life prediction system for a numerically controlled machine tool according to claim 1, characterized in that: The tool state analysis module includes, based on the understanding of the initial machining tool, by analyzing the difference between the machining tool and the initial tool, setting a scrapping threshold depth for the machining tool, and comparing the current length of the machining tool with the scrapping threshold depth of the machining tool to determine the scrapping state of the tool.

5. The tool life prediction system for a numerical control machine tool according to claim 4, characterized in that: The tool scrapping state includes mild damage, normal damage, and severe damage, and the normal damage, moderate loss, and severe damage are divided according to different depths of the machining dimensions of the machining tool.

6. The tool life prediction system for a numerical control machine tool according to claim 5, wherein: The tool state analysis module uses a laser displacement sensor to measure the tool length in real time and further accurately measures the wear depth of the tool through visual inspection.

7. The tool life prediction system for a numerical control machine tool according to claim 1, characterized in that: The tool life prediction module includes a data transmission unit connected to the tool state analysis module and a decision-making unit for determining the available state of the tool based on the tool life prediction value. After using the data transmission unit to transmit the data analyzed by the tool state analysis module to the tool life prediction module, the tool life prediction model is used to predict the life of the tool with the current detected machining state, and then the decision-making unit determines the subsequent use of the tool.

8. A tool life prediction method for a numerically controlled machine tool, characterized in that, Including the following steps: S1: Through the tool information acquisition module, acquire the machining information such as the machining duration, the number of machined parts, and the machining state imaging of the tool; S2: Based on the acquired machining information of the tool, perform staged state analysis on the tool through the tool state analysis module; S3: Through the analysis of the tool machining state by the tool state analysis module, and use a pre-established tool life prediction model to predict the life of the tool with the current detected machining state. Based on the prediction of the tool life, in the tool life prediction module, set a life usage threshold, and determine the continuation of the tool usage by comparing the predicted value of the tool service life with the life usage threshold.

9. A computer-readable storage medium, characterized in that, Including that the computer-readable storage medium stores instructions, and when the instructions are executed, the method as claimed in claim 8 is executed.

Images