A method and device for predicting tool life of CNC machine tools and CNC machine tools

Abstract

The present invention discloses a method and device for predicting tool life for CNC machine tools, and the CNC machine tool, relating to the technical field of CNC machine tools. The method comprises: collecting tool operation characteristic data in real time during the operation of the CNC machine tool, wherein the tool operation characteristic data includes multiple types of characteristic data; inputting the tool operation characteristic data into a tool life prediction model, and outputting a first tool life prediction result; obtaining an image of a finished product after material processing, combining it with a preset target finished product image, identifying processing similarity, and combining it with the first tool life prediction result to obtain a tool life prediction result through classification. The present invention solves the technical problem of inaccurate tool life prediction for CNC machine tools existing in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of numerically controlled machine tools, and in particular to a method and device for predicting tool life of numerically controlled machine tools and the numerically controlled machine tool. Background Art

[0002] When a CNC machine tool's turning tool experiences significant wear, it needs to be replaced. Conventional technology typically involves regularly inspecting and recording the tool's condition. When damage or near-damage (i.e., severe wear) is detected, the tool is replaced. Existing tool wear detection is labor-intensive. Furthermore, while higher-end turning tools are more reliable and less prone to sudden damage, other tools are susceptible to sudden damage. Failure to promptly identify and replace these tools can lead to material damage and even damage to the machine tool.

[0003] Some existing technologies propose using current changes in the motor driving the turning tool to determine tool damage or predict tool life. However, single-source predictions have low accuracy and require extensive training data, potentially leading to model overfitting. Consequently, existing technologies suffer from inaccurate tool life predictions for CNC machine tools. Summary of the Invention

[0004] The present application provides a method and device for predicting the tool life of a CNC machine tool, and a CNC machine tool, which are used to solve the technical problem of inaccurate tool life prediction of CNC machine tools in the prior art.

[0005] In view of the above problems, the present application provides a method and device for predicting tool life of a CNC machine tool and a CNC machine tool.

[0006] In a first aspect, the present application provides a method for predicting tool life of a CNC machine tool, the method comprising:

[0007] During the operation of the CNC machine tool, tool operation feature data is collected in real time, wherein the tool operation feature data includes multiple types of feature data;

[0008] Inputting the tool operation characteristic data into a tool life prediction model and outputting a first tool life prediction result;

[0009] The image of the finished product after material processing is obtained, combined with the preset target finished product image, the processing similarity is identified and obtained, and the tool life prediction result is obtained by classification in combination with the first tool life prediction result.

[0010] In a second aspect, the present application provides a device for predicting tool life of a CNC machine tool, the device comprising:

[0011] A tool data acquisition module is used to collect tool operation feature data in real time during the operation of the CNC machine tool, wherein the tool operation feature data includes multiple types of feature data;

[0012] A first life prediction module is used to input the tool operation characteristic data into a tool life prediction model and output a first tool life prediction result;

[0013] The life prediction classification module is used to obtain the image of the finished product after material processing, combine it with the preset target finished product image, identify and obtain the processing similarity, and combine it with the first tool life prediction result to classify and obtain the tool life prediction result.

[0014] In a third aspect, the present application provides a CNC machine tool, including the CNC machine tool tool life prediction device of the second aspect.

[0015] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0016] This application proposes a method for predicting tool life for CNC machine tools. The method collects tool operation feature data of multimodal data during the operation of the CNC machine tool, and then trains a tool life prediction model based on the multimodal data. The current tool operation feature data is input and predicted to obtain a first tool life prediction result. The image of the finished product after material processing is obtained, and the processing similarity is identified and obtained in combination with the preset target finished product image. The tool life prediction result is obtained by classification in combination with the first tool life prediction result. This application identifies the quality of material processing through a lightweight model, and then combines the tool life predicted by multimodal data, and finally classifies and predicts the tool life prediction result. With a small amount of data, the turning tool life is identified, achieving the technical effect of improving the accuracy of tool life prediction and improving the processing quality of CNC machine tools. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic flow chart of a method for predicting tool life of a CNC machine tool provided in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of the architecture of a tool life prediction model in a CNC machine tool tool life prediction method provided in an embodiment of the present application;

[0020] Figure 3A schematic diagram of the structure of a tool life prediction device for CNC machine tools provided in an embodiment of the present application.

[0021] In the accompanying drawings, the components represented by the reference numerals are described as follows:

[0022] Tool data acquisition module 11, first life prediction module 12, life prediction classification module 13. DETAILED DESCRIPTION

[0023] The present application provides a method and device for predicting the tool life of a CNC machine tool, and a CNC machine tool, so as to solve the technical problem of inaccurate tool life prediction of CNC machine tools in the prior art.

[0024] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0025] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0026] Example 1, as Figure 1 As shown, the present application provides a method for predicting tool life of a CNC machine tool, the method comprising:

[0027] S100: During the operation of the CNC machine tool, tool operation feature data is collected in real time, wherein the tool operation feature data includes multiple types of feature data;

[0028] In the embodiment of the present application, tool operation feature data is collected in real time during the operation of a CNC machine tool. Multiple types of feature data are collected to predict tool life based on multimodal data, thereby improving prediction accuracy and avoiding overfitting in training.

[0029] Step S100 of the method provided in the embodiment of the present application includes:

[0030] During the operation of the CNC machine tool, auxiliary light is used to collect material images in real time, wherein the auxiliary light is irradiated onto the material to form a pattern of a preset shape;

[0031] Real-time acquisition of spindle current signals of tool operation;

[0032] Real-time collection of lane operation audio signals;

[0033] The material image, the spindle current signal and the audio signal are aligned to obtain tool operation characteristic data, wherein the acquisition frequency of the spindle current signal and the audio signal is greater than the acquisition frequency of the material image.

[0034] In the embodiment of the present application, during the operation of the CNC machine tool, auxiliary light is used to capture material images in real time, and the auxiliary light is used to improve the quality and consistency of image capture.

[0035] For example, during the machining process of a turning tool (rough and fine turning tool), a light beam is used to illuminate the material to form a pattern on the material. When the turning tool processes the material, the pattern formed also changes due to changes in the material structure. Then, the material image of the material is collected to reflect the wear condition of the turning tool.

[0036] Among them, auxiliary light is used to form a preset pattern on the surface of the material. Compared with the single photoelectric, the formed light pattern has a larger deformation amplitude when the material is deformed by turning tool processing, and the subsequent recognition accuracy is higher.

[0037] For example, the preset pattern can match the shape of the finished material. For example, if the finished material surface is flat, the preset pattern can be a straight line. If the finished surface is conical, the preset pattern can be an inclined straight line. The preset image can be set based on the changing shape of the material, and can also be a curve.

[0038] Exemplarily, the frame rate of capturing material images is 30, that is, 30 material images are captured in 1 second.

[0039] Furthermore, during the operation, the spindle current signal and audio signal of the tool operation are collected in real time, for example, the spindle current size of the tool operation is collected, and the audio signal of the tool during the turning process is collected through a microphone.

[0040] The acquisition frequency of the spindle current signal and the audio signal is greater than the acquisition frequency of the material image. For example, the sampling frequency of the current signal data and the audio signal is 200 Hz, that is, 200 spindle current signals and audio signals are acquired in 1 second.

[0041] Furthermore, the collected material image, spindle current signal and audio signal are aligned to obtain the material image, spindle current signal and audio signal of the tool turning tool at the same moment as the tool operation feature data.

[0042] The step of "aligning the material image, the spindle current signal, and the audio signal to obtain tool operation characteristic data" in the method provided in the embodiment of the present application includes:

[0043] Obtaining an exposure time interval of the material image;

[0044] Selecting the middle moment in the exposure time interval as the alignment timestamp;

[0045] Dividing the spindle current signal and the audio signal according to the acquisition frequency of the material image to obtain a spindle current signal segment sequence and an audio signal segment sequence;

[0046] The spindle current signal segment and the audio signal segment under the alignment timestamp are indexed to obtain the aligned spindle current signal and the aligned audio signal, which are combined with the material image as tool operation feature data.

[0047] In the embodiment of the present application, the three types of data are aligned using the material image as the alignment reference.

[0048] When using auxiliary light to capture material images, exposure and image capture are performed for each image. This exposure interval includes a duration. For example, the exposure interval for the first material image is 0-0.033 seconds, and the exposure interval for the second material image is 0.033-0.066 seconds. A moment in the middle of the exposure interval is selected as the alignment timestamp for data alignment. For example, 0.016 seconds is selected as the alignment timestamp for the first material image.

[0049] During data alignment, the frequencies of the material image, spindle current signal, and audio signal must be consistent. Therefore, the spindle current signal and audio signal data are divided according to the material image acquisition frequency. For example, the spindle current signal and audio signal acquired within one second are divided into 30 segments, resulting in a spindle current signal segment sequence and an audio signal segment sequence. The spindle current signal segment sequence includes 30 spindle current signal segments, and the audio signal segment sequence includes 30 audio signal segments.

[0050] According to the alignment timestamp, the spindle current signal segment and audio signal segment under the alignment timestamp are indexed, that is, the spindle current signal segment and audio signal segment with the alignment timestamp in the corresponding time period are used as the aligned spindle current signal and aligned audio signal of the material image.

[0051] The material image, the alignment spindle current signal and the alignment audio signal are integrated to obtain the tool operation characteristic data after alignment.

[0052] The step of “dividing the spindle current signal and the audio signal according to the acquisition frequency of the material image to obtain a spindle current signal segment sequence and an audio signal segment sequence” in the method provided in the embodiment of the present application includes:

[0053] According to the acquisition frequency of the material image, configure the division time step;

[0054] The spindle current signal and the audio signal are divided according to the division time step to obtain a spindle current signal segment sequence and an audio signal segment sequence, wherein there is an overlapping portion between two adjacent spindle current signal segments and audio signal segments.

[0055] In the embodiment of the present application, when the spindle current signal and the audio signal are divided, there is an overlap between two adjacent spindle current signal segments and audio signal segments. This facilitates time-series-based tool life prediction and improves accuracy.

[0056] Therefore, based on the material image acquisition frequency, when configuring the time step, the time step should be larger than the average time step of 1 second divided by the material image acquisition frequency. For example, the average time step of 1 second divided by the material image acquisition frequency is 1 second divided by 30 = 0.033 seconds, while the actual time step is set to 0.05 seconds. Therefore, when the spindle current signal and audio signal are divided according to the time step, there will be overlap between adjacent spindle current signal segments and audio signal segments.

[0057] Exemplarily, when the spindle current signal and audio signal are divided according to the division time step, 0-0.05 seconds is the first segment of the spindle current signal and audio signal, and 0.033-0.083 is the second segment of the spindle current signal and audio signal. In this way, the spindle current signal segment sequence and audio signal segment sequence of the acquisition frequency of the material image are divided. For example, the spindle current signal segment sequence and the audio signal segment sequence include 30 spindle current signal segments and audio signal segments, and there are overlapping parts in two adjacent spindle current signal segments and audio signal segments.

[0058] S200: inputting the tool operation characteristic data into a tool life prediction model, and outputting a first tool life prediction result;

[0059] In an embodiment of the present application, tool operation characteristic data is obtained during collection and processing, and is input into a trained tool life prediction model to perform a preliminary tool life prediction, and a first tool life prediction result is obtained as an output.

[0060] The steps for constructing the tool life prediction model include:

[0061] Based on the tool operation data in the historical time, a sample tool operation feature data set is processed and obtained. Each set of sample tool operation feature data is annotated to obtain the sample first tool life prediction result set as the supervised training data set. Among them, each sample first tool life prediction result includes the annotated information of the early stage of tool wear, the middle stage of tool wear, and the late stage of tool wear.

[0062] Constructing a material image feature extraction path, a current signal feature extraction path, an audio signal feature extraction path, a fusion module, and a classification module within the tool life prediction model;

[0063] The supervised training data set is used to perform supervised training, verification and testing on the tool life prediction model until convergence, thereby obtaining a tool life prediction model.

[0064] like Figure 2 As shown, in this embodiment of the present application, based on historical tool operation data, multimodal data related to tool life is first collected and processed to form a sample tool operation feature dataset. Specifically, this sample tool operation feature dataset includes three key data types: image data generated during material processing, spindle motor current signal data, and audio signal data during processing. Alignment processing is performed through the steps described above to obtain the sample tool operation feature dataset.

[0065] After obtaining a dataset of sample tool operation characteristics, each set of sample tool operation characteristic data is annotated using the tool life annotation standard. The annotations cover the three stages of tool wear: early tool wear (tool performance is essentially intact, exhibiting only minor surface wear, and tool sharpness is essentially maintained); mid-stage tool wear (tool performance declines somewhat, surface wear is more pronounced, and machining quality begins to fluctuate); and late stage tool wear (tool performance declines significantly, wear is significant, and there is a risk of tool breakage). Annotation can be performed manually or by measuring tool wear. Ultimately, the first set of tool life prediction results for the sample is obtained, which serves as the supervised training dataset for the model.

[0066] like Figure 2 As shown in FIG, the network structure of the tool life prediction model is constructed. The network structure of the tool life prediction model includes a material image feature extraction path, a current signal feature extraction path, an audio signal feature extraction path, a fusion module and a classification module.

[0067] The material image feature extraction path is built based on a 2D convolutional neural network, extracting features from the material image and flattening them into one-dimensional vectors. The current signal feature extraction path and the audio signal feature extraction path are built based on the first and second GRU neural networks, extracting features from the current and audio signals and outputting them as one-dimensional vectors. The fusion module is built based on a DNN and specifically includes a fully connected layer and a dropout layer. The classification module is built based on the third GRU neural network and a softmax activation function. The output categories of the tool life prediction model include the early stage of tool wear, the middle stage of tool wear, and the late stage of tool wear. The GRU output at each moment includes historical information. To better integrate this historical information, a splicing layer is also included. This layer is placed after the material image feature extraction path, the current signal feature extraction path, and the audio signal feature extraction path, respectively. The output of a full second is spliced ​​and then fed into the fusion module for fusion. The amount of data after fusion is large, so the dropout parameter is set to 0.3.

[0068] The tool life prediction model is trained using a supervised training dataset. During training, the model parameters are adjusted through backpropagation and iterative optimization based on the output of the tool life prediction model, gradually improving the model's accuracy in identifying tool life stages. Model training also includes validation and testing steps to evaluate the model's generalization and predictive performance. The training loss function uses cross-entropy loss, and the optimizer uses the Adam algorithm to update the model until the model converges on the validation set, achieving, for example, 90% accuracy. This results in the final tool life prediction model.

[0069] Based on the trained tool life prediction model, the currently collected tool operation feature data is input into the tool life prediction model, and the first tool life prediction result is output, which includes one of the early stage of tool wear, the middle stage of tool wear or the late stage of tool wear.

[0070] S300: Obtaining a finished product image after material processing, combining it with a preset target finished product image, identifying and obtaining a processing similarity, combining it with the first tool life prediction result, and classifying and obtaining a tool life prediction result.

[0071] Furthermore, in order to improve the accuracy of tool life prediction analysis, tool life analysis is also performed based on the similarity between the material after processing and the standard.

[0072] The image of the finished product after material processing is obtained and combined with the preset target product image of the expected processing (i.e., the image of the finished product after standard processing) to identify the similarity between the two. If the tool life is good and wear is minimal, the similarity between the finished product image and the target product image should be high. Conversely, if the tool is severely worn, the similarity should be low. After identifying the similarity between the two, the processing similarity is obtained. Combined with the first tool life prediction result, the final tool life prediction result is classified to obtain the tool life prediction result.

[0073] By combining the similarity between the finished product image after processing and the target finished product image, tool life prediction classification can be performed to improve the accuracy and reliability of tool life prediction analysis.

[0074] Step S300 in the method provided in the embodiment of the present application includes:

[0075] After the CNC machine tool completes processing of the material, the image of the processed product is acquired, and the target finished product image preset by the material processing is obtained;

[0076] Based on the twin network, a finished product similarity recognition network is constructed;

[0077] The processed finished product image and the target finished product image are input into the finished product similarity recognition network, image features are extracted and Euclidean distance is calculated to obtain processing similarity.

[0078] In this embodiment of the present application, after a CNC machine tool completes material processing, the system captures an image of the finished product and obtains a pre-processed image of the target finished product to further assess processing quality and tool condition. These two images, respectively, reflect the actual and ideal shapes of the finished product, providing a basis for quantitative evaluation of processing quality.

[0079] Based on the Siamese Network, a finished product similarity recognition network is constructed, which includes two convolutional neural network paths and can extract features from the input finished product images.

[0080] The currently acquired processed product image and the target finished product image are input into the finished product similarity recognition network, the image features are extracted, and the Euclidean distance between the two image features is calculated to obtain the processing similarity between the processed product image and the target finished product image.

[0081] The relationship between the Euclidean distance threshold and the processing similarity can be pre-set. The processing similarity can be, for example, classified as low or high. If the Euclidean distance is less than the threshold, the similarity is low; if the Euclidean distance is greater than the threshold, the similarity is high. In this way, the processing similarity between the processed product image and the target product image is identified. The processing similarity can be classified as either low or high.

[0082] Furthermore, step S300 in the method provided in the embodiment of the present application further includes:

[0083] Constructing a tool life classifier;

[0084] The first tool life prediction result and the processing similarity are input into the tool life classifier to obtain a tool life prediction result.

[0085] In an embodiment of the present application, a tool life classifier is constructed to perform final tool life prediction classification based on the first tool life prediction result and the processing similarity.

[0086] Among them, the tool life classifier is constructed, including:

[0087] Obtaining a sample first tool life prediction result set at the early stage of tool wear, the middle stage of tool wear, and the late stage of tool wear, and collecting a sample processing similarity set;

[0088] Traversing the combination of the sample first tool life prediction result set and the sample processing similarity set, and marking the sample tool life prediction result for each combination to obtain the sample tool life prediction result set;

[0089] A tool life classifier is constructed by using the sample first tool life prediction result set, the sample processing similarity set and the sample tool life prediction result set.

[0090] In the embodiment of the present application, a set of sample first tool life prediction results for the early, mid, and late stages of tool wear is first obtained. Specifically, the sample first tool life prediction results include one of the early, mid, and late stages of tool wear. The sample first tool life prediction result set serves as one of the classification data for classifying and obtaining the tool life prediction results.

[0091] A sample processing similarity set is further obtained. Exemplarily, the sample processing similarities include two types, namely, low similarity and high similarity.

[0092] Each sample first tool life prediction result and each sample processing similarity in the sample first tool life prediction result set and the sample processing similarity set are traversed and combined to obtain multiple combinations, such as six combinations of early tool wear-low similarity, mid-term tool wear-low similarity, late tool wear-low similarity, early tool wear-high similarity, mid-term tool wear-high similarity and late tool wear-high similarity.

[0093] Label the sample tool life prediction results for each combination. Sample tool life prediction results include early, mid-term, and late stages. For example, if the early stage of tool wear has a low similarity, the sample tool life prediction result corresponding to the late stage is the late stage; if the mid-term of tool wear has a low similarity, the sample tool life prediction result corresponding to the late stage is the late stage; if the late stage of tool wear has a low similarity, the sample tool life prediction result corresponding to the early stage of tool wear has a high similarity, the sample tool life prediction result corresponding to the mid-term of tool wear has a high similarity, the sample tool life prediction result corresponding to the mid-term of tool wear has a high similarity, and the sample tool life prediction result corresponding to the late stage of tool wear has a high similarity is the mid-term. In this way, a set of sample tool life prediction results is obtained.

[0094] A tool life classifier is constructed using the sample first tool life prediction result set, the sample processing similarity set and the sample tool life prediction result set. Specifically, a mapping relationship between the combination of the sample first tool life prediction result set, the sample processing similarity set and the sample tool life prediction result set is constructed to obtain the tool life classifier.

[0095] After obtaining the first tool life prediction result and identifying the machining similarity, the two are input into the tool life classifier to obtain the tool life prediction result corresponding to the combination of the two, completing the tool life prediction and classification. The tool life prediction result is specifically one of the early, mid, and late stages. If the tool life prediction result is late, the tool can be replaced.

[0096] In summary, the embodiments of the present application have at least the following technical effects:

[0097] The embodiment of the present application proposes a method for predicting tool life for CNC machine tools. The method collects tool operation feature data from multimodal data during the operation of the CNC machine tool, then trains a tool life prediction model based on the multimodal data, performs input prediction on the current tool operation feature data, obtains a first tool life prediction result, then obtains an image of the finished product after material processing, combines it with a preset target finished product image, identifies the processing similarity, and classifies it to obtain a tool life prediction result based on the first tool life prediction result. The present application uses a lightweight model to identify the quality of material processing, then combines it with the tool life predicted using multimodal data, and finally classifies and predicts to obtain a tool life prediction result. With a relatively small amount of data, it can identify the tool life of a turning tool, thereby achieving the technical effect of improving the accuracy of tool life prediction and improving the processing quality of CNC machine tools.

[0098] Example 2, as Figure 3 As shown, based on the same inventive concept as the method for predicting the tool life of a CNC machine tool provided in Example 1, an embodiment of the present invention further provides a device for predicting the tool life of a CNC machine tool, comprising:

[0099] The tool data acquisition module 11 is used to collect tool operation feature data in real time during the operation of the CNC machine tool, wherein the tool operation feature data includes multiple types of feature data;

[0100] A first life prediction module 12 is used to input the tool operation characteristic data into a tool life prediction model and output a first tool life prediction result;

[0101] The life prediction classification module 13 is used to obtain the image of the finished product after material processing, identify and obtain the processing similarity in combination with the preset target finished product image, and classify and obtain the tool life prediction result in combination with the first tool life prediction result.

[0102] In one embodiment, the tool data collection module 11 is further configured to:

[0103] During the operation of the CNC machine tool, auxiliary light is used to collect material images in real time, wherein the auxiliary light is irradiated onto the material to form a pattern of a preset shape;

[0104] Real-time acquisition of spindle current signals of tool operation;

[0105] Real-time collection of lane operation audio signals;

[0106] The material image, the spindle current signal and the audio signal are aligned to obtain tool operation characteristic data, wherein the acquisition frequency of the spindle current signal and the audio signal is greater than the acquisition frequency of the material image.

[0107] The material image, spindle current signal and audio signal are aligned to obtain tool operation feature data, including:

[0108] Obtaining an exposure time interval of the material image;

[0109] Selecting the middle moment in the exposure time interval as the alignment timestamp;

[0110] Dividing the spindle current signal and the audio signal according to the acquisition frequency of the material image to obtain a spindle current signal segment sequence and an audio signal segment sequence;

[0111] The spindle current signal segment and the audio signal segment under the alignment timestamp are indexed to obtain the aligned spindle current signal and the aligned audio signal, which are combined with the material image as tool operation feature data.

[0112] The spindle current signal and the audio signal are divided according to the acquisition frequency of the material image to obtain a spindle current signal segment sequence and an audio signal segment sequence, including:

[0113] According to the acquisition frequency of the material image, configure the division time step;

[0114] The spindle current signal and the audio signal are divided according to the division time step to obtain a spindle current signal segment sequence and an audio signal segment sequence, wherein there is an overlapping portion between two adjacent spindle current signal segments and audio signal segments.

[0115] In one embodiment, the first life prediction module 12 is further configured to:

[0116] Based on the tool operation data in the historical time, a sample tool operation feature data set is processed and obtained. Each set of sample tool operation feature data is annotated to obtain the sample first tool life prediction result set as the supervised training data set. Among them, each sample first tool life prediction result includes the annotated information of the early stage of tool wear, the middle stage of tool wear, and the late stage of tool wear.

[0117] Constructing a material image feature extraction path, a current signal feature extraction path, an audio signal feature extraction path, a fusion module, and a classification module within the tool life prediction model;

[0118] The supervised training data set is used to perform supervised training, verification and testing on the tool life prediction model until convergence, thereby obtaining a tool life prediction model.

[0119] In one embodiment, the lifespan prediction classification module 13 is further configured to:

[0120] After the CNC machine tool completes processing of the material, the image of the processed product is acquired, and the target finished product image preset by the material processing is obtained;

[0121] Based on the twin network, a finished product similarity recognition network is constructed;

[0122] The processed finished product image and the target finished product image are input into the finished product similarity recognition network, image features are extracted and Euclidean distance is calculated to obtain processing similarity.

[0123] In one embodiment, the lifespan prediction classification module 13 is further configured to:

[0124] Constructing a tool life classifier;

[0125] The first tool life prediction result and the processing similarity are input into the tool life classifier to obtain a tool life prediction result.

[0126] Among them, building a tool life classifier includes:

[0127] Obtaining a sample first tool life prediction result set at the early stage of tool wear, the middle stage of tool wear, and the late stage of tool wear, and collecting a sample processing similarity set;

[0128] Traversing the combination of the sample first tool life prediction result set and the sample processing similarity set, and marking the sample tool life prediction result for each combination to obtain the sample tool life prediction result set;

[0129] A tool life classifier is constructed by using the sample first tool life prediction result set, the sample processing similarity set and the sample tool life prediction result set.

[0130] Example three, based on the same inventive concept of a CNC machine tool tool life prediction device provided in Example two, this embodiment of the application also provides a CNC machine tool, which includes a CNC machine tool tool life prediction device provided in Example two, and also includes any structure of a CNC machine tool in the prior art.

[0131] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0132] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0133] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A method for predicting the life of a CNC machine tool turning tool, characterized in that: The method comprises: During the operation of the CNC machine tool, the turning tool operation characteristic data is collected, wherein the turning tool operation characteristic data includes multiple types of characteristic data, and the multiple types of characteristic data include material images, spindle current signals, and audio signals; Inputting the turning tool operation characteristic data into a turning tool life prediction model, and outputting a first turning tool life prediction result; Obtaining a finished product image after material processing, combining it with a preset target finished product image, identifying and obtaining processing similarity, combining it with the first turning tool life prediction result, and classifying and obtaining a turning tool life prediction result, including: Construct a turning tool life classifier; Inputting the first turning tool life prediction result and the processing similarity into the turning tool life classifier to obtain a turning tool life prediction result; Among them, building a turning tool life classifier includes: Obtain the first turning tool life prediction result set of samples in the early stage of turning tool wear, the middle stage of turning tool wear, and the late stage of turning tool wear, and collect the sample processing similarity set; Traversing the combination of the sample first turning tool life prediction result set and the sample processing similarity set, and marking the sample turning tool life prediction result for each combination to obtain the sample turning tool life prediction result set; A turning tool life classifier is constructed by using the sample first turning tool life prediction result set, the sample processing similarity set and the sample turning tool life prediction result set.

2. The method for predicting the life of a CNC machine tool turning tool according to claim 1, wherein: During the operation of the CNC machine tool, the tool operation characteristic data is collected, including: During the operation of CNC machine tools, auxiliary light is used to collect material images; Collect the spindle current signal of the turning tool; Collect the audio signal of the turning tool operation; The material image, the spindle current signal and the audio signal are aligned to obtain turning tool operation characteristic data, wherein the acquisition frequency of the spindle current signal and the audio signal is greater than the acquisition frequency of the material image.

3. The method for predicting the life of a CNC machine tool turning tool according to claim 2, wherein: The material image, spindle current signal and audio signal are aligned to obtain turning tool operation characteristic data, including: Obtaining an exposure time interval of the material image; Selecting the middle moment in the exposure time interval as the alignment timestamp; Dividing the spindle current signal and the audio signal according to the acquisition frequency of the material image to obtain a spindle current signal segment sequence and an audio signal segment sequence; The spindle current signal segment and the audio signal segment under the index alignment timestamp are obtained to obtain the aligned spindle current signal and the aligned audio signal, which are combined with the material image as turning tool operation feature data.

4. The method for predicting the life of a CNC machine tool turning tool according to claim 3, wherein: According to the acquisition frequency of the material image, the spindle current signal and the audio signal are divided to obtain a spindle current signal segment sequence and an audio signal segment sequence, including: According to the acquisition frequency of the material image, configure the division time step; The spindle current signal and the audio signal are divided according to the division time step to obtain a spindle current signal segment sequence and an audio signal segment sequence, wherein there is an overlapping portion between two adjacent spindle current signal segments and audio signal segments.

5. The method for predicting the life of a CNC machine tool turning tool according to claim 4, wherein: The steps to construct the turning tool life prediction model include: Based on the turning tool operation data in the historical time period, a sample turning tool operation feature dataset is processed and each set of sample turning tool operation feature data is annotated to obtain the sample first turning tool life prediction result set as the supervised training dataset. Among them, the prediction result of each sample first turning tool life includes the annotated information of the early, middle and late stages of turning tool wear. Constructing a material image feature extraction path, a current signal feature extraction path, an audio signal feature extraction path, a fusion module, and a classification module within the turning tool life prediction model; The supervised training data set is used to perform supervised training, verification and testing on the turning tool life prediction model until convergence, thereby obtaining the turning tool life prediction model.

6. The method for predicting the life of a CNC machine tool turning tool according to claim 1, wherein: Obtain the finished product image after material processing, combine it with the preset target finished product image, and identify the processing similarity, including: After the CNC machine tool completes processing of the material, the image of the processed product is acquired, and the target finished product image preset by the material processing is obtained; Based on the twin network, a finished product similarity recognition network is constructed; The processed finished product image and the target finished product image are input into the finished product similarity recognition network, image features are extracted and Euclidean distance is calculated to obtain processing similarity.

7. A device for predicting the life of a CNC machine tool turning tool, characterized in that: For implementing the method for predicting the life of a CNC machine tool turning tool according to any one of claims 1 to 6, the device comprises: A turning tool data acquisition module is used to collect turning tool operation characteristic data during the operation of the CNC machine tool, wherein the turning tool operation characteristic data includes multiple types of characteristic data; A first life prediction module, configured to input the turning tool operation characteristic data into a turning tool life prediction model and output a first turning tool life prediction result; The life prediction classification module is used to obtain the image of the finished product after material processing, combine it with the preset target finished product image, identify and obtain the processing similarity, and combine it with the first turning tool life prediction result to classify and obtain the turning tool life prediction result.

8. A CNC machine tool, characterized in that: It includes the CNC machine tool turning tool life prediction device as described in claim 7.

Images