A method and device for optimizing and recommending cutting parameters of a tool

Abstract

The present application provides a method and device for optimizing and recommending tool cutting parameters. The method provided in the present application includes: collecting basic correlation data of tool cutting parameters, and constructing a basic data information library based on the basic correlation data; collecting real-time tool cutting information; constructing a tool cutting parameter recommendation model based on the real-time completed tool cutting information and the basic data information library, and the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut; performing cutting operations according to the optimal cutting parameters; predicting the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, monitoring the real-time cutting force of the cutting operation, and adjusting the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force. The method and device for optimizing and recommending tool cutting parameters provided in the present application are used to accurately obtain and update tool cutting parameters.

Description

Technical Field

[0001] The present application relates to the field of tool cutting technology, and in particular to a method and device for optimizing and recommending tool cutting parameters. Background Art

[0002] In the field of mechanical processing, the reasonable setting of tool cutting parameters is the key factor to ensure processing efficiency, processing quality and tool life. With the development of modern manufacturing industry, the requirements for processing accuracy and efficiency are getting higher and higher, and the optimization of cutting parameters has become particularly important. Cutting parameters directly affect the effect of the cutting process, and then affect the efficiency, cost and safety of the entire production process.

[0003] The existing tool cutting parameter settings mainly rely on the operator's experience, intuition and manuals, lacking scientific data support and accurate calculations. Once the parameters are set, they are often adjusted only after problems occur. Improper parameter settings may lead to problems such as excessive tool wear, poor workpiece surface quality or machine tool overload. In severe cases, the surface roughness of the workpiece will increase, affecting the final quality and appearance of the workpiece, and the dimensional accuracy will not meet the standards, which will also cause the tool to chip or break. At the same time, the chipping or breakage of the tool may also cause safety accidents, causing harm to the operator and leading to unexpected situations. After the tool chipping or breakage, the tool will be replaced, and frequent tool replacement will also affect processing efficiency. In addition, improper cutting parameter settings will increase the load on the machine tool, affecting the accuracy and stability of the machine tool, while accelerating the wear and aging of the machine tool.

[0004] In recent years, when setting cutting parameters through cutting parameter software, although the error problem of manual setting can be avoided to a certain extent, these team building is more concerned with how to complete the cutting task, and relatively less about tool wear and machine tool performance. While pursuing processing efficiency, it is impossible to guarantee processing quality and cutting safety. Summary of the invention

[0005] In view of this, the present application provides a method and device for optimizing and recommending tool cutting parameters, which are used to accurately acquire and update tool cutting parameters.

[0006] Specifically, the present application is implemented through the following technical solutions:

[0007] The first aspect of the present application provides a method for optimizing and recommending cutting parameters of a tool, the method comprising:

[0008] Collecting basic correlation data of tool cutting parameters, and building a basic data information base based on the basic correlation data, wherein the basic data information base is used to characterize the combination relationship between each basic correlation data, wherein the basic correlation data is the influencing factors of multiple dimensions of tool cutting operations, and the basic data information base includes a first correlation relationship between the influencing factors of each dimension and a second correlation relationship between each influencing factor of each dimension in the multiple dimensions;

[0009] Collecting real-time tool cutting information, wherein the real-time tool cutting information at least includes real-time work status information to be cut and real-time completed tool cutting information;

[0010] Building a tool cutting parameter recommendation model based on the tool cutting information completed in real time and the basic data information library, wherein the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut;

[0011] Perform cutting operations according to the optimal cutting parameters;

[0012] The target cutting force is predicted based on the tool cutting force pre-trained large model and the optimal cutting parameters, and the real-time cutting force of the cutting operation is monitored. The tool cutting force pre-trained large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force.

[0013] A second aspect of the present application provides a device for optimizing and recommending cutting parameters of a tool, the device comprising a construction module, a collection module, a processing module and an adjustment module; wherein;

[0014] The construction module is used to collect basic correlation data of tool cutting parameters, and to construct a basic data information base based on the basic correlation data, wherein the basic data information base is used to characterize the combination relationship between each basic correlation data, wherein the basic correlation data is the influencing factors of multiple dimensions of tool cutting operations, and the basic data information base includes a first correlation relationship between the influencing factors of each dimension and a second correlation relationship between each influencing factor of each dimension in the multiple dimensions;

[0015] The acquisition module is used to acquire real-time tool cutting information, wherein the real-time tool cutting information at least includes real-time working state information to be cut and real-time completed tool cutting information;

[0016] The construction module is further used to construct a tool cutting parameter recommendation model based on the tool cutting information completed in real time and the basic data information library, and the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut;

[0017] The processing module is used to perform cutting operations according to the optimal cutting parameters;

[0018] The adjustment module is used to predict the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, and monitor the real-time cutting force of the cutting operation. The tool cutting force pre-trained large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force.

[0019] The optimization and recommendation method and device of tool cutting parameters provided by the present application first collect basic correlation data of tool cutting parameters, build a basic data information base based on the basic correlation data, the basic data information base is used to characterize the combination relationship between each basic correlation data, the basic correlation data is the influencing factors of multiple dimensions of tool cutting operation, the basic data information base includes the first correlation relationship between the influencing factors of each dimension and the second correlation relationship between each influencing factor of each dimension in multiple dimensions, and then collect real-time tool cutting information, the real-time tool cutting information at least includes the real-time working state information to be cut and the real-time completed tool cutting information, and then build a tool cutting parameter recommendation model based on the real-time completed tool cutting information and the basic data information base, the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut, and then performs cutting operation according to the optimal cutting parameters, and finally predicts the target cutting force based on the tool cutting force pre-training large model and the optimal cutting parameters, monitors the real-time cutting force of the cutting operation, and the tool cutting force pre-training large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force. First, through the establishment of the first correlation relationship and the second correlation relationship, the correlation relationship of the influencing factors of each cutting parameter is directly fixed, and there is no need for the model to perform repeated learning and prediction, which improves the efficiency and accuracy of the later model prediction. Second, through the basic data information base of multiple dimensions, the cutting parameter recommendation model and the tool cutting force pre-training large model are used to realize the cutting parameter recommendation. Compared with the traditional setting method that relies on experience and intuition, it is more scientific and accurate. Specifically, the first model is used to predict the optimal cutting parameters. However, in the actual process, the issuance of cutting parameter instructions cannot complete accurate control. At this time, the inaccurate execution of control instructions can be adjusted again through the monitoring of actual cutting force, ensuring that the expected effect of the optimal cutting parameters can be accurately implemented, thereby improving the accuracy of cutting control. Furthermore, the basic data information base covers detailed data of multiple dimensions, ensuring the comprehensiveness and accuracy of the data. At the same time, the basic data information base provides solid data support for the cutting parameter recommendation model and the cutting force pre-training large model. By deeply analyzing the combination relationship in the basic data information base, a scientific basis is provided for optimizing the cutting process. Furthermore, the tool cutting parameter recommendation model analyzes the real-time tool cutting information of multiple dimensions, and can accurately predict and recommend the optimal cutting parameters. Furthermore, the cutting force pre-training model can be used to monitor the cutting force in real time, obtain real-time cutting force data, and dynamically adjust the optimal cutting parameters to optimize the cutting process and ensure the stability and safety of the cutting force. In this way, through the basic data information library, the tool cutting parameter recommendation model and the cutting force pre-training model, the intelligent recommendation of cutting parameters and the intelligent control of the cutting process can be achieved, ensuring the accuracy and reliability of the cutting parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of Example 1 of the optimization recommendation method for tool cutting parameters provided in this application;

[0021] Figure 2 A partial schematic diagram of the first embodiment of the basic data information database provided by this application;

[0022] Figure 3 A hardware structure diagram of a device for optimizing and recommending cutting parameters of a tool in the present application;

[0023] Figure 4 This is a structural schematic diagram of Example 1 of the device for optimizing and recommending tool cutting parameters provided in this application. DETAILED DESCRIPTION

[0024] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0025] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0026] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0027] The present application provides a method and device for optimizing and recommending tool cutting parameters, which are used to accurately acquire and update tool cutting parameters.

[0028] Specific embodiments are given below to introduce the technical solution of the present application in detail.

[0029] Figure 1 This is a flow chart of Example 1 of the optimization recommendation method for tool cutting parameters provided in this application. Please refer to Figure 1 , the method provided in this embodiment may include:

[0030] S101. Collect basic correlation data of tool cutting parameters, and build a basic data information base based on the basic correlation data, wherein the basic data information base is used to characterize the combination relationship between each basic correlation data, and the basic correlation data are influencing factors of multiple dimensions of tool cutting operations. The basic data information base includes a first correlation relationship between the influencing factors of each dimension and a second correlation relationship between each influencing factor of each dimension in the multiple dimensions.

[0031] Specifically, the basic associated data is a series of basic data that affects the tool cutting process. Any data of an object related to the cutting process can be used as the basic associated data. In specific implementation, the basic associated data of tool cutting parameters stored historically can be obtained.

[0032] It should be noted that the specific dimensions of the basic associated data of tool cutting parameters are determined according to actual needs, and in this embodiment, there is no limitation on this. In specific implementation, the dimensions of the basic associated data of tool cutting parameters may include machine tool dimensions, tool holder dimensions, tool dimensions, tool cutting parameter dimensions, workpiece material dimensions, processing technology dimensions, etc. Dimensions refer to objects of basic associated data, such as processed tools, processing machines, etc. Correspondingly, the basic data information library may include a machine tool basic data information library, a tool holder basic data information library, a tool basic data information library, a tool cutting parameter basic data information library, a workpiece material basic data information library, and a processing technology data basic data information library, etc.

[0033] Furthermore, a basic data information database of machine tools is constructed, and the data attributes include manufacturer, model, maximum speed, spindle interface type, spindle diameter, spindle speed, spindle positioning accuracy, motor power, control system brand and model, number of feed axes, control method, worktable size, travel range, maximum processing length, repeat positioning accuracy, straightness, parallelism, maintenance cycle, last maintenance time, cooling system, lubrication system, additional axes, automatic tool changing system, tailstock, and usage history.

[0034] Furthermore, a basic data information database of the tool holder is constructed, and the data attributes include model, manufacturer, production year, serial number, serial number of the tool holder, tool holder type, tool holder diameter, tool holder length, tool holder width, clamping length, taper, material, hardness, density, interface type, maximum torque, clamping method, maximum load weight, maximum speed, balance level, sensor type, power requirements, coolant interface, lubrication method, tool holder surface treatment process, heat treatment process of the tool holder, last maintenance time, maintenance cycle, and usage history.

[0035] Furthermore, a basic tool data information database is constructed, and the data attributes include model, manufacturer, serial number, diameter, length, effective length, blade length, shank diameter, taper, tool material, hardness, coating, number of blades, helix angle, rake angle, back angle, blade inclination angle, blade shape, life, wear status, number of uses, maintenance cycle, connection type, connection accuracy, clamping force, tool surface treatment process, heat treatment process undergone by the tool, production batch, production date, warranty period, and usage history.

[0036] Furthermore, a basic data information database of tool cutting parameters is constructed, and the data attributes include cutting speed, feed rate, cutting depth, and cutting width.

[0037] Furthermore, a basic data information database of processed workpiece materials is constructed, and the data attributes include workpiece material name, workpiece material length, workpiece material width, workpiece material height, workpiece material material, workpiece material hardness, workpiece material strength, workpiece material toughness, workpiece material thermal conductivity, workpiece material thermal expansion coefficient, workpiece material hole diameter, workpiece material weight, workpiece material surface roughness, and workpiece material usage history.

[0038] Furthermore, a basic data information database of machining process data is constructed, and the data attributes include turning, milling, drilling, boring, grinding, tapping, and rolling.

[0039] It should be noted that the basic associated data of tool cutting parameters can be obtained through multi-dimensional usage history. In specific implementation, the basic associated data of tool cutting parameters can be obtained through the machine tool usage history, the tool handle usage history, the tool material, the tool usage history, the workpiece material material, and the workpiece material usage history. Further, the machine tool usage history includes the assembled tool handle, the tool, and the processed workpiece materials and quantity; the tool handle usage history includes the assembled machine tool, the tool, and the processed workpiece materials and quantity; the tool material includes high-speed steel, cemented carbide, ceramics, diamond, cubic boron nitride, polycrystalline diamond, polycrystalline cubic boron nitride; the tool usage history includes the assembled tool handle and the processed workpiece materials and quantity; the workpiece material material includes carbon steel, alloy steel, stainless steel, aluminum, titanium, gold, silver, aluminum alloy, titanium alloy, nickel alloy, magnesium alloy, polyethylene, polypropylene, polyvinyl chloride, polycarbonate, nylon, alumina ceramic, silicon nitride ceramic, composite material; the workpiece material usage history further includes the machine tool, tool handle, tool, tool cutting and processing technology required for processing.

[0040] Furthermore, the data in the basic data information base have a combination relationship. In specific implementation, the combination relationship can be determined according to the actual matching relationship and influence relationship of the data during tool cutting.

[0041] It should be noted that all data included in the basic associated data are influencing factors in the cutting operation.

[0042] Furthermore, in each dimension of the basic associated data, such as the tool dimension, the relationship between the various data is a first associated relationship, such as the relationship between the geometric parameters of the tool and the drill bit. Figure 2 This is a partial schematic diagram of the first embodiment of the basic data information database provided by this application. Figure 2 ,The figure only shows the first correlation relationship between the ,influencing factors in the dimension of tool geometric parameters, and two influencing factors with ,influence factors are connected by a short line, representing the first correlation relationship between the ,corresponding influencing factors.

[0043] Furthermore, between different dimensions, the relationship between each influencing factor is a second association relationship. For example, in one embodiment, the relationship between the influencing factor in the tool geometry dimension: the number of teeth, and the influencing factor in the workpiece material dimension: non-metallic material, is a second association relationship.

[0044] Furthermore, the basic correlation data of tool cutting parameters can be derived from standards, patents, process / assembly manuals, cutting manuals, manufacturer standard technical manuals, user manuals, manufacturer technical support, evaluation reports, industry analysis reports, development trend reports, technical analysis reports, academic journals, papers, magazines and books, online network platforms, daily practical experience, and daily maintenance and repair report files related to organic processing products.

[0045] S102, collecting real-time tool cutting information, wherein the real-time tool cutting information at least includes real-time work status information to be cut and real-time completed tool cutting information.

[0046] Specifically, the real-time tool cutting information refers to information directly related to the tool cutting state that is acquired and recorded in real time during the cutting process. It should be noted that the real-time tool cutting information can reflect the actual situation of the current cutting operation.

[0047] Furthermore, the data type of the real-time tool cutting information may include structured text, semi-structured text, unstructured text, tables, documents, pictures, audio and video related to the machining scene. Furthermore, the data source of the real-time tool cutting information may include the technical data of the machine tool, tool, workpiece, material and tool manufacturer. Furthermore, the data collection method of the real-time tool cutting information may include data collection, system docking, manual import and / or input, etc.

[0048] Furthermore, after the real-time tool cutting information is collected, data processing can be performed on it. In specific implementation, duplicate records in the real-time tool cutting information can be deleted to ensure the uniqueness of the real-time tool cutting information.

[0049] Furthermore, for real-time tool cutting information containing missing values, if there are fewer missing values, the records containing missing values ​​can be directly deleted. For data with more missing values ​​that are crucial to the analysis, mean interpolation, modeling prediction, multiple interpolation and other methods can be used to fill the missing values.

[0050] Furthermore, statistical methods (Z-score, IQR) can be used to identify and process outliers in the data, and delete or replace them with reasonable values. Non-numeric data (text, categorical data) in real-time tool cutting information can also be converted into numerical data.

[0051] Furthermore, after obtaining the real-time tool cutting information, key features related to tool cutting parameters and their influencing factors can be extracted from the real-time tool cutting information: material hardness, tool material and shape, machine tool power, etc.

[0052] Furthermore, the real-time working status information to be cut refers to the state in which the cutting operation has not been performed, and the specific working status information of each part. At this time, the cutting parameters under the working status information are unknown and need to be predicted by the model provided by the present invention and adjusted based on the pre-trained large model.

[0053] Furthermore, the tool cutting information completed in real time refers to the information fed back by the tool after the current cutting operation has been completed in the adjacent time. At this time, the cutting parameters are known, and the basic data information library is corrected with such parameters to improve the accuracy of the database and the association relationship, and at the same time improve the accuracy of the model predicting the cutting force. In specific implementation, the tool cutting information completed in real time can include the actual cutting parameters of the tool during the cutting process, the cutting effect, the tool wear, and the abnormal conditions that may occur during the cutting process.

[0054] When this step is implemented, sensors can be installed on the machine tool and the tool according to the cutting requirements and machine tool characteristics to accurately capture the real-time tool cutting information.

[0055] Optionally, the tool cutting information completed in real time includes data of each object in each link of the cutting operation process, and the data lengths and data expression forms of different objects in different links are different;

[0056] After collecting the real-time tool cutting information, the method further includes:

[0057] (1) Extracting target features associated with cutting parameters from the tool cutting information completed in real time based on correlation analysis.

[0058] Specifically, the data of each object in each link of the cutting operation process may be data extracted from different stages of processing, different use states of the tool, different cutting states of the material, and the like.

[0059] It is understandable that, due to the different durations and work contents of different stages of the cutting operation, the data lengths and data expression forms of different objects in different links are different.

[0060] It should be noted that the target features can be extracted based on the traditional correlation analysis method or the neural network correlation analysis method, which is not limited in this step. In the specific implementation, the data related to the cutting parameters are extracted, the features of these data are obtained, and they are determined as the target features.

[0061] In specific implementation, correlation analysis and principal component analysis (PCA) methods can be used to select features that have a significant impact on the tool cutting information completed in real time. The real-time tool cutting information is converted from the original feature space to a new low-dimensional space through the principal component analysis method, retaining the main change direction in the real-time tool cutting information while reducing the complexity of the data.

[0062] (2) Each target feature is segmented to obtain a data segment set of all target features, wherein the lengths of the data segments in the data segment set are the same.

[0063] Specifically, the specific length of each data segment is determined according to actual needs, and is not limited in this embodiment. In specific implementation, the target feature is divided according to the determined length to obtain multiple data segments.

[0064] Furthermore, all data segments are combined to obtain a data segment set.

[0065] (3) Marking each data segment according to the processing stage to which the target feature belongs and the corresponding tool status.

[0066] Specifically, the processing stage is determined according to each link in the cutting operation process.

[0067] Furthermore, according to the acquisition time corresponding to the data segment, the processing stage and the corresponding tool state corresponding to each data point are determined, the processing stages and tool states belonging to the same category are grouped, and the data segments with classification numbers are marked with the processing stages and tool states.

[0068] The optimization and recommendation method for tool cutting parameters provided in this embodiment first extracts the target features associated with the cutting parameters from the tool cutting information completed in real time based on correlation analysis, and then segments each target feature to obtain a data segment set of all target features, in which each data segment has the same length, and finally marks each data segment according to the processing stage to which the target feature belongs and the corresponding tool state. In this way, by segmenting the tool cutting information completed in real time according to the processing stage and the corresponding tool state, the cutting characteristics and parameter changes in each stage and state can be analyzed more accurately. In this way, it is helpful to capture more detailed data features and change trends, thereby improving the accuracy and effectiveness of data analysis. Further, by adding corresponding labels to each segment of data to complete the marking, it is helpful to quickly identify and classify abnormal situations in the data, and take corresponding measures to make adjustments in a timely manner. In this way, through segmentation and marking processing, a unified data set is formed, which helps to eliminate the problem of data islands, improve the reliability of tool cutting information completed in real time, and provide more comprehensive data support for the optimization of cutting parameters.

[0069] A specific embodiment is given below to introduce the modification process of the basic data information database in detail:

[0070] (1) Identifying the real-time relationship between various parameters based on the tool cutting information completed in real time.

[0071] In specific implementation, the corresponding relationship between different data in the real-time tool cutting information can be determined by acquiring the attribute information obtained when the real-time tool cutting information is completed, and the corresponding relationship can be analyzed using data analysis tools and techniques to identify the real-time correlation between various parameters.

[0072] It should be noted that, since the correlation relationship of data may be different under different cutting conditions, it is necessary to obtain the current real-time correlation in real time.

[0073] (2) Locating the corresponding first association relationship and second association relationship in the basic data information library based on various parameters.

[0074] In specific implementation, according to the correspondence between the tool cutting information completed in real time and the data in the basic data information library, data of the same data type are determined to correspond to each other. The first association relationship and the second association relationship between the corresponding data are also determined to correspond. For example, in one embodiment, the drill diameter data and the drill data in the basic data information library have a first association relationship, and the drill diameter data and the drill data in the tool cutting information completed in real time and the basic data information library are determined to correspond to each other.

[0075] (3) Modifying the first association relationship and the second association relationship according to the real-time association.

[0076] In specific implementation, the real-time relationship can directly replace the corresponding first and second association relationships to obtain new first and second association relationships. When the real-time relationship does not have a corresponding first and second association relationship, the real-time relationship can be directly filled into the basic data information library to obtain new first and second association relationships.

[0077] The optimization and recommendation method for tool cutting parameters provided in this embodiment first identifies the real-time association between various parameters based on the tool cutting information completed in real time, then locates the corresponding first association relationship and second association relationship in the basic data information library based on each parameter, and finally corrects the first association relationship and the second association relationship according to the real-time association. In this way, through the feedback and dynamic correction of the tool cutting information completed in real time, the combination relationship in the basic data information library is closer to the actual cutting process, thereby improving the accuracy of the cutting parameter recommendation. Furthermore, as the processing conditions and cutting status change, the basic data information library can update and adjust the combination relationship in real time. This dynamic correction mechanism enables the basic data information library to better adapt to diverse processing needs and improve the adaptability and flexibility of the system.

[0078] S103, constructing a tool cutting parameter recommendation model based on the real-time tool cutting information and the basic data information library, wherein the tool cutting parameter recommendation model predicts optimal cutting parameters based on the real-time working state information to be cut.

[0079] Specifically, the tool cutting parameter recommendation model is used to predict and recommend optimal cutting parameters based on the tool cutting information completed in real time and the data in the basic data information library.

[0080] In specific implementation, the tool cutting parameter recommendation model includes the corresponding relationship between the tool cutting parameters and the data in the basic data information library, so the optimal cutting parameters can be found according to the tool cutting information completed in real time.

[0081] A specific embodiment is given below to introduce in detail the training process of the tool cutting parameter recommendation model:

[0082] (1) Structuring the real-time tool cutting parameters and extracting entity and relationship information.

[0083] Specifically, the real-time tool cutting parameters are structured and processed by using a traditional structured processing method or a neural network structured processing method. This is not limited in the present embodiment.

[0084] In specific implementation, the real-time tool cutting parameters can be cleaned, format converted, and normalized to ensure the quality and consistency of the data and obtain structured data.

[0085] Furthermore, the entities corresponding to the data and the corresponding relationship information between different entities are determined in the structured data.

[0086] (2) Searching for a parameter position corresponding to the real-time tool cutting parameter according to the basic data information database.

[0087] Specifically, the real-time tool cutting parameters are input into the basic data information base, and the positions of the parameters corresponding to the real-time tool cutting parameters in the planned data information base are obtained.

[0088] It should be noted that the combination relationship between each basic associated data can be determined as the combination relationship between the real-time tool cutting parameters through the parameter position.

[0089] (3) Instantiating the found parameter positions based on the extracted entity and relationship information to complete the instantiation of the basic data information library, and training the tool cutting parameter recommendation model based on the instantiated basic data information library.

[0090] In the specific implementation, in this step, the entity is filled into the basic data information base, and the entity is instantiated through the combination relationship stored in the basic data information base.

[0091] It should be noted that the instantiated basic data information library not only includes the parameter values ​​of each dimension, but also includes the first association relationship and the second association relationship corresponding to the parameter values ​​of each dimension.

[0092] Furthermore, a benchmark model is trained based on the instantiated basic data information database to obtain a trained tool cutting parameter recommendation model.

[0093] In the specific implementation, in this embodiment, based on the instantiated basic data information library, data related to machine tools, tool holders, tools, tool cutting parameters, workpiece materials, and processing techniques can be extracted, and real-time tool cutting parameters can be used as target labels. The weights of the benchmark model can be adjusted through the SGD algorithm to complete the model training of the tool cutting parameter recommendation model.

[0094] The optimization recommendation method for tool cutting parameters provided in this embodiment first structures the real-time tool cutting parameters, extracts entity and relationship information, and then searches for parameter positions corresponding to the real-time tool cutting parameters based on the basic data information base, and then instantiates the found parameter positions based on the extracted entity and relationship information to complete the instantiation of the basic data information base, and trains the tool cutting parameter recommendation model based on the instantiated basic data information base. In this way, by corresponding the real-time tool cutting parameters to the basic data information base and instantiating them, the tool cutting parameter recommendation model can learn the complex relationship between the cutting parameters. Furthermore, as the processing conditions and cutting requirements continue to change, the tool cutting parameter recommendation model is trained through real-time tool cutting parameters so that it can adapt to various processing scenarios and material changes, thereby improving the flexibility and response speed of the production process.

[0095] Another specific embodiment is given below to introduce in detail the process of training the tool cutting parameter recommendation model:

[0096] (1) Locating the cutting parameter node position from the instantiated basic data information library.

[0097] Specifically, the cutting parameter node may be an important data type related to the cutting operation. In specific implementation, all instantiated data in the basic data information library may be determined as cutting parameter nodes, and the cutting parameter node positions thereof may be obtained.

[0098] (2) Locating the cutting work status information corresponding to the cutting parameter node position based on the instantiated first association relationship and the second association relationship.

[0099] In specific implementation, the cutting working state information can be obtained through the corresponding relationship between the cutting working state information and the cutting parameter node position.

[0100] (3) The cutting parameters are taken as output, the cutting working status information is taken as input, and the correlation between the input and output is taken as auxiliary information to construct the first training sample.

[0101] Specifically, the first training sample includes cutting parameters, cutting working state information, and auxiliary data of the associated relationship.

[0102] (4) Training the tool cutting parameter recommendation model based on the first training sample, wherein the tool cutting parameter recommendation model is used to predict the tool cutting parameters and the influence of each parameter value on the cutting force.

[0103] In a specific implementation, the first training sample is input into the benchmark model to learn the correlation between the cutting parameters and the cutting working state information, and obtain a trained tool cutting parameter recommendation model.

[0104] The optimization recommendation method for tool cutting parameters provided in this embodiment first locates the cutting parameter node position from the instantiated basic data information library, and then locates the cutting working state information corresponding to the cutting parameter node position based on the instantiated first association relationship and the second association relationship, and then takes the cutting parameter as output and the cutting working state information as input, and the association relationship between the input and the output as auxiliary information, and constructs the first training sample, and finally trains the tool cutting parameter recommendation model based on the first training sample, and the tool cutting parameter recommendation model is used to predict the influence of the tool cutting parameters and each parameter value on the cutting force. In this way, the tool cutting parameter recommendation model is trained based on the carefully constructed first training sample. During the training process, the model can learn the complex relationship between the cutting parameters and the cutting working state, thereby improving the accuracy of the model in predicting the cutting parameters and their influence on the cutting force. It is helpful to obtain a more reasonable cutting parameter setting in the actual cutting process, and then obtain accurate optimal cutting parameters.

[0105] A specific example is given below to introduce in detail the use of the tool cutting parameter recommendation model:

[0106] (1) Extracting feature information of each dimension from the real-time working status information to be cut.

[0107] Specifically, feature information of each dimension in the real-time working status information to be cut can be obtained based on a traditional feature extraction method, or feature information of each dimension in the real-time working status information to be cut can be obtained based on a feature extraction method of a neural network. This is not limited in this embodiment.

[0108] Furthermore, the real-time working status information to be cut may be preprocessed and then feature extracted.

[0109] (2) Inputting the feature information of each dimension into the pre-trained tool cutting parameter recommendation model.

[0110] In the specific implementation, the extracted feature vector is passed as input data to the trained tool cutting parameter recommendation model.

[0111] (3) The tool cutting parameter recommendation model identifies the correlation between the feature information of each dimension and the cutting parameters, and predicts the optimal cutting parameters and the influence of each cutting parameter on the cutting force.

[0112] In specific implementation, the tool cutting parameter recommendation model predicts the best cutting parameter combination under the current cutting state based on the potential correlation between the input feature information and the cutting parameters.

[0113] The optimization and recommendation method for tool cutting parameters provided in this embodiment first extracts the feature information of each dimension in the real-time working state information to be cut, and then inputs the feature information of each dimension into the pre-trained tool cutting parameter recommendation model, and then the tool cutting parameter recommendation model identifies the correlation between the feature information of each dimension and the cutting parameters, and predicts the optimal cutting parameters and the influence of each cutting parameter on the cutting force. In this way, by extracting the real-time working state information to be cut in real time, it is possible to quickly respond to changes in the cutting environment and ensure that the adjustment of the cutting parameters is always based on the latest real-time working state information to be cut. Furthermore, the tool cutting parameter recommendation model can predict the optimal cutting parameters under the current cutting conditions based on the complex correlation between these feature information and the cutting parameters by analyzing the feature information of each dimension. In this way, the optimal cutting parameters with high reliability and good accuracy can be obtained.

[0114] S104, performing cutting operation according to the optimal cutting parameters.

[0115] Specifically, cutting operation is the operation process of using a tool to cut a workpiece.

[0116] In specific implementation, the machine tool is set with corresponding parameters according to the optimal cutting parameters. For example, in one embodiment, the cutting speed, feed rate, cutting depth and other parameters are adjusted according to the optimal cutting parameters so that the machine tool performs cutting operations according to the recommended optimal parameters.

[0117] S105. Predict the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, monitor the real-time cutting force of the cutting operation, and adjust the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force by the tool cutting force pre-trained large model.

[0118] Specifically, the tool cutting force pre-training large model is used to predict and simulate the cutting force generated during the cutting process. It should be noted that the model is trained with a large amount of historical cutting data and cutting force data, and can more accurately predict the cutting force that may be generated under given cutting parameters.

[0119] Furthermore, the target cutting force is the cutting force that may be generated by the machine tool predicted by the optimal cutting parameters.

[0120] Furthermore, the real-time cutting force can be collected in real time through sensors installed on the machine tool.

[0121] Optionally, before predicting the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, the method further includes:

[0122] (1) Determining target feature information whose correlation with cutting force is greater than a preset threshold based on the basic data information database and the tool cutting information completed in real time.

[0123] Specifically, based on the static data in the basic data information library and the tool cutting information completed in real time, a feature selection method is used to evaluate the information in the basic data information library to obtain a correlation value corresponding to each feature data.

[0124] Furthermore, the specific value of the preset threshold is determined according to actual needs and is not limited in this embodiment.

[0125] Further, the correlation value is compared with a preset threshold, and feature data corresponding to a correlation value greater than the preset threshold is determined as target feature information.

[0126] (2) Using the target feature information and the corresponding cutting force as training samples, training a tool cutting force pre-training model, and determining the target correlation relationship between each cutting parameter and the cutting force.

[0127] In the specific implementation, the selected target feature information and the corresponding cutting force data are input into the benchmark model as training samples, and the model is trained using machine learning algorithms such as neural networks. The model's prediction performance is optimized by adjusting model parameters and hyperparameters such as learning rate. It is understandable that during the training process, the model will gradually learn the complex correlation between various cutting parameters and cutting force, and can accurately predict cutting force under new cutting conditions.

[0128] In specific implementation, the cross-validation technology can be used to evaluate the tool cutting force pre-training large model to ensure the generalization ability and stability of the tool cutting force pre-training large model. At the same time, the hyperparameters of the tool cutting force pre-training large model can be tuned through grid search, random search or Bayesian optimization to further improve the prediction performance of the tool cutting force pre-training large model. Furthermore, the gradient boosting tree ensemble learning method can be used to fuse multiple trained single tool cutting force pre-training large models to improve the accuracy and stability of the prediction.

[0129] The optimization and recommendation method for tool cutting parameters provided in this embodiment determines the target feature information with a correlation degree with the cutting force greater than a preset threshold value based on the basic data information base and the tool cutting information completed in real time; the target feature information and the corresponding cutting force are used as training samples to train the tool cutting force pre-training large model to determine the target correlation relationship between each cutting parameter and the cutting force. In this way, by screening out the target feature information with a correlation degree with the cutting force greater than the preset threshold value, the high correlation and representativeness of the training samples are ensured, thereby improving the prediction accuracy of the tool cutting force pre-training large model.

[0130] A specific embodiment is given below to introduce in detail the use process of the tool cutting force pre-training large model:

[0131] (1) Determine the gap between the target cutting force and the real-time cutting force.

[0132] In specific implementation, the monitoring system collects cutting force data in real time during the cutting process, obtains the real-time cutting force, compares the real-time cutting force with the target cutting force, and calculates the difference between the two. This difference reflects the degree of deviation between the current cutting parameter setting and the target state. It should be noted that the difference between the target cutting force and the real-time cutting force can be obtained by methods such as subtraction.

[0133] (2) Determine a set of candidate adjustment solutions based on the pre-trained large model of tool cutting force and the gap.

[0134] Specifically, a large pre-trained model of tool cutting force is used to predict different cutting parameter combinations and evaluate their potential impact on the cutting force.

[0135] In specific implementation, the gap between the real-time cutting force and the target cutting force is input into the tool cutting force pre-training large model to generate a series of candidate cutting parameter adjustment schemes to form a candidate adjustment scheme set.

[0136] (3) Screening a target adjustment solution from the candidate adjustment solution set based on the tool cutting parameter recommendation model.

[0137] In the specific implementation, each solution in the candidate adjustment solution set is input into the tool cutting parameter recommendation model, and the candidate solutions are evaluated and ranked by comprehensively considering factors such as cutting efficiency, tool life and processing quality.

[0138] Furthermore, the tool cutting parameter recommendation model selects the optimal target adjustment scheme. It can be understood that the target adjustment scheme has the best performance in balancing cutting efficiency, tool life and processing quality.

[0139] (4) Adjusting each cutting parameter according to the target adjustment plan, and reducing the gap between the real-time cutting force and the target cutting force after the adjustment.

[0140] Specifically, the cutting parameters are adjusted according to the target adjustment plan.

[0141] It should be noted that after adjustment, the gap between the real-time cutting force and the target cutting force is narrowed.

[0142] The optimization and recommendation method for tool cutting parameters provided in this embodiment first determines the gap between the target cutting force and the real-time cutting force, and then determines a set of candidate adjustment schemes based on the tool cutting force pre-training large model and the gap, and then screens the target adjustment scheme from the candidate adjustment scheme set based on the tool cutting parameter recommendation model, and finally adjusts each cutting parameter according to the target adjustment scheme, and the gap between the real-time cutting force and the target cutting force is narrowed after adjustment. In this way, by real-time monitoring of the real-time cutting force in the cutting process and comparing it with the target cutting force, the deviation in the cutting process can be immediately discovered, and the cutting parameters can be adjusted in time to ensure that the cutting process is always in the optimal state. Further, the tool cutting force pre-training large model and the tool cutting parameter recommendation model are used to realize the intelligent adjustment of the cutting parameters. These two models analyze and learn a large amount of historical data through machine learning algorithms, and can automatically recommend and adjust the cutting parameters, reduce manual intervention, and improve the accuracy and efficiency of the adjustment. In this way, by continuously narrowing the gap between the real-time cutting force and the target cutting force, it can be ensured that the cutting process is always close to the ideal state, which helps to improve the machining accuracy and surface quality of the workpiece.

[0143] A specific embodiment is given below to introduce the process of obtaining the target adjustment scheme in detail:

[0144] (1) A target correlation relationship between various cutting parameters and cutting force is determined based on the tool cutting force pre-training large model.

[0145] Specifically, the tool cutting force pre-training large model is used to analyze and determine the target correlation relationship between the cutting parameters and the cutting force. In specific implementation, the tool cutting force pre-training large model generates the target correlation relationship through the correlation between the cutting parameters and the cutting force.

[0146] (2) Determining a set of candidate adjustment solutions based on the gap and the target association relationship.

[0147] In the specific implementation, a series of possible cutting parameter adjustment schemes are generated according to the gap between the real-time cutting force and the target cutting force, as well as the target association relationship obtained through the pre-trained large model, forming a set of candidate adjustment schemes.

[0148] It should be noted that the tool cutting force pre-trained large model analyzes the causes of the gap, and based on this, changes the cutting parameters to obtain a set of candidate adjustment solutions.

[0149] (3) Determining the influence of each cutting parameter on the cutting force based on the tool cutting parameter recommendation model.

[0150] Specifically, the tool cutting parameter recommendation model is used to analyze each of the candidate adjustment solutions and quantify the specific impact of different cutting parameters on the cutting force change. It should be noted that the tool cutting parameter recommendation model can be used to evaluate the effects of each solution and select the optimal solution.

[0151] (4) Selecting a target adjustment plan from the set of candidate adjustment plans based on the degree of impact.

[0152] In the specific implementation, based on the analysis of the influence of cutting parameters on cutting force, the optimal target adjustment scheme is selected from the candidate adjustment schemes. It can be understood that the target adjustment scheme can minimize the gap between the real-time cutting force and the target cutting force while meeting the processing requirements, while considering multiple factors such as cutting efficiency, tool life and processing quality.

[0153] The optimization recommendation method for tool cutting parameters provided in this embodiment first determines the target association relationship between each cutting parameter and the cutting force based on the tool cutting force pre-training large model, and then determines the candidate adjustment scheme set based on the gap and the target association relationship, and then determines the degree of influence of each cutting parameter on the cutting force based on the tool cutting parameter recommendation model, and finally selects the target adjustment scheme from the candidate adjustment scheme set based on the degree of influence. In this way, the target association relationship between the cutting parameters and the cutting force can be accurately established through the tool cutting force pre-training large model, and the influence of each parameter on the cutting force in the actual cutting process can be more accurately reflected. Further, in the real-time cutting process, by continuously monitoring the gap between the real-time cutting force and the target cutting force, and based on this gap and the target association relationship, a series of candidate adjustment scheme sets are automatically generated. In this way, the tool cutting parameter recommendation model is used to quantify the degree of influence of the cutting parameters on the cutting force in different candidate adjustment scheme sets, and the target adjustment scheme finally selected not only takes into account the need to narrow the cutting force gap, but also integrates multiple factors such as cutting efficiency, tool life and processing quality, so that an accurate and reliable target adjustment scheme can be obtained.

[0154] The optimization and recommendation method and device for tool cutting parameters provided in the present embodiment first collect basic correlation data of tool cutting parameters, and build a basic data information base based on the basic correlation data, the basic data information base is used to characterize the combination relationship between each basic correlation data, the basic correlation data is the influencing factors of multiple dimensions of tool cutting operation, the basic data information base includes a first correlation relationship between the influencing factors of each dimension and a second correlation relationship between each influencing factor of each dimension in multiple dimensions, and then collect real-time tool cutting information, the real-time tool cutting information at least includes real-time working status information to be cut and real-time completed tool cutting information, and then build a tool cutting parameter recommendation model based on the real-time completed tool cutting information and the basic data information base, the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working status information to be cut, and then performs cutting operation according to the optimal cutting parameters, and finally predicts the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, monitors the real-time cutting force of the cutting operation, and the tool cutting force pre-trained large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force. In this way, the cutting parameter recommendation is realized based on the tool cutting parameter recommendation model and the tool cutting force pre-training model through the basic data information database of multiple dimensions, which is more scientific and accurate than the traditional setting method that relies on experience and intuition. Furthermore, the basic data information database covers detailed data of multiple dimensions, ensuring the comprehensiveness and accuracy of the data. At the same time, the basic data information database provides solid data support for the cutting parameter recommendation model and the cutting force pre-training model, and provides a scientific basis for optimizing the cutting process by deeply analyzing the combination relationship in the basic data information database. Furthermore, the tool cutting parameter recommendation model analyzes the real-time tool cutting information of multiple dimensions, and can accurately predict and recommend the best cutting parameters. Furthermore, the tool cutting force pre-training model can monitor the cutting force in real time and compare and analyze it with the preset target. When it is found that there is a gap between the cutting force and the target, a set of candidate adjustment solutions is generated based on the target association relationship between the cutting force and the cutting parameters, and the tool cutting parameter recommendation model is used to further analyze the influence of each cutting parameter on the cutting force, and finally the best adjustment solution is selected. This dynamic adjustment mechanism ensures that the cutting process is always in the optimal state, effectively improving the processing efficiency and processing quality. In this way, production efficiency and processing quality can be improved by intelligently determining the most suitable cutting parameters.

[0155] Corresponding to the aforementioned embodiment of a method for optimizing and recommending tool cutting parameters, the present application also provides an embodiment of a device for optimizing and recommending tool cutting parameters.

[0156] The embodiment of the device for optimizing and recommending tool cutting parameters in the present application can be applied to a device for optimizing and recommending tool cutting parameters. The device embodiment can be implemented by software, or by hardware, or by a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, the processor of the device for optimizing and recommending tool cutting parameters reads the corresponding computer program instructions in the non-volatile memory into the memory and runs them. From the hardware level, if Figure 3 As shown, it is a hardware structure diagram of the tool cutting parameter optimization recommendation device of the present application, except Figure 3 In addition to the processor, memory, network interface, and non-volatile memory shown, the tool cutting parameter optimization recommendation device in the embodiment may also include other hardware according to the actual function of the tool cutting parameter optimization recommendation device, which will not be described in detail.

[0157] Figure 4 This is a schematic diagram of the structure of the first embodiment of the device for optimizing and recommending cutting parameters of a tool provided in this application. Figure 4 , the device provided in this embodiment includes a construction module 410, a collection module 420, a processing module 430 and an adjustment module 440; wherein;

[0158] The construction module 410 is used to collect basic correlation data of tool cutting parameters, and to construct a basic data information base based on the basic correlation data, wherein the basic data information base is used to characterize the combination relationship between each basic correlation data, wherein the basic correlation data is the influencing factors of multiple dimensions of tool cutting operations, and the basic data information base includes a first correlation relationship between the influencing factors of each dimension and a second correlation relationship between each influencing factor of each dimension in the multiple dimensions;

[0159] The acquisition module 420 is used to acquire real-time tool cutting information, wherein the real-time tool cutting information at least includes real-time working state information to be cut and real-time completed tool cutting information;

[0160] The construction module 410 is further used to construct a tool cutting parameter recommendation model based on the real-time tool cutting information and the basic data information library, and the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut;

[0161] The processing module 430 is used to perform cutting operations according to the optimal cutting parameters;

[0162] The adjustment module 440 is used to predict the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, and monitor the real-time cutting force of the cutting operation. The tool cutting force pre-trained large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force.

[0163] The device of this embodiment can be used to perform Figure 1 The steps, specific implementation principles and implementation processes of the method embodiment shown are similar and will not be repeated here.

[0164] Optionally, the construction module 410 is specifically used to extract target features associated with cutting parameters from the tool cutting information completed in real time based on correlation analysis;

[0165] The construction module 410 is further specifically used to segment each target feature to obtain a data segment set of all target features, wherein the lengths of the data segments in the data segment set are the same;

[0166] The construction module 410 is further specifically used to mark each data segment according to the processing stage to which the target feature belongs and the corresponding tool state.

[0167] Optionally, the acquisition module 420 is specifically used to identify the real-time association between various parameters based on the tool cutting information completed in real time;

[0168] The acquisition module 420 is further specifically used to locate the corresponding first association relationship and second association relationship in the basic data information library based on various parameters;

[0169] The acquisition module 420 is further specifically configured to modify the first association relationship and the second association relationship according to the real-time association.

[0170] Optionally, the construction module 410 is further specifically used to structure the real-time tool cutting parameters and extract entity and relationship information;

[0171] The construction module 410 is further specifically used to search for a parameter position corresponding to the real-time tool cutting parameter according to the basic data information library;

[0172] The construction module 410 is also specifically used to instantiate the found parameter positions based on the extracted entity and relationship information to complete the instantiation of the basic data information library, and train the tool cutting parameter recommendation model based on the instantiated basic data information library.

[0173] Optionally, the construction module 410 is further specifically used to locate the cutting parameter node position from the instantiated basic data information library;

[0174] The construction module 410 is further specifically used to locate the cutting work status information corresponding to the cutting parameter node position based on the instantiated first association relationship and the second association relationship;

[0175] The construction module 410 is further specifically used to construct a first training sample by taking the cutting parameters as output and the cutting work status information as input, and the correlation between the input and the output as auxiliary information;

[0176] The construction module 410 is further specifically used to train the tool cutting parameter recommendation model based on the first training sample, and the tool cutting parameter recommendation model is used to predict the tool cutting parameters and the influence of each parameter value on the cutting force.

[0177] Optionally, the construction module 410 is further specifically used to extract feature information of each dimension from the real-time working status information to be cut;

[0178] The construction module 410 is further specifically used to input feature information of each dimension into the pre-trained tool cutting parameter recommendation model;

[0179] The construction module 410 is also specifically used for the tool cutting parameter recommendation model to identify the correlation between the feature information of each dimension and the cutting parameters, and predict the optimal cutting parameters and the influence of each cutting parameter on the cutting force.

[0180] Optionally, the construction module 410 is further specifically used to determine target feature information having a correlation degree with the cutting force greater than a preset threshold based on the basic data information library and the tool cutting information completed in real time;

[0181] The construction module 410 is also specifically used to use the target feature information and the corresponding cutting force as training samples, train a tool cutting force pre-training model, and determine the target correlation relationship between each cutting parameter and the cutting force.

[0182] Optionally, the adjustment module 440 is specifically used to determine the gap between the target cutting force and the real-time cutting force;

[0183] The adjustment module 440 is further specifically used to determine a set of candidate adjustment solutions based on the tool cutting force pre-trained large model and the gap;

[0184] The adjustment module 440 is further specifically configured to select a target adjustment solution from the candidate adjustment solution set based on the tool cutting parameter recommendation model;

[0185] The adjustment module 440 is further specifically used to adjust various cutting parameters according to the target adjustment scheme, so that the gap between the real-time cutting force and the target cutting force is reduced after the adjustment.

[0186] Optionally, the adjustment module 440 is further specifically used to determine the target association relationship between each cutting parameter and the cutting force based on the tool cutting force pre-trained large model;

[0187] The adjustment module 440 is further specifically configured to determine a set of candidate adjustment solutions based on the gap and the target association relationship;

[0188] The adjustment module 440 is further specifically used to determine the influence of each cutting parameter on the cutting force based on the tool cutting parameter recommendation model;

[0189] The adjustment module 440 is further specifically configured to select a target adjustment solution from the set of candidate adjustment solutions based on the degree of impact.

[0190] Please continue to refer to Figure 3 The present application also provides an optimization and recommendation device for tool cutting parameters, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any one of the methods provided in the first aspect of the present application are implemented.

[0191] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of any one of the methods provided in the present application are implemented.

[0192] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0193] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0194] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for optimizing and recommending cutting parameters of a tool, characterized in that: The method comprises: Collecting basic correlation data of tool cutting parameters, and building a basic data information base based on the basic correlation data, wherein the basic data information base is used to characterize the combination relationship between each basic correlation data, wherein the basic correlation data is the influencing factors of multiple dimensions of tool cutting operations, and the basic data information base includes a first correlation relationship between the influencing factors of each dimension and a second correlation relationship between each influencing factor of each dimension in the multiple dimensions; Collecting real-time tool cutting information, wherein the real-time tool cutting information at least includes real-time work status information to be cut and real-time completed tool cutting information; Building a tool cutting parameter recommendation model based on the tool cutting information completed in real time and the basic data information library, wherein the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut; Perform cutting operations according to the optimal cutting parameters; The target cutting force is predicted based on the tool cutting force pre-trained large model and the optimal cutting parameters, and the real-time cutting force of the cutting operation is monitored. The tool cutting force pre-trained large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force.

2. The method according to claim 1, characterized in that The tool cutting information completed in real time includes data of each object in each link of the cutting operation process, and the data length and data expression form of different links and different objects are different; After collecting the real-time tool cutting information, the method further includes: Extracting target features associated with cutting parameters from the tool cutting information completed in real time based on correlation analysis; Segmenting each target feature to obtain a data segment set of all target features, wherein each data segment in the data segment set has the same length; Each data segment is marked according to the processing stage to which the target feature belongs and the corresponding tool state.

3. The method according to claim 1, characterized in that After collecting the real-time tool cutting information, the method further includes: Identify the real-time correlation between various parameters based on the tool cutting information completed in real time; Locate the corresponding first association relationship and second association relationship in the basic data information library based on the various parameters; The first association relationship and the second association relationship are modified according to the real-time association.

4. The method according to claim 1, characterized in that The constructing of a tool cutting parameter recommendation model based on the tool cutting information completed in real time and the basic data information library comprises: Structuring the real-time tool cutting parameters and extracting entity and relationship information; Searching for a parameter position corresponding to the real-time tool cutting parameter according to the basic data information database; The found parameter positions are instantiated based on the extracted entity and relationship information to complete the instantiation of the basic data information library, and the tool cutting parameter recommendation model is trained according to the instantiated basic data information library.

5. The method according to claim 4, characterized in that The step of training the tool cutting parameter recommendation model according to the instantiated basic data information library comprises: Locate the cutting parameter node position from the instantiated basic data information library; Locating the cutting work status information corresponding to the cutting parameter node position based on the instantiated first association relationship and the second association relationship; The cutting parameters are used as output, the cutting working state information is used as input, and the correlation between the input and the output is used as auxiliary information to construct a first training sample; The tool cutting parameter recommendation model is trained based on the first training sample, and the tool cutting parameter recommendation model is used to predict the tool cutting parameters and the influence of each parameter value on the cutting force.

6. The method according to claim 4, characterized in that The tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut, including: Extracting feature information of each dimension from the real-time working status information to be cut; Inputting the feature information of each dimension into the pre-trained tool cutting parameter recommendation model; The tool cutting parameter recommendation model identifies the correlation between the feature information of each dimension and the cutting parameters, and predicts the optimal cutting parameters and the influence of each cutting parameter on the cutting force.

7. The method according to claim 1, characterized in that Before predicting the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, it also includes: Determine target feature information whose correlation with cutting force is greater than a preset threshold based on the basic data information base and the tool cutting information completed in real time; The target feature information and the corresponding cutting force are used as training samples to train a large pre-training model of tool cutting force and determine the target correlation relationship between each cutting parameter and the cutting force.

8. The method according to claim 1, characterized in that The tool cutting force pre-training large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force, including: determining a difference between the target cutting force and the real-time cutting force; Determine a set of candidate adjustment solutions based on the tool cutting force pre-trained large model and the gap; Selecting a target adjustment solution from the candidate adjustment solution set based on the tool cutting parameter recommendation model; Each cutting parameter is adjusted according to the target adjustment scheme, and after the adjustment, the gap between the real-time cutting force and the target cutting force is reduced.

9. The method according to claim 8, characterized in that The step of selecting a target adjustment solution from the candidate adjustment solution set based on the tool cutting parameter recommendation model includes: A target correlation relationship between each cutting parameter and the cutting force determined based on the tool cutting force pre-trained large model; Determine a set of candidate adjustment solutions based on the gap and the target association relationship; Determining the influence of each cutting parameter on the cutting force based on the tool cutting parameter recommendation model; A target adjustment solution is screened from the set of candidate adjustment solutions based on the degree of impact.

10. A device for optimizing and recommending cutting parameters of a tool, characterized in that: The device includes a construction module, a collection module, a processing module and an adjustment module; wherein, The construction module is used to collect basic correlation data of tool cutting parameters, and to construct a basic data information base based on the basic correlation data, wherein the basic data information base is used to characterize the combination relationship between each basic correlation data, wherein the basic correlation data is the influencing factors of multiple dimensions of tool cutting operations, and the basic data information base includes a first correlation relationship between the influencing factors of each dimension and a second correlation relationship between each influencing factor of each dimension in the multiple dimensions; The acquisition module is used to acquire real-time tool cutting information, wherein the real-time tool cutting information at least includes real-time working state information to be cut and real-time completed tool cutting information; The construction module is further used to construct a tool cutting parameter recommendation model based on the tool cutting information completed in real time and the basic data information library, and the tool cutting parameter recommendation model predicts the optimal cutting parameters based on the real-time working state information to be cut; The processing module is used to perform cutting operations according to the optimal cutting parameters; The adjustment module is used to predict the target cutting force based on the tool cutting force pre-trained large model and the optimal cutting parameters, and monitor the real-time cutting force of the cutting operation. The tool cutting force pre-trained large model adjusts the optimal cutting parameters based on the gap between the target cutting force and the real-time cutting force.

Images