Intelligent machine tool tool wear acousto-optic detection method

Abstract

This invention relates to an intelligent machine tool tool wear acoustic-optical detection method. It pre-constructs a tool type matching model, a tool cutting wear acoustic model of tools of various materials cutting workpieces of different materials, and a tool cutting wear optical characteristic model. Based on these models, it obtains tool cutting wear models and a set of workpiece waste characteristic parameters formed when tools of different materials cut workpieces of different materials. By acquiring real-time tool cutting process parameters during the cutting process, and then using the tool cutting wear model and these parameters, it determines the tool wear condition during the cutting process. Thus, based on the acoustic-optical characteristics of the tool during the cutting process, it achieves accurate judgment of the wear condition of tools of different materials, improving the intelligent detection efficiency of tool wear on machine tools.

Description

Technical Field

[0001] This invention relates to the field of intelligent machine tools, and in particular to an acoustic-optical detection method for tool wear in intelligent machine tools. Background Technology

[0002] With the continuous development of intelligent manufacturing technology, intelligent machine tool technology has become one of the key technologies for realizing intelligent manufacturing. Intelligent machine tools refer to the improvement of existing CNC machine tools through the application of information technology and intelligent technology to achieve intelligent perception and detection during the cutting process.

[0003] Existing CNC machine tools typically include a controller and cutting tools. The controller controls the cutting tools to cut the workpiece according to a preset process to obtain the desired product. During the cutting process, the cutting force generated by the cutting tool cuts the workpiece, and friction is generated between the tool and the workpiece. This mutual friction causes wear on the cutting tool, shortening its service life. Once the cutting tool is severely worn, it will affect the quality of the cut workpiece, reduce the cutting efficiency of the CNC machine tool, and decrease the yield of the cut products.

[0004] Therefore, accurately detecting the wear of intelligent machine tool cutting tools and obtaining timely information about the wear is crucial for predicting the remaining service life of the tools so as to promptly prompt tool replacement. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent machine tool tool wear acoustic and optical detection method in light of the above-mentioned prior art.

[0006] The technical solution adopted by this invention to solve the above-mentioned technical problems is: an intelligent machine tool tool wear acoustic-optical detection method, characterized by comprising the following steps:

[0007] Step 1: Pre-construct tool type sub-matching models that represent different types of tools when cutting is not performed, and form a tool type matching model from all the tool type sub-matching models; wherein, each type of tool corresponds one-to-one with its tool type sub-matching model;

[0008] Step 2: Pre-construct tool cutting wear sound sub-models for the wear generated by tools of different materials during the cutting of workpieces of different materials, and form a tool cutting wear sound model from all the tool cutting wear sound sub-models; wherein, in each tool cutting wear sound sub-model, the tool cutting wear sound sub-model includes an initial cutting wear sound model for the initial cutting wear stage, an intermediate cutting wear sound model for the intermediate cutting wear stage, and a severe cutting wear sound model for the severe cutting wear stage;

[0009] Step 3: Pre-construct tool cutting wear optical feature sub-models for the wear generated by tools of different materials during the cutting of workpieces of different materials, and form a tool cutting wear optical feature model from all the tool cutting wear optical feature sub-models; wherein, in each tool cutting wear optical feature sub-model, there are three types of tool cutting wear optical feature models for the initial cutting wear stage, the intermediate cutting wear stage, and the severe cutting wear stage; the optical feature is the refractive index of the tool;

[0010] Step 4: Pre-construct a set of workpiece waste characteristic parameters for workpieces of various materials that are cut by tools of various materials under different wear conditions; wherein, the workpiece waste characteristic parameters correspond one-to-one with the workpiece material itself, the material tool that performs the cutting action on it, and the wear state of the tool; the workpiece waste characteristic parameters are the gray-scale mean of the histogram corresponding to the workpiece waste and the temperature mean of the infrared image corresponding to the workpiece waste.

[0011] Step 5: Based on the pre-constructed tool cutting wear sound model, tool cutting wear optical feature model, workpiece information and workpiece waste chip feature parameter set, tool cutting wear sub-models are obtained for different types of tools cutting workpieces of different materials under different wear states, and all tool cutting wear sub-models are combined to form the tool cutting wear model;

[0012] Step 6: Obtain tool information, workpiece material, and workpiece waste chip characteristic parameter set during the actual cutting process, and form real-time tool cutting process parameters together with the obtained tool information, workpiece material, and workpiece waste chip characteristic parameter set; wherein, the tool information includes tool material type, real-time tool sound characteristics, and real-time tool surface optical characteristics;

[0013] Step 7: Determine the tool wear condition during the cutting process based on the obtained tool cutting wear model and real-time tool cutting process parameters.

[0014] Improved, in the aforementioned intelligent machine tool tool wear acoustic-optical detection method, the construction process of the tool type matching model is as follows: steps a1 to a5:

[0015] Step a1: Apply the same load voltage to tools of each material type within the same first time period to obtain the current and resistance values ​​of each material type tool; wherein, the total number of material types of tools is denoted as I, the total number of resistance values ​​of the i-th material type tool within the first time period is denoted as M, and the m-th resistance value of the i-th material type tool within the first time period is denoted as r. i,m , 1≤i≤I, 1≤m≤M;

[0016] Step a2: Light is emitted from the same light source onto each type of tool when it is not cutting, within the same second time period, to obtain the optical feature set of each type of tool. The optical feature set is the set of refractive indices of the tool. The total number of refractive indices of the i-th type of tool within the second time period is denoted as K, and the k-th refractive index value of the i-th type of tool within the second time period is denoted as n. i,k , 1≤k≤K;

[0017] Step a3: Based on the obtained sets of electrical resistance and optical characteristics of tools of various material types, obtain the tool type matching parameter feature values ​​for each material type tool; wherein, the tool type matching parameter feature values ​​are as follows:

[0018]

[0019] Where, χ i,T To characterize the tool type matching parameter feature value of the i-th material tool, α i These are intermediate parameters obtained based on the resistance value of the cutting tool of the i-th material. β is the average resistance of the cutting tool of material type i; i These are intermediate parameters obtained based on the optical characteristics of the cutting tool of the i-th material. It is the average refractive index of the cutting tool of the i-th material;

[0020] Step a4: Pre-set the actual values ​​of the tool resistance and the actual values ​​of the optical feature set of each material type of tool when not performing cutting, using a labeling method, to obtain the actual values ​​of the tool type matching parameters for each material type of tool; wherein, the actual value of the tool type matching parameter for the i-th material type tool is labeled as χ. i,s :

[0021]

[0022] Step a5: Based on the obtained feature values ​​of tool type matching parameters for various material tools and the actual values ​​of the tool type matching parameters, a tool type matching model is formed; wherein, the tool type matching model is as follows:

[0023]

[0024] Compared with the prior art, the advantages of the present invention are as follows:

[0025] Firstly, in the intelligent machine tool tool wear acoustic-optical detection method of this invention, a tool type matching model, a tool cutting wear sound model, and a tool cutting wear optical characteristic model are pre-constructed when tools of different materials are cutting workpieces of different materials. Then, based on the tool cutting wear sound model and the tool cutting wear optical characteristic model, tool cutting wear models formed when tools of different materials are cutting workpieces of different materials are obtained, as well as a set of workpiece waste chip characteristic parameters when tools of different materials are cutting different materials under different wear states. By acquiring the real-time process parameters of tool cutting during the cutting process, and based on the obtained tool cutting wear model and real-time process parameters (including workpiece waste chip characteristic parameters), the tool wear condition during the cutting process is determined (specifically, whether the tool is in the early stage of cutting wear, the middle stage of cutting wear, or the severe stage of cutting wear). Thus, based on the acoustic-optical characteristics of the tool during the cutting process, the wear condition of tools of different materials is accurately judged, improving the intelligent detection efficiency of tool wear on machine tools.

[0026] Secondly, because this invention considers and integrates the characteristics of the tool during the cutting of workpieces of different materials and the characteristics of the workpiece waste generated after the cutting process as judgment factors when judging the tool wear condition, it can make a comprehensive judgment from both the characteristics of the tool performing the cutting action and the characteristics of the workpiece waste generated after the cutting action, thereby improving the accuracy of the judgment and avoiding the one-sidedness of judging the tool wear condition by relying solely on the characteristics of the tool. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the intelligent machine tool tool wear acoustic-optical detection method in an embodiment of the present invention;

[0028] Figure 2 This refers to the tool wear recognition rate of the intelligent machine tool tool wear acoustic-optical detection method in this embodiment of the invention for tool wear conditions of different materials. Detailed Implementation

[0029] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0030] This embodiment provides an acoustic-optical detection method for intelligent machine tool tool wear. Specifically, see [link to documentation]. Figure 1 As shown, the intelligent machine tool tool wear acoustic-optical detection method includes the following steps 1 to 7:

[0031] Step 1: Pre-construct tool type sub-matching models that represent different types of tools when cutting is not performed, and form a tool type matching model from all the tool type sub-matching models; wherein, each type of tool corresponds one-to-one with its tool type sub-matching model;

[0032] Step 2: Pre-construct tool cutting wear sound sub-models for the wear generated by tools of different materials during the cutting of workpieces of different materials, and form a tool cutting wear sound model from all the tool cutting wear sound sub-models; wherein, in each tool cutting wear sound sub-model, the tool cutting wear sound sub-model includes an initial cutting wear sound model for the initial cutting wear stage, an intermediate cutting wear sound model for the intermediate cutting wear stage, and a severe cutting wear sound model for the severe cutting wear stage;

[0033] For example, in this embodiment, considering that different types of cutting tools will generate different noises due to friction between the cutting tools and the workpieces of different materials during the cutting process, and that the noise generated by each type of cutting tool at different wear stages has its own unique sound characteristics, the intelligent machine tool tool wear acoustic and optical detection method of this embodiment considers the noise generated by the cutting tool during the cutting process as a characteristic factor of the tool wear condition (i.e., the tool is at different wear stages and corresponds to different wear conditions).

[0034] Step 3: Pre-construct tool cutting wear optical feature sub-models for the wear generated by tools of different materials during the cutting of workpieces of different materials, and form a tool cutting wear optical feature model from all the tool cutting wear optical feature sub-models; wherein, in each tool cutting wear optical feature sub-model, there are three types of tool cutting wear optical feature models for the initial cutting wear stage, the intermediate cutting wear stage, and the severe cutting wear stage; the optical feature is the refractive index of the tool;

[0035] Step 4: Pre-construct a set of workpiece waste characteristic parameters for workpieces of various materials that are cut by tools of various materials under different wear conditions; wherein, the workpiece waste characteristic parameters correspond one-to-one with the workpiece material itself, the material tool that performs the cutting action on it, and the wear state of the tool; the workpiece waste characteristic parameters are the gray-scale mean of the histogram corresponding to the workpiece waste and the temperature mean of the infrared image corresponding to the workpiece waste.

[0036] Step 5: Based on the pre-constructed tool cutting wear sound model, tool cutting wear optical feature model, workpiece information and workpiece waste chip feature parameter set, tool cutting wear sub-models are obtained for different types of tools cutting workpieces of different materials under different wear states, and all tool cutting wear sub-models are combined to form the tool cutting wear model;

[0037] It should be noted that in step 5 of this embodiment, the tool material type in the cutting process is preferably determined by matching based on the tool type matching model established in step 1. The construction of the tool type matching model is described later.

[0038] Step 6: Obtain tool information, workpiece material, and workpiece waste chip characteristic parameter set during the actual cutting process, and form real-time tool cutting process parameters together with the obtained tool information, workpiece material, and workpiece waste chip characteristic parameter set; wherein, the tool information includes tool material type, real-time tool sound characteristics, and real-time tool surface optical characteristics;

[0039] Step 7: Determine the tool wear condition during the cutting process based on the obtained tool cutting wear model and real-time tool cutting process parameters.

[0040] Specifically, in this embodiment, the process of constructing the tool type matching model in step 1 above is as follows: steps a1 to a5:

[0041] Step a1: Apply the same load voltage to tools of each material type within the same first time period to obtain the current and resistance values ​​of each material type tool; wherein, the total number of material types of tools is denoted as I, the total number of resistance values ​​of the i-th material type tool within the first time period is denoted as M, and the m-th resistance value of the i-th material type tool within the first time period is denoted as r. i,m , 1≤i≤I, 1≤m≤M; The resistance value of the tool is the quotient between the load voltage and the current value passing through the tool.

[0042] Step a2: Light is emitted from the same light source onto each type of tool when it is not cutting, within the same second time period, to obtain the optical feature set of each type of tool. The optical feature set is the set of refractive indices of the tool. The total number of refractive indices of the i-th type of tool within the second time period is denoted as K, and the k-th refractive index value of the i-th type of tool within the second time period is denoted as n. i,k , 1≤k≤K;

[0043] Step a3: Based on the obtained sets of electrical resistance and optical characteristics of tools of various material types, obtain the tool type matching parameter feature values ​​for each material type tool; wherein, the tool type matching parameter feature values ​​are as follows:

[0044]

[0045] Where, χ i,T To characterize the tool type matching parameter feature value of the i-th material tool, α i These are intermediate parameters obtained based on the resistance value of the cutting tool of the i-th material. β is the average resistance of the cutting tool of material type i; i These are intermediate parameters obtained based on the optical characteristics of the cutting tool of the i-th material. It is the average refractive index of the cutting tool of the i-th material;

[0046] Step a4: Pre-set the actual values ​​of the tool resistance and the actual values ​​of the optical feature set of each material type of tool when not performing cutting, using a labeling method, to obtain the actual values ​​of the tool type matching parameters for each material type of tool; wherein, the actual value of the tool type matching parameter for the i-th material type tool is labeled as χ. i,s :

[0047]

[0048] Step a5: Based on the obtained feature values ​​of tool type matching parameters for various material tools and the actual values ​​of the tool type matching parameters, a tool type matching model is formed; wherein, the tool type matching model is as follows:

[0049]

[0050] In other words, based on the tool type matching model formed in step a5, when the difference between the actual value of the tool type matching parameter and the characteristic value of the tool type matching parameter for any material tool is less than the preset tool type matching parameter difference Δχ min When this occurs, it indicates that the tool of any material and the tool corresponding to the matching parameter feature value of any tool type belong to the same material.

[0051] As a method for constructing the sound sub-model of tool cutting wear, in this embodiment, the construction process of the above-mentioned sound sub-model of tool cutting wear is as follows: steps b1 to b9:

[0052] Step b1: Pre-collect noise signal sequences and corresponding noise amplitude value sequences of tools of various material types during the entire process of cutting workpieces of different materials; wherein, the noise signal sequence of the tool of material i during the entire process of cutting workpiece of material j is labeled as S. ij Noise signal sequence S ij The noise amplitude value sequence is labeled A ij A ij ={a ij,q};Noise amplitude value sequence A ij The total number of noise amplitude values ​​within the range is denoted as Q, a ij,q It is the q-th noise amplitude value in the noise amplitude value sequence, 1≤q≤Q; the total number of material types of all workpieces is marked as J, 1≤j≤J;

[0053] Step b2: Based on the obtained noise amplitude value sequence of each material type tool during the entire process of cutting workpieces of different materials, calculate the cutting wear sound characteristics of each material type tool during the cutting process of workpieces of different materials; wherein, the cutting wear sound characteristics of the i-th material tool during the cutting process of the j-th material tool are denoted as δ. ij,q :

[0054]

[0055] It should be noted that as the tool continuously performs cutting operations on the workpiece, the wear of the tool during the cutting process will also increase. This will result in the noise generated by the tool cutting the workpiece becoming louder and sharper, that is, the amplitude of the noise signal will vary greatly.

[0056] Step b3: Based on the obtained noise amplitude value sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the noise amplitude variation value that occurs when the tool cuts workpieces of different materials, and take the first noise amplitude variation value that exceeds the first preset noise amplitude variation value as the first noise amplitude singular value; wherein, the first noise amplitude singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as...

[0057]

[0058] in, The first preset noise amplitude change value;

[0059] Step b4: Based on the obtained first noise amplitude singular values ​​and noise amplitude value sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the initial wear sound characteristics of tools of each material type when they begin to enter the initial wear stage during the cutting of workpieces of different materials; wherein, the initial wear sound characteristics of tools of material i entering the initial wear stage when cutting workpieces of material j are denoted as δ. ij,初期 :

[0060]

[0061] Step b5: Based on the obtained noise amplitude value sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the noise amplitude variation value that occurs when the tool cuts workpieces of different materials, and take the first noise amplitude variation value that exceeds the second preset noise amplitude variation value as the second noise amplitude singular value; wherein, the second noise amplitude singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as...

[0062]

[0063] in, This is the second preset noise amplitude change value;

[0064] Step b6: Based on the obtained second noise amplitude singular values ​​and noise amplitude value sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the mid-cutting wear sound characteristics of tools of each material type when they begin to enter the mid-cutting wear stage during the cutting of workpieces of different materials; wherein, the mid-cutting wear sound characteristics of the tool of material i entering the mid-cutting wear stage when cutting workpiece of material j are marked as δ. ij,中期 :

[0065]

[0066] Step b7: Based on the obtained noise amplitude value sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the noise amplitude variation value that occurs when the tool cuts workpieces of different materials, and take the first noise amplitude variation value that exceeds the third preset noise amplitude variation value as the third noise amplitude singular value; wherein, the third noise amplitude singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as...

[0067]

[0068] in, The third preset noise amplitude change value;

[0069] Step b8: Based on the obtained third noise amplitude singular values ​​and noise amplitude value sequences corresponding to the cutting of workpieces of different materials by tools of different material types, calculate the severe wear sound characteristics of tools of different material types when they begin to enter the severe wear stage during the cutting of workpieces of different materials; wherein, the severe wear sound characteristics of tools of material i entering the severe wear stage when cutting workpieces of material j are marked as δ. ij,严重 :

[0070]

[0071] Step b9: Based on the obtained sound characteristics of initial cutting wear, mid-cutting wear, and severe cutting wear, obtain tool cutting wear sound sub-models for the wear generated by tools of different material types during the cutting of workpieces of different materials. All tool cutting wear sound sub-models are then combined to form a tool cutting wear sound model. The tool cutting wear sound model is as follows:

[0072]

[0073] Regarding the aforementioned optical characteristic model of tool cutting wear, in the intelligent machine tool tool wear acousto-optic detection method of this embodiment, the construction process of the aforementioned optical characteristic model of tool cutting wear is as follows: steps c1 to c9:

[0074] Step c1: Pre-collect the refractive index sequence of each type of tool during the entire process of cutting workpieces of different materials; wherein, the refractive index sequence of the tool of type i during the entire process of cutting workpiece of type j is labeled as N. ij n ij ={n ij,p};Refractive index sequence N ij The total amount of refractive index within is denoted as P,n ij,p It is the refractive index sequence N ij The p-th refractive index value within the range, 1≤p≤P;

[0075] Step c2: Based on the obtained refractive index sequence of each material type tool during the entire process of cutting workpieces of different materials, calculate the cutting wear optical characteristics of each material type tool during the cutting process of workpieces of different materials; wherein, the cutting wear optical characteristics of the i-th material tool during the cutting process of the j-th material tool are denoted as η. ij,q :

[0076]

[0077] It should be noted that as the tool continuously performs cutting operations on the workpiece, the wear of the tool during the cutting process will also increase. This results in the tool surface becoming increasingly rough when cutting the workpiece. As the roughness of the tool surface increases, the refractive index of the tool surface will also decrease, and this will be manifested as a large change in the value of the refractive index of the tool surface.

[0078] Step c3: Based on the obtained refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the refractive index change value that occurs when the tool cuts workpieces of different materials, and take the first refractive index change value that is less than or equal to the first preset refractive index change value as the first refractive index singular value; wherein, the first refractive index singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as...

[0079]

[0080] in, This is the first preset refractive index change value;

[0081] Step c4: Based on the obtained first refractive index singular values ​​and refractive index sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the initial wear optical characteristics of tools of each material type when they begin to enter the initial wear stage during the cutting process of workpieces of different materials; wherein, the initial wear optical characteristics of tools of material i entering the initial wear stage when cutting workpieces of material j are denoted as η. ij,初期 :

[0082]

[0083] Step c5: Based on the obtained refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the refractive index change value that occurs when the tool cuts workpieces of different materials, and take the first refractive index change value that is less than the first preset refractive index change value and greater than or equal to the second preset refractive index change value as the second refractive index singular value; wherein, the second refractive index singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as...

[0084]

[0085] in, This is the second preset refractive index change value;

[0086] Step c6: Based on the obtained second refractive index singular values ​​and refractive index sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the mid-cutting wear optical characteristics at the beginning of the mid-cutting wear stage for tools of each material type during the cutting of workpieces of different materials; wherein, the mid-cutting wear optical characteristics at the beginning of the mid-cutting wear stage for tools of material i during the cutting of workpieces of material j are denoted as η. ij,中期 :

[0087]

[0088] Step c7: Based on the obtained refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the refractive index change value that occurs when the tool cuts workpieces of different materials, and take the first refractive index change value that is less than the second preset refractive index change value and greater than or equal to the third preset refractive index change value as the third refractive index singular value; wherein, the third refractive index singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as...

[0089]

[0090] in, This is the third preset refractive index change value;

[0091] Step c8: Based on the obtained singular values ​​of the third refractive index and the refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the optical characteristics of severe wear at the beginning of the severe wear stage when tools of each material type start cutting workpieces of different materials; wherein, the optical characteristics of severe wear at the beginning of the severe wear stage when the tool of material i starts cutting workpiece of material j are marked as η. ij,严重 :

[0092]

[0093] Step c9: Based on the obtained optical characteristics of initial cutting wear, mid-cutting wear, and severe cutting wear, obtain sub-models of tool cutting wear optical characteristics during the cutting of workpieces of different materials, and form a tool cutting wear optical characteristic model from all the tool cutting wear optical characteristic sub-models; the tool cutting wear optical characteristic model is as follows:

[0094]

[0095] Specifically, in this embodiment, in step 5, the process of forming the tool cutting wear model includes the following steps d1 to d9:

[0096] Step d1: Collect the noise amplitude value sequence of the noise signal sequence corresponding to the cutting of different material workpieces by different material tools within the same first preset number of cuttings, and calculate the cutting wear sound characteristics of each material tool when cutting different material workpieces.

[0097] Step d2: Based on the cutting wear sound model and the obtained cutting wear sound characteristics, determine the tool wear condition of each type of tool; wherein, the tool wear condition is defined as the tool being in the initial cutting wear stage, the middle cutting wear stage, or the severe cutting wear stage.

[0098] Step d3: A pre-designated operator observes the tool wear of each material type during the cutting of workpieces of different materials. Based on the observed tool wear and the tool wear determined by the tool cutting wear sound model, the operator obtains a first tool wear recognition rate for each material type. The first tool wear recognition rate corresponding to the cutting of a workpiece of material type j by a tool of material type i is denoted as ξ. ij,1 ;

[0099] Step d4: Collect the refractive index sequence of different material tools during the entire process of cutting different material workpieces within the same first preset number of cuts, and calculate the cutting wear optical characteristics of each material tool when cutting different material workpieces.

[0100] Step d5: Based on the aforementioned optical characteristic model of tool cutting wear and the obtained optical characteristics of each cutting wear, determine the tool wear condition of each type of tool; wherein, the tool wear condition is defined as the tool being in the initial cutting wear stage, the middle cutting wear stage, or the severe cutting wear stage;

[0101] Step d6: The pre-selected personnel observe the tool wear of each type of tool during the cutting of workpieces of different materials. Based on the observed tool wear and the tool wear determined by the tool cutting wear optical feature model, the pre-selected personnel obtain a second tool wear recognition rate. The second tool wear recognition rate corresponding to the cutting of workpieces of type j by a tool of type i is denoted as ξ. ij,2 ;

[0102] Step d7: Based on the obtained first tool wear recognition rate and second tool wear recognition rate, determine the tool wear fusion recognition rate based on the fusion of the tool cutting wear sound model and the tool cutting wear optical feature model; wherein, the tool wear fusion recognition rate corresponding to the cutting of the j-th material workpiece by the tool of the i-th material is marked as ξ. ij ξ ij =ξ ij,1 ·ξ ij,2 ;

[0103] Step d8: Based on the obtained set of chip characteristic parameters for each workpiece, calculate the chip wear characterization factor for each material when cutting workpieces of different materials with tools of different materials; wherein, the calculation method for the chip wear characterization factor is as follows: steps d81 to d84:

[0104] Step d81: Pre-acquire workpiece waste images and infrared images of workpiece waste obtained when the tool cuts workpieces of different materials under different wear conditions. Pre-process each workpiece waste image to obtain pre-processed workpiece waste images corresponding to different wear conditions of the tool. The pre-processed workpiece waste images correspond one-to-one with the type of tool used, the wear condition of the tool, and the material of the workpiece being cut.

[0105] Step d82: Process each preprocessed workpiece waste image to obtain the workpiece waste histogram for each preprocessed workpiece waste image.

[0106] Step d83: Process the histogram of waste chips for each workpiece to obtain the grayscale mean value of the histogram of waste chips for each material workpiece.

[0107] Step d84: Based on the obtained infrared images of the waste chips of each workpiece, process them to obtain the average temperature of the waste chips corresponding to each material of the workpiece.

[0108] Step b5: Based on the obtained average grayscale value and average temperature value, obtain the workpiece chip wear characterization factor corresponding to the chip wear state of each material workpiece; the calculation method of the workpiece chip wear characterization factor is as follows:

[0109]

[0110] Where, λ i,t,j G represents the workpiece chip wear characterization factor when a tool of material i cuts a workpiece of material j at time t. i,t,j T is the average gray value of the workpiece chip histogram corresponding to the cutting of the i-th material workpiece at time t when the workpiece is cut by the i-th material workpiece. i,t,j G is the average temperature and average grayscale value of the workpiece waste generated when a tool of material i cuts a workpiece of material j at time t. i,t,j Average temperature T i,t,j And the wear condition of the tool when cutting the workpiece corresponds one-to-one.

[0111] Step d9: Based on the tool cutting wear sound model, the tool cutting wear optical feature model, the obtained tool wear fusion recognition rate, and the workpiece chip wear characterization factor, a tool cutting wear model is formed; wherein, the tool cutting wear model is as follows:

[0112]

[0113] To improve the accuracy of detecting tool wear conditions for machine tool cutting tools, this embodiment employs a convolutional neural network-based approach to optimize the constructed tool cutting wear model, resulting in an optimized tool cutting wear model. Specifically, the process of forming the optimized tool cutting wear model includes the following steps:

[0114] Step S1: Based on the different wear states of the cutting tool, the reference value of the cutting tool wear state corresponding to one or several wear states of the cutting tool and the workpiece chip characteristic parameters are used as training samples, while the reference value of the cutting tool wear state corresponding to other wear states of the cutting tool and the workpiece chip characteristic parameters are used as test samples.

[0115] Step S2: Input the training samples into a convolutional autoencoder for model pre-training to obtain the model pre-training results;

[0116] Step S3: Continue model training through a convolutional neural network to obtain an initial model of tool cutting wear for this material.

[0117] Step S4: Use the obtained initial model of tool cutting wear to judge the tool wear state of the test sample and output the tool wear state judgment result;

[0118] Step S5: Based on the error between the actual tool wear state of the tool and the obtained tool wear state judgment result, adjust the parameters of the initial tool cutting wear model, and perform iterative update processing on the initial tool cutting wear model after adjusting the model parameters to obtain the final optimized initial tool cutting wear model.

[0119] To test the efficiency of the intelligent machine tool tool wear acoustic-optical detection method in detecting tool wear in this embodiment, an experimental simulation was conducted using this method to detect tool wear of different material tools during the cutting of workpieces of different materials. For details of the experimental simulation, please refer to [link to simulation details]. Figure 2 As shown. Among them, in this Figure 2 middle:

[0120] "Alloy-steel" refers to the detection and recognition rate of tool wear of alloy tools in the process of cutting steel workpieces using the intelligent machine tool wear acoustic and optical detection method of this embodiment;

[0121] "Alloy-Iron" indicates the detection and recognition rate of tool wear of alloy tools in the process of cutting iron workpieces using the intelligent machine tool wear acoustic and optical detection method of this embodiment;

[0122] The same logic applies to other curves.

[0123] from Figure 2 As can be seen, the intelligent machine tool tool wear acoustic and optical detection method of this embodiment has a higher and higher detection and recognition rate of tool wear as the degree of wear increases.

[0124] Although preferred embodiments of the present invention have been described in detail above, it should be clearly understood that various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent machine tool tool wear detection method using acoustic and optical methods, characterized in that, Includes the following steps: Step 1: Pre-construct tool type sub-matching models that represent different types of tools when cutting is not performed, and form a tool type matching model from all the tool type sub-matching models; wherein, each type of tool corresponds one-to-one with its tool type sub-matching model; Step 2: Pre-construct tool cutting wear sound sub-models for the wear generated by tools of different materials during the cutting of workpieces of different materials, and form a tool cutting wear sound model from all the tool cutting wear sound sub-models; wherein, in each tool cutting wear sound sub-model, the tool cutting wear sound sub-model includes an initial cutting wear sound model for the initial cutting wear stage, an intermediate cutting wear sound model for the intermediate cutting wear stage, and a severe cutting wear sound model for the severe cutting wear stage; Step 3: Pre-construct tool cutting wear optical feature sub-models for the wear generated by tools of different materials during the cutting of workpieces of different materials, and form a tool cutting wear optical feature model from all the tool cutting wear optical feature sub-models; wherein, in each tool cutting wear optical feature sub-model, there are three types of tool cutting wear optical feature models for the initial cutting wear stage, the intermediate cutting wear stage, and the severe cutting wear stage; the optical feature is the refractive index of the tool; Step 4: Pre-construct a set of workpiece waste characteristic parameters for workpieces of various materials that are cut by tools of various materials under different wear conditions; wherein, the workpiece waste characteristic parameters correspond one-to-one with the workpiece material itself, the material tool that performs the cutting action on it, and the wear state of the tool; the workpiece waste characteristic parameters are the gray-scale mean of the histogram corresponding to the workpiece waste and the temperature mean of the infrared image corresponding to the workpiece waste. Step 5: Based on the pre-constructed tool cutting wear sound model, tool cutting wear optical feature model, workpiece information and workpiece waste chip feature parameter set, tool cutting wear sub-models are obtained for different types of tools cutting workpieces of different materials under different wear states, and all tool cutting wear sub-models are combined to form the tool cutting wear model; Step 6: Obtain tool information, workpiece material, and workpiece waste chip characteristic parameter set during the actual cutting process, and form real-time tool cutting process parameters together with the obtained tool information, workpiece material, and workpiece waste chip characteristic parameter set; wherein, the tool information includes tool material type, real-time tool sound characteristics, and real-time tool surface optical characteristics; Step 7: Determine the tool wear condition during the cutting process based on the obtained tool cutting wear model and real-time tool cutting process parameters.

2. The intelligent machine tool tool wear acoustic-optical detection method according to claim 1, characterized in that, The process of constructing the tool type matching model is as follows: steps a1 to a5: Step a1, loading the same load voltage to each material type cutter in the same first time period, obtaining the current value and resistance value of each material type cutter; wherein the total number of material types of each material type cutter is marked as I, the total number of resistance values of the obtained i-th material type cutter in the first time period is marked as M, and the m-th resistance value of the i-th material type cutter in the first time period is marked as r i,m , 1≤i≤I, 1≤m≤M; Step a2: Light is emitted from the same light source onto each type of tool when it is not cutting, within the same second time period, to obtain the optical feature set of each type of tool. The optical feature set is the set of refractive indices of the tool. The total number of refractive indices of the i-th type of tool within the second time period is denoted as K, and the k-th refractive index value of the i-th type of tool within the second time period is denoted as n. i,k , 1≤k≤K; Step a3: Based on the obtained sets of electrical resistance and optical characteristics of tools of various material types, obtain the tool type matching parameter feature values ​​for each material type tool; wherein, the tool type matching parameter feature values ​​are as follows: Where, χ i,T To characterize the tool type matching parameter feature value of the i-th material tool, α i These are intermediate parameters obtained based on the resistance value of the cutting tool of the i-th material. β is the average resistance of the cutting tool of material type i; i These are intermediate parameters obtained based on the optical characteristics of the cutting tool of the i-th material. It is the average refractive index of the cutting tool of the i-th material; Step a4: Pre-set the actual values ​​of the tool resistance and the actual values ​​of the optical feature set of each material type of tool when not performing cutting, using a labeling method, to obtain the actual values ​​of the tool type matching parameters for each material type of tool; wherein, the actual value of the tool type matching parameter for the i-th material type tool is labeled as χ. i,s : Step a5: Based on the obtained feature values ​​of tool type matching parameters for various material tools and the actual values ​​of the tool type matching parameters, a tool type matching model is formed; wherein, the tool type matching model is as follows:

3. The intelligent machine tool tool wear acoustic-optical detection method according to claim 2, characterized in that, The construction process of the tool cutting wear sound sub-model is as follows: steps b1 to b9: Step b1: Pre-collect noise signal sequences and corresponding noise amplitude value sequences of tools of various material types during the entire process of cutting workpieces of different materials; wherein, the noise signal sequence of the tool of material i during the entire process of cutting workpiece of material j is labeled as S. ij Noise signal sequence S ij The noise amplitude value sequence is labeled A ij A ij ={a ij,q };Noise amplitude value sequence A ij The total number of noise amplitude values ​​within the range is denoted as Q, a ij,q It is the q-th noise amplitude value in the noise amplitude value sequence, 1≤q≤Q; the total number of material types of all workpieces is marked as J, 1≤j≤J; Step b2: Based on the obtained noise amplitude value sequence of each material type tool during the entire process of cutting workpieces of different materials, calculate the cutting wear sound characteristics of each material type tool during the cutting process of workpieces of different materials; wherein, the cutting wear sound characteristics of the i-th material tool during the cutting process of the j-th material tool are denoted as δ. ij,q : Step b3: Based on the obtained noise amplitude value sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the noise amplitude variation value that occurs when the tool cuts workpieces of different materials, and take the first noise amplitude variation value that exceeds the first preset noise amplitude variation value as the first noise amplitude singular value; wherein, the first noise amplitude singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as... in, The first preset noise amplitude change value; Step b4: Based on the obtained first noise amplitude singular values ​​and noise amplitude value sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the initial wear sound characteristics of tools of each material type when they begin to enter the initial wear stage during the cutting of workpieces of different materials; wherein, the initial wear sound characteristics of tools of material i entering the initial wear stage when cutting workpieces of material j are denoted as δ. ij,初期 : Step b5: Based on the obtained noise amplitude value sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the noise amplitude variation value that occurs when the tool cuts workpieces of different materials, and take the first noise amplitude variation value that exceeds the second preset noise amplitude variation value as the second noise amplitude singular value; wherein, the second noise amplitude singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as... in, This is the second preset noise amplitude change value; Step b6: Based on the obtained second noise amplitude singular values ​​and noise amplitude value sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the mid-cutting wear sound characteristics of tools of each material type when they begin to enter the mid-cutting wear stage during the cutting of workpieces of different materials; wherein, the mid-cutting wear sound characteristics of the tool of material i entering the mid-cutting wear stage when cutting workpiece of material j are marked as δ. ij,中期 : Step b7: Based on the obtained noise amplitude value sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the noise amplitude variation value that occurs when the tool cuts workpieces of different materials, and take the first noise amplitude variation value that exceeds the third preset noise amplitude variation value as the third noise amplitude singular value; wherein, the third noise amplitude singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as... in, The third preset noise amplitude change value; Step b8: Based on the obtained third noise amplitude singular values ​​and noise amplitude value sequences corresponding to the cutting of workpieces of different materials by tools of different material types, calculate the severe wear sound characteristics of tools of different material types when they begin to enter the severe wear stage during the cutting of workpieces of different materials; wherein, the severe wear sound characteristics of tools of material i entering the severe wear stage when cutting workpieces of material j are marked as δ. ij,严重 : Step b9: Based on the obtained sound characteristics of initial cutting wear, mid-cutting wear, and severe cutting wear, obtain tool cutting wear sound sub-models for the wear generated by tools of different material types during the cutting of workpieces of different materials. All tool cutting wear sound sub-models are then combined to form a tool cutting wear sound model. The tool cutting wear sound model is as follows:

4. The intelligent machine tool tool wear acoustic-optical detection method according to claim 3, characterized in that, The process of constructing the optical characteristic model of tool cutting wear is as follows: steps c1 to c9: Step c1: Pre-collect the refractive index sequence of each type of tool during the entire process of cutting workpieces of different materials; wherein, the refractive index sequence of the tool of type i during the entire process of cutting workpiece of type j is labeled as N. ij n ij ={n ij,p };Refractive index sequence N ij The total amount of refractive index within is denoted as P,n ij,p It is the refractive index sequence N ij The p-th refractive index value within the range, 1≤p≤P; Step c2: Based on the obtained refractive index sequence of each material type tool during the entire process of cutting workpieces of different materials, calculate the cutting wear optical characteristics of each material type tool during the cutting process of workpieces of different materials; wherein, the cutting wear optical characteristics of the i-th material tool during the cutting process of the j-th material tool are denoted as η. ij,q : Step c3: Based on the obtained refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the refractive index change value that occurs when the tool cuts workpieces of different materials, and take the first refractive index change value that is less than or equal to the first preset refractive index change value as the first refractive index singular value; wherein, the first refractive index singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as... in, This is the first preset refractive index change value; Step c4: Based on the obtained first refractive index singular values ​​and refractive index sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the initial wear optical characteristics of tools of each material type when they begin to enter the initial wear stage during the cutting process of workpieces of different materials; wherein, the initial wear optical characteristics of tools of material i entering the initial wear stage when cutting workpieces of material j are denoted as η. ij,初期 : Step c5: Based on the obtained refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the refractive index change value that occurs when the tool cuts workpieces of different materials, and take the first refractive index change value that is less than the first preset refractive index change value and greater than or equal to the second preset refractive index change value as the second refractive index singular value; wherein, the second refractive index singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as... in, This is the second preset refractive index change value; Step c6: Based on the obtained second refractive index singular values ​​and refractive index sequences corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the mid-cutting wear optical characteristics at the beginning of the mid-cutting wear stage for tools of each material type during the cutting of workpieces of different materials; wherein, the mid-cutting wear optical characteristics at the beginning of the mid-cutting wear stage for tools of material i during the cutting of workpieces of material j are denoted as η. ij,中期 : Step c7: Based on the obtained refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the refractive index change value that occurs when the tool cuts workpieces of different materials, and take the first refractive index change value that is less than the second preset refractive index change value and greater than or equal to the third preset refractive index change value as the third refractive index singular value; wherein, the third refractive index singular value that occurs when the tool of material i-th type cuts the workpiece of material j-th type throughout the entire process is marked as... in, This is the third preset refractive index change value; Step c8: Based on the obtained singular values ​​of the third refractive index and the refractive index sequence corresponding to the cutting of workpieces of different materials by tools of each material type, calculate the optical characteristics of severe wear at the beginning of the severe wear stage when tools of each material type start cutting workpieces of different materials; wherein, the optical characteristics of severe wear at the beginning of the severe wear stage when the tool of material i starts cutting workpiece of material j are marked as η. ij,严重 : Step c9: Based on the obtained optical characteristics of initial cutting wear, mid-cutting wear, and severe cutting wear, obtain sub-models of tool cutting wear optical characteristics during the cutting of workpieces of different materials, and form a tool cutting wear optical characteristic model from all the tool cutting wear optical characteristic sub-models; the tool cutting wear optical characteristic model is as follows:

5. The intelligent machine tool tool wear acoustic-optical detection method according to claim 4, characterized in that, In step 5, the process of forming the tool cutting wear model includes the following steps: Step d81: Pre-acquire workpiece waste images and infrared images of workpiece waste obtained when the tool cuts workpieces of different materials under different wear conditions. Pre-process each workpiece waste image to obtain pre-processed workpiece waste images corresponding to different wear conditions of the tool. The pre-processed workpiece waste images correspond one-to-one with the type of tool, the wear condition of the tool, and the material of the workpiece being cut. Step d82: Process each preprocessed workpiece waste image to obtain the workpiece waste histogram for each preprocessed workpiece waste image. Step d83: Process the histogram of waste chips for each workpiece to obtain the grayscale mean value of the histogram of waste chips for each material workpiece. Step d84: Based on the obtained infrared images of the waste chips of each workpiece, process them to obtain the average temperature of the waste chips corresponding to each material of the workpiece. Step d85: Based on the obtained average gray value and average temperature value, obtain the workpiece waste wear characterization factor that represents the waste wear state of each material workpiece. Step d86: Based on the pre-constructed tool cutting wear sound model, tool cutting wear optical characteristic model, and corresponding workpiece waste wear characterization factor, a tool cutting wear model is obtained when tools of different materials cut workpieces of different materials under different wear conditions.

Images