On-line monitoring method of internal field temperature during milling process

Abstract

The invention discloses an online monitoring method for internal field temperature in a milling process, comprising: step one: data acquisition: using an infrared thermal imager to photograph a tool and a workpiece in real time to obtain a tool external field temperature thermal map; step two: data processing: extracting the temperature of the tool external field surface point; step three: heat transfer analysis: establishing a heat conduction direct problem model for milling; step four: virtual simulation: simulating the heat transfer process of the milling cutter in milling, applying heat fluxes of different sizes to the tool internal field, solving the tool internal temperature field, and obtaining the temperature rise sensitivity coefficient of each point inside the tool at different temperatures; step five: heat transfer inverse solution: constructing a heat transfer inverse problem model, treating the heat transfer inverse solution problem as an optimization problem, and solving the heat flux size of the tool internal field at each moment; step six: solving the internal field temperature: substituting the temperature rise sensitivity coefficient of each point inside the tool and the heat flux size of the internal field into the heat conduction direct problem model to solve the tool internal field temperature.

Description

Technical Field

[0001] The invention belongs to the technical field of mechanical cutting processing, and specifically relates to an online monitoring method for internal field temperature during milling processing. Background Art

[0002] During the machining process, a large amount of heat generated by friction and plastic deformation forms three heat sources, namely the first deformation zone (shear slip zone), the second deformation zone (tool chip contact zone) and the tertiary deformation zone (extrusion friction rebound zone). Figure 1 As shown in the figure, the temperature rise of the tool is mainly due to the heat generated in the tool-chip contact area entering the tool through the contact heat conduction between the tool and the chip. A large amount of heat entering the tool is concentrated in the tiny tool-chip interface area, resulting in extremely high temperatures in this area. Existing studies have shown that the heat generated in the tool-chip contact area is the main cause of tool wear. The extremely high temperature caused by it causes the chemical elements of the tool and the workpiece to diffuse with each other in the solid state. This diffusion wear changes the original composition and structure of the material, making the tool fragile. In addition, the increase in temperature will also intensify the oxidation reaction between the oxygen in the air and the alloy elements in the tool, producing softer oxides that fall off during the cutting process. This oxidation wear further aggravates the wear of the tool.

[0003] Milling is an intermittent machining operation. Since the single cutting edge of the milling cutter is not in continuous contact with the workpiece, the thermal load on the cutting edge is periodically discontinuous, so it needs to withstand fatigue cycles of thermal cooling and thermal loading. Compared with turning, the tool is subjected to more severe thermal loads during milling than turning, which will inevitably increase tool wear, shorten tool life, and affect machining accuracy and production efficiency. Therefore, by online monitoring of the maximum temperature in the tool-chip contact area, severe tool wear caused by excessively high temperatures in the tool-chip contact area can be avoided, which is of great significance for preventing premature tool aging and reducing machining costs.

[0004] According to the research on tool internal field temperature by scholars at home and abroad, the current methods for measuring tool internal field temperature during cutting are mainly divided into two categories, namely direct measurement method and indirect measurement method.

[0005] The direct measurement method refers to the use of test equipment to directly measure the internal field temperature of the tool. The current mainstream direct measurement methods include thermocouple temperature measurement technology and infrared imager temperature measurement technology. Thermocouples can be used to measure the temperature of a certain point near the tool-chip contact area. However, due to (i) the extremely high temperature gradient in the tool-chip contact area, the temperature within 1 mm near the peak temperature is hundreds of degrees Celsius away from the peak temperature; and (ii) the embedded thermocouple will affect the mechanical and thermal properties of the tool, increase tool wear, reduce tool life, and interfere with the normal conduction of heat flow in the tool. Therefore, it is difficult to use thermocouples to directly measure the temperature of the tool's internal field. Since the chips will block the tool-chip contact area during the cutting process, the infrared thermal imager cannot capture the highest temperature in the tool-chip contact area. In order to solve this problem, some scholars have used infrared thermal imagers to measure the internal temperature of transparent tool yttrium aluminum garnet materials during machining. Although the related research on transparent material tools directly measured the temperature of the tool-chip contact area using infrared thermal imagers, the physical properties and shape of the transparent material itself cannot meet the needs of actual tools and cannot be used in engineering. Therefore, it is difficult to directly measure the temperature inside the tool using an infrared thermal imager.

[0006] The indirect measurement method uses the easy-to-measure variables in the measurement process to measure the variables to be measured through various mathematical calculations and estimation methods based on the mathematical relationship between the easy-to-measure variables and the variables to be measured. According to the different indirect measurement models, indirect measurement can be divided into three categories: analytical model method, artificial intelligence method, and heat transfer inverse method.

[0007] The analytical model method solves the tool internal field temperature by establishing an analytical model of the relationship between cutting force and cutting heat. However, since the heat distribution in the tool-chip contact area is affected by the thermophysical properties of the tool-workpiece material, cutting process parameters, cutting fluid, tool wear condition, tool-chip contact pressure and other factors, and there is a complex coupling of mechanical and thermal behaviors in the tiny tool-chip contact area, it is difficult to accurately estimate the heat distribution in the tool-chip contact area. At present, heat distribution is often calculated based on many simplified assumptions, such as considering the tool-chip contact as completely close contact, thereby ignoring the effect of contact thermal resistance on heat distribution. Therefore, under the condition that the heat distribution coefficient cannot be accurately estimated, the accuracy of the model is difficult to guarantee.

[0008] The artificial intelligence method uses a large amount of data to train the ANN artificial neural network model, which has the characteristics of high prediction accuracy and fast response speed. However, in actual engineering, it is difficult to measure the maximum temperature in the tool-chip contact area by direct measurement. The data obtained by finite element simulation can only be used to train the artificial intelligence model instead of the actual real data. Since both the finite element method and the artificial intelligence model have errors, their simultaneous application will lead to a cascade decline in solution accuracy, making it difficult to accurately predict the internal field temperature of the tool.

[0009] The inverse heat transfer method is a method of obtaining the internal field temperature of the tool by solving the inverse problem of heat conduction. Generally speaking, the use of the temperature of a certain position of the tool that can be directly measured by the test equipment to estimate the heat flux flowing into the tool is called the "inverse problem of heat conduction". However, the existing inverse heat transfer model regards the increase in cutting temperature as the result of the continuous action of heat flux in a single tool-chip contact area, while the increase in milling temperature is the result of the periodic intermittent action of heat flux in the tool-chip contact area on multiple milling cutter edges. Therefore, the existing inverse heat transfer model cannot be used for online monitoring of the internal field temperature of the tool during the milling process. Summary of the invention

[0010] In view of this, an object of the present invention is to provide a method for online monitoring of the internal field temperature during milling processing, so as to achieve the purpose of online monitoring of the internal field temperature of the tool during the milling process in real time.

[0011] In order to achieve the above object, the present invention provides the following technical solutions:

[0012] A method for online monitoring of internal field temperature during milling processing comprises the following steps:

[0013] Step 1: Data collection: During the milling process, use an infrared thermal imager to take real-time photos of the tool and workpiece to obtain a tool external field temperature thermal map for extracting the external field temperature;

[0014] Step 2: Data processing: extract the temperature of the tool external field surface point in the tool external field temperature heat map;

[0015] Step 3: Heat transfer analysis: Establish the heat conduction direct problem model of milling;

[0016] Step 4: Virtual simulation: simulate the heat transfer process of the milling cutter during milling, apply heat fluxes of different sizes to the internal field of the tool, solve the internal temperature field of the tool, and obtain the temperature rise sensitivity coefficient of each point inside the tool at different temperatures;

[0017] Step 5: Heat transfer inverse solution: Construct an inverse heat transfer problem model, regard the heat transfer inverse solution problem as an optimization problem, and solve the heat flux size in the tool internal field at each time;

[0018] Step 6: Solve the internal field temperature: Substitute the temperature rise sensitivity coefficient of each point inside the tool and the heat flux of the internal field into the heat conduction direct problem model to solve the internal field temperature of the tool.

[0019] Furthermore, in step 3, the method for constructing the heat conduction forward problem model is:

[0020] Assuming that the heat flux in the tool-chip contact area at any time is known, the corresponding temperature field calculated by Duhamel integration is:

[0021]

[0022] Where T(x,y,z,t) represents the temperature of the point with coordinates (x,y,z) at time t; T0(x,y,z,t) represents the initial temperature of the point with coordinates (x,y,z); represents the temperature rise caused by the unit pulse heat flux from the initial moment to the moment t at the point with coordinates (x, y, z); q(λ) represents the heat flux entering the tool at the moment λ;

[0023] The time domain is discretized into intervals of length Δt, namely: Δt, 2Δt, 3Δt...kΔt, where k is the number of time steps and (k-1)Δt≤t k ≤kΔt; then the discretization form of the temperature field is:

[0024]

[0025] Among them, T(x i ,y i ,z i ,k) means the coordinates at the kth moment are (x i ,y i ,z i ) point; T0 represents the initial temperature; q n represents the heat flux entering the tool at the kth moment; is the temperature rise sensitivity coefficient, and:

[0026]

[0027] in Indicates the temperature rise sensitivity coefficient at the knth moment; represents the temperature rise caused by the unit pulse heat flux at the k-n+1th moment; Represents the temperature rise caused by the unit pulse heat flux at the knth moment.

[0028] Furthermore, in step 4, finite element simulation is used to obtain the temperature rise sensitivity coefficient at different temperatures. The method is as follows:

[0029] 1) Pre-processing: Establish a three-dimensional model of the milling cutter, import it into the finite element simulation software, specify the tool material, set the material's thermal physical performance parameters, and divide the mesh;

[0030] 2) Determine the size of the tool-chip contact area: The length of the tool-chip contact area is determined by the back cutting amount, and the width of the tool-chip contact area is:

[0031]

[0032] Among them, lcontact Indicates the width of the tool-chip contact area; t c is the undeformed cutting thickness; φ n is the normal friction angle; λ is the average friction angle of the rake face; α n is the normal rake angle of the tool;

[0033] 3) Apply heat flux: Apply different periodic intermittent heat fluxes in the tool-chip contact area, and the values ​​are:

[0034] q(t)=c j =j×10 6 W / m 2 , (j=1, 2, 3...J)

[0035] The cycle time is: T = πD / v, D is the diameter of the milling cutter, and v is the cutting speed;

[0036] Peak load time: t = l a / v,l a is the contact arc length between the milling cutter and the workpiece;

[0037] 4) Determine the heat transfer conditions: The convection heat transfer coefficient during milling is:

[0038]

[0039] Where h is the convection heat transfer coefficient between the tool surface and the air; k is the thermal conductivity of air; D is the diameter of the milling cutter; Re r is the rotation Reynolds number; ω is the angular velocity of the milling cutter; ρ is the air density; μ is the dynamic viscosity coefficient of air; N u represents the average Nusselt number;

[0040] 5) Solve the finite element model and record the temperature at the same position in the finite element simulation as the surface temperature measured by the infrared thermal imager, denoted as T m (c j ).;

[0041] 6) Let q(t) = c j +1, re-solve the finite element model and record the temperature rise caused by the unit pulse heat flux Get the temperature rise sensitivity coefficient

[0042] Furthermore, in step 5, the least squares function is used to obtain the deviation between the temperature measured by the infrared thermal imager and the temperature solved from the direct problem, and the sequential Tikhonov regularization method is used for optimization to obtain:

[0043]

[0044] Among them, Tm,k (q k ) indicates that the mth measurement point is measured at time k by the heat flux q k The resulting temperature; Y m,k represents the temperature of the mth measurement point measured by the infrared thermal imager at time k; αΩ(q k ,q k-1 ,q k-2 ) represents the regularization term, and:

[0045]

[0046] Among them, α represents the regularization parameter; α1(q k -q k-1 ) 2 , α2(q k -2q k-1 +q k-2 ) 2 denote the 0th, 1st and 2nd order regularization terms, respectively, and α0, α1 and α2 denote the 0th, 1st and 2nd order regularization parameters, respectively;

[0047] Substituting the discretized form of the temperature field, we can obtain the inverse estimate of the heat flux in the tool internal field at each moment:

[0048]

[0049] in, represents the temperature of the mth measurement point at time k caused by the heat flux at all previous times, and:

[0050]

[0051] in, represents the temperature rise sensitivity coefficient of the mth measurement point at the initial moment; It represents the temperature rise sensitivity coefficient of the mth measurement point at the time k-n+1.

[0052] The beneficial effects of the present invention are:

[0053] The method for online monitoring of internal field temperature during milling processing of the present invention takes into account that the milling cutter needs to withstand fatigue cycles of thermal cooling and thermal loading during milling processing, utilizes the inverse heat transfer method and combines Duhamel's thermal superposition theorem to realize online monitoring of the internal field temperature of the tool during milling processing; the present invention does not require additional hardware equipment except for an infrared thermal imager, is easy to implement in engineering, and is operational; the computational complexity of the present invention is low, and can be directly applied to edge-end computing, so it is easy to realize online monitoring of the internal field temperature in engineering.

[0054] Specifically, in the direct measurement method, compared with thermocouple measurement, the present invention does not affect the performance of the tool and the material, and can measure the peak temperature of the tool's internal field; the present invention measures the tool's external field temperature by an infrared thermal imager, which will not be affected by chip shielding, and its application does not need to be performed under specific tool materials;

[0055] In the indirect measurement method, compared with the analytical model, the present invention does not need to accurately estimate the heat distribution in the tool-chip contact area, and no additional sensors are required to monitor the cutting force in actual engineering applications; compared with the artificial intelligence prediction method, the present invention does not need a large amount of engineering data to train the model, and has better practicality when it is difficult to obtain engineering data to train the model.

[0056] Note: The "inner field" mentioned in this article refers to the tool-chip contact area where the tool contacts the chip; the "outer field" refers to the remaining surface of the tool that is not in contact with the chip. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to make the purpose, technical solution and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration:

[0058] Figure 1 Schematic diagram of three cutting deformation zones;

[0059] Figure 2 It is a flowchart of the method for online monitoring of internal field temperature during milling process of the present invention;

[0060] Figure 3 This is a schematic diagram of tool field temperature monitoring and infrared thermal imager installation;

[0061] Figure 4 The thermal image of the tool field temperature taken by the infrared thermal imager;

[0062] Figure 5 Schematic diagram of the tool-chip contact area during milling;

[0063] Figure 6 It is a curve diagram of heat flux variation in the tool-chip contact area during the milling process;

[0064] Figure 7 Schematic diagram of convective heat transfer during milling;

[0065] Figure 8 A technical roadmap for solving the heat conduction forward problem;

[0066] Fig. 9 This is the installation structure diagram of the milling processing experimental equipment;

[0067] Fig.10 It is the temperature curve diagram of the inner field of the tool;

[0068] Fig.11 Comparison curve of calculated temperature and measured temperature values ​​at different measuring points outside the tool field. DETAILED DESCRIPTION

[0069] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0070] like Figure 2 The figure is a flowchart of the method for online monitoring of the internal temperature during milling processing of the present invention. The method for online monitoring of the internal temperature during milling processing of this embodiment includes the following steps:

[0071] Step 1: Data acquisition: During the milling process, an infrared thermal imager is used to photograph the tool and workpiece in real time to obtain a tool external field temperature thermal map for extracting the external field temperature.

[0072] The present invention uses an infrared thermal imager to obtain the external temperature of the tool, and solves the heat flux in the tool-chip contact area through the heat transfer inverse method, and then solves the internal temperature of the tool, thereby solving the problem that the infrared thermal imager cannot directly obtain the internal temperature of the tool. Figure 3 shown.

[0073] Step 2: Data processing: extract the temperature of the tool external field surface point in the tool external field temperature heat map.

[0074] The thermal image obtained by the infrared thermal imager is as follows: Figure 4 In order to extract useful information from the thermal image, the supporting software of the infrared thermal imager is used to obtain the maximum temperature of the tool surface and the temperature of the adjacent pixel grid from the thermal image in real time, and these temperatures are used as the external field temperature of the tool to be input in this embodiment to solve the internal field temperature. The temperature of each point obtained is recorded as Y m (m=1,2,3……)

[0075] Step 3: Heat transfer analysis: Establish a heat conduction forward problem model for milling.

[0076] like Figure 5 As shown in the figure, when the milling cutter is cutting, the cutting edge of the tool cuts the workpiece intermittently. When the main cutting edge contacts the workpiece, there will be a surface heat source in the tool-chip contact area whose area changes with time. When the cutting edge cuts into the workpiece, the effective area of ​​the surface heat source increases with time, and then remains unchanged until the cutting edge cuts out of the workpiece, and the effective area decreases to zero with time. Therefore, Figure 6As shown in the figure, during the milling process, the heat flux in the inner field tool-chip contact zone exhibits periodic and discontinuous characteristics.

[0077] The three-dimensional transient heat conduction problem of the tool can be expressed by the following heat transfer differential equation:

[0078]

[0079] Where T represents the temperature of the point in the tool with the heat source as the origin and coordinates (x, y, z) at the tth moment; ρ, k and C p represent density, thermal conductivity and specific heat respectively.

[0080] The initial conditions are:

[0081] T(x,y,z,t)=T0 (2)

[0082] Wherein, T0 represents the temperature at the initial moment.

[0083] The boundary conditions of the tool-chip contact area are:

[0084]

[0085] in, represents the derivative of the normal line stretched outward along the surface of the object; q(t) represents the heat flux entering the tool at time t; k * Indicates the thermal conductivity of the tool.

[0086] The boundary conditions of the surface of the tool exposed to the environment are:

[0087]

[0088] Where h represents the convective heat transfer coefficient between the tool surface and the air, T ∞ Indicates the ambient temperature.

[0089] The boundary conditions between the tool and the chuck are:

[0090]

[0091] Among them, h * Represents the contact thermal conductivity between the tool and the chuck.

[0092] The boundary conditions of heat conduction can be determined by the following mechanism model:

[0093] The length of the chip contact area is determined by the actual cutting depth, and the width l contact Determined by the following formula:

[0094]

[0095] Among them, tc is the undeformed cutting thickness, which is determined by the cutting process parameters; φ n is the normal friction angle, which is determined by the normal front angle α n and the average friction angle λ; λ is the average friction angle of the rake face, which is determined by the normal rake angle of the tool; α n is the normal rake angle of the tool.

[0096] like Figure 7 As shown in the figure, the heat in the inner field tool-chip contact area is conducted in the tool, and is convectively cooled with the environment during the tool rotation. When the milling cutter rotates, the convective heat transfer between the tool surface and the surrounding air can be simplified as the heat dissipation process of the cylindrical surface in the air. The convective heat transfer coefficient formula can be expressed as:

[0097]

[0098] Where h is the convection heat transfer coefficient between the tool surface and the air; k is the thermal conductivity of air; D is the diameter of the milling cutter; Re r is the rotation Reynolds number; ω is the angular velocity of the milling cutter; ρ is the air density; μ is the dynamic viscosity coefficient of air; N u represents the average Nusselt number.

[0099] Since only a small part of the milling cutter is in contact with the air inside the chuck, adiabatic boundary conditions can be assumed.

[0100] Equations (1)-(5) represent the forward problem of heat conduction. Therefore, if the thermophysical parameters of all materials, boundary conditions, and the heat flux applied to the tool-chip contact area are clearly known, the internal field temperature of the tool can be obtained by solving the above equations.

[0101] Since the thermophysical parameters of the material (such as thermal conductivity, specific heat, etc.) and boundary conditions often change with the cutting temperature during the actual cutting process, the heat conduction problem presents nonlinear characteristics, which increases the difficulty of solving the problem and reduces the efficiency of solving it. Therefore, this embodiment uses a combination of Duhamel theory and finite element method to quickly solve the problem and meet the needs of online monitoring, such as Figure 8 shown.

[0102] When solving a heat conduction problem in which the heat source term or the boundary condition term both vary with time, the Duhamel theorem is used to link it with the same heat conduction problem in which the boundary condition and / or the heat source term do not vary with time. Specifically, the heat conduction forward problem model of this embodiment is constructed as follows: Assuming that the heat flux at any time in the tool-chip contact area is known, the corresponding temperature field calculated by the Duhamel integral is:

[0103]

[0104] Where T(x,y,z,t) represents the temperature of the point with coordinates (x,y,z) at time t; T0(x,y,z,t) represents the initial temperature of the point with coordinates (x,y,z); represents the temperature rise caused by the unit pulse heat flux from the initial moment to the moment t at the point with coordinates (x, y, z); q(λ) represents the heat flux entering the tool at the moment λ;

[0105] The time domain is discretized into intervals of length Δt, namely: Δt, 2Δt, 3Δt...kΔt, where k is the number of time steps and (k-1)Δt≤t k ≤kΔt; then the discretization form of the temperature field is:

[0106]

[0107] Among them, T(x i ,y i ,z i ,k) means the coordinates at the kth moment are (x i ,y i ,z i ) point; T0 represents the initial temperature; q n represents the heat flux entering the tool at the kth moment; is the temperature rise sensitivity coefficient, and:

[0108]

[0109] in, Indicates the temperature rise sensitivity coefficient at the knth moment; represents the temperature rise caused by the unit pulse heat flux at the k-n+1th moment; Represents the temperature rise caused by the unit pulse heat flux at the knth moment.

[0110] Since the direct problem is nonlinear, at a certain point in time The value is related to the temperature at that point. When the tool temperature field changes, The value also changes accordingly. In order to obtain the tool under different temperature fields The present invention uses the finite element method to establish The offline database is used for online monitoring.

[0111] Step 4: Virtual simulation: Use simulation software to simulate the heat transfer process of the milling cutter during milling, apply heat fluxes of different sizes to the internal field of the tool, solve the internal temperature field of the tool, and obtain the temperature rise sensitivity coefficient of each point inside the tool at different temperatures.

[0112] In order to obtain different temperature fields The present embodiment uses finite element analysis to analyze the heat transfer process inside the tool, aiming to simulate the temperature field inside the tool under different heat fluxes. This embodiment uses finite element simulation to obtain the temperature rise sensitivity coefficient at different temperatures The method is as follows:

[0113] 1) Pre-processing: Establish a three-dimensional model of the milling cutter, import it into the finite element simulation software, specify the tool material, set the material's thermal physical performance parameters, and divide the mesh;

[0114] 2) Determine the size of the tool-chip contact area: The length of the tool-chip contact area is determined by the depth of the cutter. Substituting the tool geometry angle into equation (6), we can obtain the width of the tool-chip contact area l contact ;

[0115] 3) Apply heat flux: Apply different periodic intermittent heat fluxes in the tool-chip contact area, and the values ​​are:

[0116] q(t)=c j =j×10 6 W / m 2 , (j=1, 2, 3...J)

[0117] The cycle time is: T = πD / v, D is the diameter of the milling cutter, and v is the cutting speed;

[0118] Peak load time: t = l a / v,l a is the contact arc length between the milling cutter and the workpiece;

[0119] 4) Determine the heat transfer conditions: Substitute the tool geometry parameters, milling process parameters and environmental parameters into equation (7) to obtain the convective heat transfer coefficient during the milling process;

[0120] 5) Solve the finite element model and record the temperature at the same position in the finite element simulation as the surface temperature measured by the infrared thermal imager, denoted as T m (c j ).;

[0121] 6) Let q(t) = c j +1, re-solve the finite element model and record the temperature rise caused by the unit pulse heat flux Substituting it into formula (10), we get the temperature rise sensitivity coefficient

[0122] The value of J depends on the nonlinearity of the model, and under different values ​​of J, the temperature of the temperature field in the finite element analysis needs to include the temperature from the tool cutting in to the tool cutting out in the actual milling process. When monitoring the temperature of the tool-chip contact area online, taking the kth moment as an example, first read the temperature Y of the corresponding surface point measured by the infrared thermal imager in real time.m , by making Y m T corresponding to different J values ​​at the current moment m (c j ) to minimize the difference. Will Substitute into equation (9) and solve the direct problem.

[0123] Step 5: Heat transfer inverse solution: Construct an inverse heat transfer problem model, regard the heat transfer inverse problem as an optimization problem, and solve the heat flux in the tool internal field at each time.

[0124] At present, the idea of ​​solving the inverse problem is usually to transform the inverse problem of heat conduction into an optimization problem, so that the deviation between the temperature measured by the infrared thermal imager and the temperature calculated from the direct problem is minimized. The traditional method of solving the minimum deviation is to use the least squares function, which is as follows:

[0125]

[0126] Among them, T m,k (q k ) indicates that the mth measurement point is measured at time k by the heat flux q k The resulting temperature; Y m,k represents the temperature of the mth measurement point measured by the infrared thermal imager at time k;

[0127] The inverse problem of heat conduction is also called an "ill-posed" problem. In practice, a small measurement error will cause a huge fluctuation in the model solution. Therefore, a suitable regularization method is needed to adapt it. This embodiment uses the sequential Tikhonov regularization method to solve the ill-posed inverse problem.

[0128]

[0129] Among them, αΩ(q k ,q k-1 ,q k-2 ) represents the regularization term, and:

[0130]

[0131] Among them, α represents the regularization parameter; α1(q k -q k-1 ) 2 , α2(q k -2q k-1 +q k-2 ) 2denote the 0th, 1st, and 2nd order regularization terms, respectively, and α0, α1, and α2 denote the 0th, 1st, and 2nd order regularization parameters, respectively. Penalties are applied according to the order. The value of α is determined by experimental trial and error to strike a balance between smoothing the solution and minimizing the temperature deviation.

[0132] Substitute equation (9) into equation (12) and solve L(q k ) to obtain the minimum value of q k The inverse estimator of :

[0133]

[0134] in, Represents the temperature rise sensitivity coefficient of the mth measurement point at the initial moment.

[0135] in, It represents the temperature of the mth measurement point at time k caused by the heat flux at all previous times, that is:

[0136]

[0137] Once the heat flux q at each time is known k As well as the temperature rise sensitivity coefficient of each point in the tool-chip contact area under different temperature fields, the transient temperature of the tool internal field changing with time can be calculated through Equation 9.

[0138] Step 6: Solve the internal field temperature: Substitute the temperature rise sensitivity coefficient of each point inside the tool and the heat flux of the internal field into the heat conduction direct problem model to solve the internal field temperature of the tool.

[0139] The method for online monitoring of internal field temperature during milling processing of the present invention is verified below with reference to specific examples.

[0140] 1. Milling processing experimental test environment

[0141] In order to prove the feasibility of the online monitoring method of internal field temperature during the milling process proposed in the present invention, this embodiment takes a specific milling experiment as an example. The experiment was carried out on a VCML850 CNC milling machine. The infrared thermal imager model was NEC H2640. Data processing and finite element analysis were performed using a Windows system computer with Intel (R) Core (TM) i7-10750H CPU and NVIDIA GeForce RTX 2060. ANSYS WORKBENCH 2021R1 was used as virtual testing software. In order to verify the method of this embodiment, this embodiment uses a KISTLER 5167A built-in digital acquisition charge amplifier and a 9139AA cutting force meter to measure the cutting force data during the machining process. The tool material used in the experiment is cemented carbide, and its specific geometric angles and thermal physical properties are shown in Tables 1 and 2. The processing parameters are shown in Table 3.

[0142] Table 1 Geometric parameters of carbide milling cutters

[0143]

[0144] Table 2 Thermal physical parameters of cemented carbide milling cutter temperature change

[0145]

[0146] Table 3 Experimental processing parameters

[0147]

[0148] During the milling process, an infrared thermal imager is used to collect the external temperature data of the tool. The overall installation layout and relative position relationship between the various experimental equipment are shown in the figure below. Fig. 9 As shown in the figure, the cutting force meter is fixed horizontally on the machine tool workbench with 4 clamps, connected to the charge amplifier and then directly connected to the computer through a network cable. The sampling frequency of the infrared thermal imager used in the experiment is 450 frames per minute, which meets the requirements of online monitoring. At the same time, the tool external field temperature heat map collected by the infrared thermal imager is exported in real time to the matching thermal map analysis software InfReC Analyzer NS9500 Standard to extract the specific value of the tool external field temperature.

[0149] 2. Experimental results analysis

[0150] according to Figure 2The technical route demonstrated uses the surface temperature captured by the infrared thermal imager as the input of the model to obtain the maximum value of the tool internal field temperature in the actual milling process. From the start of processing to 3.5 seconds of processing, the tool internal field temperature rises, and the processing tends to be stable after 3.5 seconds. The tool internal field temperature change curve of the stable processing process is shown in Fig.10 As shown by the solid black line in the middle. As can be seen from the figure, the tool internal field shows a trend of oscillating upward, and the final temperature varies around 300 degrees Celsius, with a maximum of 350.19 degrees. During the milling process, there is a short separation between the tool and the workpiece after each pass, and the temperature of the milling cutter will drop sharply at this time. Therefore, the entire internal field temperature curve shows an alternating rise and fall state.

[0151] In order to verify the accuracy of the proposed method, this embodiment verifies the solved internal field temperature. First, the tool internal field temperature obtained by the method used in this embodiment is directly verified by combining the cutting force mechanism model with the finite element method. The specific steps are as follows:

[0152] 1) According to the research of Merchant ME, based on the shear plane model shown in formula (16), the total heat flux q in the tool-chip contact area on the rake face during cutting is calculated by the cutting force measured by the cutting force meter: sh :

[0153]

[0154] Among them, F c 、F f 、F sh They represent the tangential force, feed force and the composite shear force of the two, γ represents the tool rake angle; τ sh A represents the shear stress in the tool-chip contact area. sh Indicates the area of ​​the tool-chip contact zone; V c represents the cutting speed and λ represents the chip compression ratio.

[0155] 2) As shown in equation (17), the heat q flowing into the tool tool From the heat distribution coefficient α * It is determined that usually about 4-10% of the friction heat flows into the tool. * = 0.04 is used to calculate the heat flowing into the tool.

[0156] q tool =α * q sh (17)

[0157] 3) With k = 0.5s as the time interval, the time domain is discretized into intervals of length Δt: Δt, 2Δt, 3Δt...kΔt, where k is the number of time steps. The heat flux entering the tool at the kth moment is solved by the above method.

[0158] 4) The heat flux q at each time k As a heat source input into the ANSYS finite element model, solve the tool transient temperature field, output the tool internal field temperature at each moment, such as Fig.10 Shown by the red line.

[0159] In addition, this embodiment also adopts an indirect verification method, that is, using the method described in this embodiment to solve the external field temperature and compare it with the data collected by the infrared thermal imager. The specific steps are as follows: extract the external field temperature of the tool surface from the data preprocessing in step 2 as the input of the method described in the present invention, solve the external field point temperature at another position and verify its consistency with the measured temperature.

[0160] The above verification method uses root mean square error (RMS), mean absolute error (MAE) and relative root mean square error (RRMS) as evaluation indicators. The effectiveness of the method is judged by comparing the mean absolute error and root mean square error of the calculated results with the measured values. The specific definitions are shown in equations (18)-(20).

[0161]

[0162]

[0163] RRMS=RMS / T max (20)

[0164] in, is the temperature value output by the method, y i is the measured temperature value. For direct verification, T max is the maximum value of the internal field temperature obtained in the mechanism-finite element combination method. For indirect verification, T max is the maximum temperature actually obtained by the infrared thermal imager. The RMS, MAE and RRMS of each point are shown in Table 4. The comparison between the temperature calculated by the indirect verification method and the measured temperature value is shown in Fig.11 shown.

[0165] Table 4 RME, MAE and RRMS

[0166]

[0167] By observation Fig.10It can be seen that the tool internal field temperature curve solved by the method of this embodiment in the stable cutting stage can be well fitted with the tool internal field temperature curve solved by the method of combining mechanism and finite element. The root mean square error and absolute average error values ​​of the two are 23.77 and 20.78 respectively, and the relative root mean square error is 6.8%. Therefore, this embodiment can more accurately realize the measurement of the tool internal field temperature. In addition, Fig.11 As shown, the tool external field temperature curve is solved by the method of this embodiment and compared with the actual measured temperature. The maximum root mean square error and absolute average error of the two are 4.06 and 3.51 respectively, which indirectly proves the accuracy of this embodiment.

[0168] Table 5 shows the computational complexity comparison of the method of this embodiment and the mechanism-finite element combined method in the process of verifying the accuracy of the method of this embodiment, where the solution time refers to the time required to use each method to solve the internal field temperature of the tool at a single moment.

[0169] Table 5 Computational cost of solving the heat conduction problem by the method of this embodiment and the finite element method

[0170]

[0171] It can be seen that using the finite element method to solve the tool external field temperature is time-consuming. In contrast, the method of this embodiment has low computational complexity while achieving the same effect and can meet the needs of online monitoring.

[0172] The above-described embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or changes made by those skilled in the art based on the present invention are within the protection scope of the present invention. The protection scope of the present invention shall be subject to the claims.

Claims

1. A method for online monitoring of internal field temperature during milling process. Features: The steps include: Step 1: Data collection: During the milling process, use an infrared thermal imager to take real-time photos of the tool and workpiece to obtain a tool external field temperature thermal map for extracting the external field temperature; Step 2: Data processing: extract the temperature of the tool external field surface point in the tool external field temperature heat map; Step 3: Heat transfer analysis: Establish the heat conduction direct problem model of milling; Step 4: Virtual simulation: simulate the heat transfer process of the milling cutter during milling, apply heat fluxes of different sizes to the internal field of the tool, solve the internal temperature field of the tool, and obtain the temperature rise sensitivity coefficient of each point inside the tool at different temperatures; Step 5: Heat transfer inverse solution: Construct an inverse heat transfer problem model, regard the heat transfer inverse solution problem as an optimization problem, and solve the heat flux size in the tool internal field at each time; Step 6: Solve the internal field temperature: Substitute the temperature rise sensitivity coefficient of each point inside the tool and the heat flux of the internal field into the heat conduction direct problem model to solve the internal field temperature of the tool; In step 3, the method for constructing the heat conduction forward problem model is: Assuming that the heat flux in the tool-chip contact area at any time is known, the corresponding temperature field calculated by Duhamel integration is: Where T(x,y,z,t) represents the temperature of the point with coordinates (x,y,z) at time t; T 0 (x,y,z,t) represents the initial temperature of the point with coordinates (x,y,z); represents the temperature rise caused by the unit pulse heat flux from the initial moment to the moment t at the point with coordinates (x, y, z); q(λ) represents the heat flux entering the tool at the moment λ; The time domain is discretized into intervals of length Δt, namely: Δt, 2Δt, 3Δt...kΔt, where k is the number of time steps and (k-1)Δt≤t k ≤kΔt; then the discretization form of the temperature field is: Among them, T(x i ,y i ,z i ,k) means the coordinates at the kth moment are (x i ,y i ,z i ) point temperature; T 0 represents the initial temperature; q n represents the heat flux entering the tool at the kth moment; is the temperature rise sensitivity coefficient, and: in, Indicates the temperature rise sensitivity coefficient at the knth moment; represents the temperature rise caused by the unit pulse heat flux at the k-n+1th moment; It represents the temperature rise caused by the unit pulse heat flux at the knth moment; In step 4, finite element simulation is used to obtain the temperature rise sensitivity coefficient at different temperatures. The method is as follows: 1) Pre-processing: Establish a three-dimensional model of the milling cutter, import it into the finite element simulation software, specify the tool material, set the material's thermal physical performance parameters, and divide the mesh; 2) Determine the size of the tool-chip contact area: The length of the tool-chip contact area is determined by the back cutting amount, and the width of the tool-chip contact area is: Among them, l contact Indicates the width of the tool-chip contact area; t c is the undeformed cutting thickness; φ n is the normal friction angle; λ is the average friction angle of the rake face; α n is the normal rake angle of the tool; 3) Apply heat flux: Apply different periodic intermittent heat fluxes in the tool-chip contact area, and the values ​​are: q(t)=c j =j×10 6 W / m 2 ,(j=1,2,3……J) The cycle time is: T = πD / v, D is the diameter of the milling cutter, and v is the cutting speed; Peak load time: t = l a / v,l a is the contact arc length between the milling cutter and the workpiece; 4) Determine the heat transfer conditions: The convection heat transfer coefficient during milling is: Where h is the convection heat transfer coefficient between the tool surface and the air; k is the thermal conductivity of air; D is the diameter of the milling cutter; Re r is the rotation Reynolds number; ω is the angular velocity of the milling cutter; ρ is the air density; μ is the dynamic viscosity coefficient of air; N u represents the average Nusselt number; 5) Solve the finite element model and record the temperature at the same position in the finite element simulation as the surface temperature measured by the infrared thermal imager, denoted as T m (c j ); 6) Let q(t) = c j +1, re-solve the finite element model and record the temperature rise caused by the unit pulse heat flux Get the temperature rise sensitivity coefficient 2. According to the method for online monitoring of internal field temperature during milling processing according to claim 1, Features: In step 5, the least squares function is used to obtain the deviation between the temperature measured by the infrared thermal imager and the temperature solved from the direct problem, and the sequential Tikhonov regularization method is used for optimization to obtain: Among them, T m,k (q k ) indicates that the mth measurement point is measured at time k by the heat flux q k The resulting temperature; Y m,k represents the temperature of the mth measurement point measured by the infrared thermal imager at time k; αΩ(q k ,q k-1 ,q k-2 ) represents the regularization term, and: Among them, α represents the regularization parameter; α 1 (q k -q k-1 ) 2 , α 2 (q k -2q k-1 aq k-2 ) 2 denote the 0th, 1st and 2nd order regularization terms respectively, α 0 , α 1 and α 2 denote the 0th, 1st, and 2nd order regularization parameters respectively; Substituting the discretized form of the temperature field, we can obtain the inverse estimate of the heat flux in the tool internal field at each moment: in, represents the temperature of the mth measurement point at time k caused by the heat flux at all previous times, and: in, It represents the temperature rise sensitivity coefficient of the mth measurement point at the initial moment; It represents the temperature rise sensitivity coefficient of the mth measurement point at the time k-n+1.

Images