Multi-scale finite element simulation method for abrasive belt grinding temperature field of robot

Abstract

The invention discloses a multi-scale finite element simulation method for a robot abrasive belt grinding temperature field, and relates to the technical field of grinding simulation. According to the multi-scale model provided by the invention, a micro cutting area and a macro workpiece surface temperature field are deeply studied, the workpiece surface in the cutting area is divided into a grinding area and a heat transfer area according to the contact condition of the workpiece surface and abrasive particles, the heat generation and heat conduction processes in the abrasive belt grinding and polishing process of a robot can be effectively simulated, and the working efficiency is improved. The temperature field distribution of the complex machining process is obtained with low calculation cost, according to the simulation model, the temperature field distribution of the workpiece surface under different machining parameters can be obtained by changing different grinding parameters, and the method has important significance on optimal selection of the grinding parameters and quality analysis of the grinding surface.

Description

Technical Field

[0001] The invention belongs to the technical field of grinding simulation, and in particular relates to a multi-scale finite element simulation method for a robot abrasive belt grinding temperature field. Background Art

[0002] As the core component of aero-engines, turbine blades have the characteristics of complex surface contours, high processing difficulty and high dimensional accuracy requirements, which are difficult to meet with traditional processing methods. Robotic belt grinding is widely used in engine blade preparation as an efficient and flexible finishing method. However, when difficult-to-process materials (such as nickel-based high-temperature alloys) are ground, the grinding heat generated can easily cause surface burns, which will seriously change the physical and chemical properties of the workpiece surface. Therefore, it is very necessary to model and analyze the temperature of robotic belt grinding to achieve belt grinding temperature prediction. During the grinding process, most of the grinding energy is converted into heat energy. Part of the heat energy is transferred to the abrasive particles, chips and the environment, while the other part is called input heat energy, which enters the workpiece and increases the temperature of the workpiece surface. The uncertainty of heat input may lead to local high temperature, resulting in damage to the workpiece surface.

[0003] Finite element method has been proven to be reliable and effective in analyzing grinding heat problems. As a flexible grinding method, the contact area between the abrasive belt and the workpiece is significantly longer than that of grinding wheel grinding. Since the cutting speed of the abrasive belt is much higher than the robot feed speed, the residence time of a point on the workpiece in the cutting area is much longer than the single cutting time of a single abrasive grain (about 104 to 105 times). This huge difference in time scale makes the computational cost of directly simulating the cutting process unbearable.

[0004] At present, the equivalent heat source method is the main method for analyzing the temperature field of belt grinding. This method predicts the grinding temperature distribution through parabolic heat flux intensity, rectangular heat flux intensity, and elliptical heat flux intensity. According to the elastic contact characteristics of belt grinding, a parabolic heat source is used for numerical analysis. However, in the actual grinding process, the heat flux changes with time, so it is challenging to simulate this dynamic process through numerical simulation. It is precisely because of the shape and elastic characteristics of the dynamic grinding zone of the contact wheel that the grinding temperature of the belt cannot be accurately predicted for a long time, and the direct multi-abrasive simulation process is too complicated and the computational cost is unbearable. Summary of the invention

[0005] In view of the shortcomings of the existing technology, this paper proposes a multi-scale finite element simulation method for the temperature field of robotic belt grinding. The entire robotic belt grinding process is discretized into a macroscopic belt machine workpiece contact finite element model, a microscopic force-heat coupling cutting zone finite element model and a macroscopic temperature field finite element model according to the scale. The temperature field distribution of the workpiece under the entire robotic belt grinding is obtained at a lower computational cost.

[0006] The technical solution of the present invention is:

[0007] A multi-scale finite element simulation method for temperature field of robot belt grinding comprises the following steps:

[0008] Step 1: Establish a macroscopic abrasive belt machine workpiece contact finite element model and use the macroscopic abrasive belt machine workpiece contact finite element model to simulate the grinding process, and obtain the normal force curve of a single abrasive grain on the abrasive belt and the length of the contact area between the abrasive belt and the workpiece during the grinding process; the contact area is the area where the groove exists after the abrasive belt is polished; the contact area length is the length of the groove on the workpiece surface after polishing;

[0009] The establishment process of the macroscopic sanding machine workpiece contact finite element model is as follows:

[0010] Firstly, a three-dimensional model of the rubber contact wheel, a three-dimensional model of the abrasive belt and a three-dimensional model of the workpiece are established, and the material parameters, analysis steps, action relationships and loads of the three-dimensional model of the rubber contact wheel, the three-dimensional model of the abrasive belt and the three-dimensional model of the workpiece are set. Then, the three-dimensional model of the rubber contact wheel, the three-dimensional model of the abrasive belt and the three-dimensional model of the workpiece are meshed respectively and divided into a plurality of units, each unit including a plurality of nodes, so as to obtain a macroscopic abrasive belt machine workpiece contact finite element model.

[0011] The rubber contact wheel comprises a rubber layer and a metal wheel, and the abrasive belt comprises a substrate and abrasive grains;

[0012] The material parameters include conductivity, density, elastic modulus, Poisson's ratio, expansion coefficient, inelastic heat share, specific heat capacity, constitutive model and damage failure model;

[0013] Step 2: According to the normal force curve of a single abrasive grain and the length of the contact area between the abrasive belt and the workpiece during the grinding process obtained in step 1, a finite element model of a microscopic force-heat coupling cutting zone is established, and the microscopic grinding process and heat transfer process are simulated using the finite element model of the microscopic force-heat coupling cutting zone to obtain the length of the heat-affected zone and the temperature distribution fitting curve of the grinding plane;

[0014] The heat affected zone is an area on the workpiece where there is a temperature difference with the ambient temperature;

[0015] The method for establishing the finite element model of the microscopic force-heat coupling cutting zone is as follows: extracting a section of the abrasive belt and the workpiece within a set range in the macroscopic abrasive belt machine workpiece contact finite element model, simplifying the substrate of the contact area from a section of circular arc to a straight line of equal length, the sand grains are still arranged on the substrate, and the workpiece is divided into a grinding area and a heat transfer area according to the length of the contact area, the length of the grinding area is the length of the contact area, and the rest is the heat transfer area, and the material parameters, analysis steps, action relationships and loads of the abrasive belt and the workpiece are set, and the mesh is re-divided to obtain the finite element model of the microscopic force-heat coupling cutting zone;

[0016] The normal force curve of the single abrasive grain is used as the amplitude of the load of the finite element model of the microscopic force-heat coupling cutting zone, and is loaded onto each abrasive grain according to time;

[0017] The set range is a range consisting of the contact area and the set distance areas on both sides thereof, and the length of the set range is greater than the length of the contact area;

[0018] Step 3: A macroscopic temperature field finite element model is established based on the regional length of the heat-affected zone and the temperature distribution fitting curve of the grinding plane obtained in step 2, and the heat source transfer process is simulated using the macroscopic temperature field finite element model to obtain the temperature field distribution of the workpiece and the temperature field distribution curve of the depth of the heat-affected zone;

[0019] The method for constructing the macroscopic temperature field finite element model is as follows: establishing a three-dimensional model of the workpiece and a three-dimensional model of a movable thin plate as an equivalent heat source, then setting material parameters, analysis steps, action relationships and loads of the three-dimensional model of the workpiece and the three-dimensional model of the movable thin plate, and finally performing meshing to obtain the macroscopic temperature field finite element model;

[0020] The length of the movable thin plate is the length of the heat-affected zone; the temperature distribution fitting curve of the grinding plane is applied to the three-dimensional model of the movable thin plate;

[0021] The heat-affected zone depth temperature field distribution curve is a temperature curve formed by the temperature along the depth direction starting from the contact point between the movable thin plate and the workpiece.

[0022] Compared with the prior art, the beneficial effects of the present invention are:

[0023] The multi-scale model proposed in the present invention conducts an in-depth study on the temperature field of the micro-cutting zone and the macro-workpiece surface, and divides the workpiece surface in the cutting zone into a grinding zone and a heat transfer zone according to the contact with the abrasive particles. It can effectively simulate the heat generation and heat conduction process in the process of robot belt grinding and polishing, and obtain the temperature field distribution of complex processing processes at a lower computing cost. According to the simulation model of the present invention, the temperature field distribution of the workpiece surface under different processing parameters can be obtained by changing different grinding parameters, which is of great significance to the optimal selection of grinding parameters and the analysis of grinding surface quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of micro-scale abrasive cutting in an embodiment of the present invention;

[0025] Among them, (a) is a schematic diagram of the temperature rise part and the heat dissipation part in the cutting zone; (b) is a schematic diagram of the normal load loading process in the cutting zone of a single abrasive;

[0026] Figure 2 It is a general flow chart of a multi-scale finite element simulation method of a temperature field of a robot belt grinding in an embodiment of the present invention;

[0027] Figure 3 Schematic diagram of a finite element model of a micro-macro thermal-mechanical coupling temperature field in an embodiment of the present invention;

[0028] Among them, (a) is the macroscopic belt sander workpiece contact finite element model; (b) is the microscopic force-heat coupling cutting zone finite element model; (c) is the macroscopic temperature field finite element model;

[0029] Figure 4 It is a schematic diagram of simulation results of a finite element model of macroscopic sanding machine workpiece contact in an embodiment of the present invention;

[0030] Among them, (a) is a schematic diagram of the groove morphology in the cutting area; (b) is the normal force curve of a single abrasive particle;

[0031] Figure 5 It is a schematic diagram of the simulation results of the finite element model of the microscopic force-heat coupling cutting zone in an embodiment of the present invention;

[0032] Among them, (a) is the surface temperature rise curve of belt grinding; (b) is the simulation calculation result of temperature distribution in the steady-state cutting zone; (c) is the fitting curve;

[0033] Figure 6 Schematic diagram of simulation results of a macroscopic temperature field finite element model in an embodiment of the present invention;

[0034] Among them, (a) is the simulation result of temperature field distribution; (b) is the depth temperature distribution curve of heat affected zone. DETAILED DESCRIPTION

[0035] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0036] The macroscopic robot belt grinding temperature field simulation is a complex process. Due to the shape and elastic properties of the dynamic grinding zone of the contact wheel, as well as the complex multi-abrasive model and heat dissipation conditions, direct macroscopic temperature field simulation will lead to simulation result errors and increased calculation amount. Therefore, the present invention chooses to discretize this grinding process into a macroscopic belt machine workpiece contact finite element model, a microscopic force-heat coupling cutting zone finite element model and a macroscopic temperature field finite element model. Based on the coupling relationship between grinding force and grinding heat at the microscale, the temperature field of the workpiece surface under different processing parameters is obtained.

[0037] The thermal-mechanical coupling analysis method is used to analyze the thermal-mechanical coupling effect in the belt grinding process, and the finite element solution equation of the temperature field is established;

[0038] According to the principle of energy conservation, we get formula (1):

[0039]

[0040] Where V is the volume of the continuous medium, v i is the velocity field, ρ is the density of the medium, t is the time, U is the internal energy per unit mass of the medium, Q is the heat flow per unit volume, and b i is the unit volume force, S is the boundary, H is the boundary heat flux, P i is the pressure on the unit boundary;

[0041] According to the principle of conservation of momentum, the force balance equation is established as shown in formula (2):

[0042]

[0043] Substituting formula (2) into formula (1), we get the energy conservation equation of thermal-mechanical coupling, as shown in formula (3):

[0044]

[0045] Among them, x j is the direction axis, σ ij is the Cauchy stress component. The pressure on the unit boundary is expressed as P using the Cauchy stress. ij =n i σ ij , P ij is the pressure per unit boundary, n i is the direction of the boundary normal, j represents the direction and j = 1, 2, 3, representing the x, y, z directions in the rectangular coordinate system respectively;

[0046] According to the principle of virtual work, the structural displacement u is establishedi Satisfied relations:

[0047]

[0048] In the formula, δ ui represents virtual displacement;

[0049] The relationship between the unit displacement vector and the node displacement vector is shown in formula (5):

[0050] W(x,t)=N(x)u N (t) (5)

[0051] Where W(x,t) is the unit displacement vector, N(x) is the shape function matrix, x is the coordinate vector, and u N (t) is the node displacement vector;

[0052] The deformation rate obtained by derivation of formula (5) with respect to time is shown in formula (6):

[0053]

[0054] Among them, S(x,t ) is the deformation rate, is the derivative vector of the node displacement vector with respect to time;

[0055] The unit temperature vector is expressed as formula (7):

[0056] T m (x,t)=T(t)B(x) (7)

[0057] Where, T m (x, t) is the element temperature vector, T(t) is the node temperature vector, and B(x) is the interpolation function of the temperature field;

[0058] The unit strain vector is formula (8):

[0059] ε(x,t)=LW(x,t)=u N (t)G(x) (8)

[0060] Among them, ε(x,t) is the element strain vector, L is the differential operator; G(x) is the geometric matrix representing the relationship between the element strain vector and the node displacement vector;

[0061] According to the definition of temperature gradient: temperature gradient is the rate of change of temperature along the normal direction (x direction) of the isothermal surface, the temperature gradient vector is constructed as shown in formula (9):

[0062]

[0063] Among them, g(x,t) is the temperature gradient vector, A(x) is the relationship matrix between the unit temperature gradient vector and the node temperature vector;

[0064] The finite element equations for analyzing the transient temperature field and thermal stress-strain field of the structure obtained from formula (9) are shown in formulas (10) and (11):

[0065]

[0066]

[0067] Among them, K u is the structural stiffness matrix, K Τ is the heat conduction matrix, M T is the thermal stiffness matrix, M u is the thermal coupling matrix, is the derivative vector of the node temperature vector with respect to time, F(t) is the mechanical load vector, C u is the heat capacity matrix, D is the dissipation vector; R is the heat load vector;

[0068] Formula (12) can be obtained from formula (10) and (11):

[0069]

[0070] Where P(t)=D+Q+K T T(t);

[0071] For the problem of contact friction heating, the heat load vector R is expressed as:

[0072] R=κF f V r (13)

[0073] Where κ is the heat-to-work conversion coefficient, F f is the friction force, V r is the relative sliding velocity of the friction surface;

[0074] Heat conduction in the micro-cutting process can be regarded as the process of heat transfer between the abrasive and the workpiece and heat transfer from the workpiece to the outside, such as Figure 1 As shown. The sliding process between the abrasive and the workpiece is dry friction, so it can be assumed that the abrasive accumulation and the workpiece have a heat source effect. According to the analysis, the temperature function is related to the time and space position, but because the time domain and the space domain are not coupled, the finite element form can be established using a partial discretization method;

[0075] Discretize the above space domain into finite units, and use the node temperature t kApproximate interpolation calculations are performed, so that the temperature t1 of any point in the unit can be obtained, so the calculated node temperature is related to time, and the interpolation function is related to the spatial domain;

[0076]

[0077] in, is the approximate temperature, n is the number of nodes on each element, s k (x, y, z) is the interpolation function, (x, y, z) is the coordinate of the point in the unit, t k (τ) is the node temperature, τ is the time;

[0078] The error function with t1 is expressed as:

[0079]

[0080] In the formula, R Ω is the constant temperature boundary condition error, is the boundary condition error in the grinding zone, is the boundary condition error outside the grinding zone, c is the specific heat capacity, λ x ,λ y ,λ z represents the thermal conductivity along the x, y, and z directions, n x , n y , n z represents the direction cosine of the normal outside the boundary, and q is the heat flux density;

[0081] Let the weighted integral of the error function be zero:

[0082]

[0083] In the formula, Ω is the solution domain, Γ a is the inner boundary of the grinding zone, Γ b is the outer boundary of the grinding zone, S1, S2 and S3 are expressed as weight functions;

[0084] Let weight function S1 = -S2 = -S3 = N j , after integral calculation, we get n node temperatures t k The matrix equation is expressed as:

[0085]

[0086] Where [C] is the specific heat capacity matrix, is the temperature-time derivative matrix, [K] is the heat conduction matrix, {T} is the node temperature matrix, and {Q1} is the temperature load matrix;

[0087] Therefore, the temperature function related to the time and space domains is discretized in the space domain and converted into the problem of solving the differential equation of n node temperatures t = (τ) under certain initial values ​​and boundary conditions.

[0088] A multi-scale finite element simulation method for the temperature field of robot belt grinding, such as Figure 2 As shown, the following steps are included:

[0089] Step 1: Establish a macroscopic abrasive belt machine workpiece contact finite element model and use the macroscopic abrasive belt machine workpiece contact finite element model to simulate the grinding process, and obtain the normal force curve of a single abrasive grain on the abrasive belt and the length of the contact area between the abrasive belt and the workpiece during the grinding process; the contact area is the area where the groove exists after the abrasive belt is polished; the contact area length is the length of the groove on the workpiece surface after polishing;

[0090] The establishment process of the macroscopic sanding machine workpiece contact finite element model is as follows:

[0091] Firstly, a three-dimensional model of the rubber contact wheel, a three-dimensional model of the abrasive belt and a three-dimensional model of the workpiece are established, and the material parameters, analysis steps, action relationships and loads of the three-dimensional model of the rubber contact wheel, the three-dimensional model of the abrasive belt and the three-dimensional model of the workpiece are set. Then, the three-dimensional model of the rubber contact wheel, the three-dimensional model of the abrasive belt and the three-dimensional model of the workpiece are meshed respectively and divided into a plurality of units, each unit including a plurality of nodes, so as to obtain a macroscopic abrasive belt machine workpiece contact finite element model.

[0092] The rubber contact wheel comprises a rubber layer and a metal wheel, and the abrasive belt comprises a substrate and abrasive grains;

[0093] The material parameters include conductivity, density, elastic modulus, Poisson's ratio, expansion coefficient, inelastic heat share, specific heat capacity, constitutive model and damage failure model;

[0094] The constitutive model adopts the Johnson-Cook model, in which the equivalent stress is a function of strain, strain rate and temperature:

[0095]

[0096] in, is the flow stress, A is the initial yield stress, B is the strain hardening parameter of the material, C is the strain rate sensitivity coefficient, r is the hardening index of the material, is the equivalent plastic strain, is the equivalent plastic strain rate, is the reference strain rate of the material, T me is the melting point temperature of the material, m is the thermal softening index of the material; T t is the material reference temperature, T0 is the temperature;

[0097] The material fracture criterion takes into account strain, strain rate, temperature and stress. The critical fracture value of each unit, that is, the damage failure model, is shown in formulas (19) and (20):

[0098]

[0099] Among them, E is the standard parameter for measuring fracture in the Johnson-Cook model, is the incremental equivalent strain, is the failure strain;

[0100]

[0101] Where D1~D5 are the material JC failure parameters, σ m is the pressure, σ' is the equivalent stress;

[0102] Based on the characteristics of robot belt grinding and polishing, a macroscopic belt grinding machine workpiece contact finite element model was established in Abaqus software. Figure 3 (a). In this embodiment, the outer circle radius of the rubber layer is 9mm, the inner circle radius is 5mm, the metal wheel radius is 5mm, and the height is 6mm. This invention uses an ultra-depth microscope to detect the surface of the sanding belt. According to the observation results, the shape of the abrasive particles is simplified to a pyramid. The abrasive particles are evenly distributed on the substrate. The sanding belt used in the present invention is a pyramid abrasive aluminum oxide sanding belt produced by VSM. After observation under an ultra-depth microscope, the sanding belt is simplified to improve calculation efficiency. Some abrasive particles are arranged on the substrate. After pre-grinding, the abrasive particles change from a pyramid to a quadrangular pyramid with a bottom surface of 0.5×0.5mm 2 The workpiece is made of high-temperature alloy material, specifically a cuboid of 6mm×4mm×1mm.

[0103] The parameter setting in this embodiment is specifically as follows: setting the analysis step, action relationship, load, and meshing in sequence in the ABAQUS finite element software.

[0104] First, create an analysis step, and select the dynamic display analysis step as the analysis step type. Create field output, select the stress component and invariant in the strain force, the volume average stress component and invariant, select the plastic strain component in the strain, the volume average plastic strain component, the equivalent plastic strain, and the volume average equivalent plastic strain, select the translation and rotation, translation and rotation velocity, and translation and rotation acceleration in the displacement / velocity / acceleration, select the reaction force and torque in the action / reaction force, and select the state in the state / field / user / time. Create history output, select the set where the single abrasive particle is located in the scope, select the translation and rotation in the displacement / velocity / acceleration as the output variable, and select the reaction force and torque in the action / reaction force.

[0105] Set the interaction relationship, select general contact type, the first surface is a simplified 27-row abrasive surface, and the second surface is the workpiece surface.

[0106] Set the load. First, set the boundary condition "symmetric / antisymmetric / completely fixed" on the bottom of the workpiece to be completely fixed. Secondly, set the boundary condition "symmetric / antisymmetric / completely fixed" on a single abrasive grain. Finally, set a reference point at the center of the contact wheel. Use this reference point as the selection area in the last step boundary condition "speed / angular velocity" and set the speed V1 = 2. Set the predefined field and set the rotation speed of the abrasive belt and contact wheel to 1111.11 rad / s.

[0107] Meshing,using the mesh control method of C3D8R, the substrate, abrasive, workpiece and rubber contact wheel are refined.,Since the elastic modulus of the metal wheel is much larger than that of the rubber, the metal wheel is regarded as an analytical rigid body and does not require a mesh.

[0108] In the modeling process, the contact between the abrasive grains on the abrasive belt and the workpiece, as well as the deformation of the contact wheel and the workpiece, are fully considered. The simulation results are shown in Figure 2. Figure 4 As shown, Figure 4 (a) shows the formation of polishing grooves and the removal of material on the workpiece surface. The specific operation is to use the query function of ABAQUS to actively measure the length of the surface groove after polishing, which is the length of the contact area. In addition, through this model, the normal force curve of a single abrasive particle can also be output, such as Figure 4 (b) The specific operation is to select "Create XY Data" in the post-processing process to output the process variables. The specific variable is the set where the single abrasive grain is located. Select this set and output the normal force curve. Analysis shows that the normal force reaches the maximum value in the middle of the polishing area and presents an approximate parabolic distribution feature. From this simulation process, the normal force curve of a single abrasive grain on the abrasive belt and the length of the contact area between the abrasive belt and the workpiece during the grinding process can be extracted.

[0109] Step 2: According to the normal force curve of a single abrasive grain and the length of the contact area between the abrasive belt and the workpiece during the grinding process, a microscopic force-heat coupling cutting zone finite element model is established, and the microscopic grinding process and heat transfer process are simulated using the microscopic force-heat coupling cutting zone finite element model to obtain the regional length of the heat-affected zone and the temperature distribution fitting curve of the grinding plane; the heat-affected zone is an area on the workpiece where there is a temperature difference with the ambient temperature;

[0110] The method for establishing the microscopic force-heat coupling cutting zone finite element model is as follows: extracting a section of abrasive belt and a workpiece within a set range in the macroscopic abrasive belt machine workpiece contact finite element model; since the arc length of the substrate of the abrasive belt in the contact area is much smaller than the radius of the rubber contact wheel, the substrate of the contact area is simplified from a section of an arc to a straight line of equal length; in addition, the sand grains are still closely arranged on the substrate; and the workpiece is divided into a grinding area and a heat transfer area according to the length of the contact area, wherein the length of the grinding area comes from the length of the contact area in step 1, and the rest is the heat transfer area; and the material parameters, analysis steps, action relationships and loads of the abrasive belt and the workpiece are set, and the mesh is re-divided to obtain the microscopic force-heat coupling cutting zone finite element model; the normal force curve of the single abrasive grain is used as the amplitude of the load of the microscopic force-heat coupling cutting zone finite element model, and is loaded onto each abrasive grain according to time; the set range is the range composed of the contact area and the set distance areas on both sides thereof, and the length of the set range is greater than the length of the contact area;

[0111] like Figure 3 As shown in (b), the grinding and polishing operation is completed by the coordinated movement of the abrasive belt and the workpiece. The contact area length and the normal force curve of a single abrasive grain can be obtained from step 1, based on which a finite element model of the microscopic force-heat coupling cutting zone can be established. The model consists of an abrasive belt and a workpiece. The abrasive belt is simplified into a row of abrasive grains and the abrasive belt matrix along the circumference of the contact wheel. The length of the simplified abrasive belt needs to be greater than the length of the contact area. The workpiece is the same as in step 1. The workpiece is further divided into a grinding area and a heat transfer area according to the length of the contact area. The length of the grinding area comes from the length of the contact area in step 1, and the rest is the heat transfer area.

[0112] Set parameters: Set the analysis step, action relationship, load, and mesh in ABAQUS finite element software.

[0113] First, create an analysis step, and select the "Dynamic, Temperature-Displacement, Display" analysis step as the analysis step type. Create field output, select stress components and invariants in strain force, select plastic strain components and logarithmic strain components in strain, select translation and rotation in displacement / velocity / acceleration, select reaction force and torque in action / reaction force, select contact stress in contact, select node temperature, unit temperature and heat flux vector in thermal, and select state in state / field / user / time. Create history output, set the scope to the point set where the starting point of the grinding area is located, and select node temperature and unit temperature in thermal as the output variable.

[0114] To set the interaction relationship, first select the surface-to-surface contact, the first surface is the bottom surface of each abrasive grain, the second surface is the workpiece surface, and then set the surface heat exchange conditions for the abrasive belt and the workpiece.

[0115] Set the load. First, use the normal force curve of a single abrasive in step 1 as the amplitude of the load and load it to each abrasive in time. Secondly, create a "displacement / rotation angle" boundary condition for the set of all abrasives, fix its U1 and UR3 coordinate systems, and set the boundary condition "symmetric / antisymmetric / completely fixed" for the substrate. Finally, set the "speed / angular velocity" boundary condition for the workpiece, set the speed V1 to -10002, and then set the predefined field to give the sanding belt and abrasive an ambient temperature of 20 degrees Celsius.

[0116] For meshing, CRE4RT mesh was used. Since the model is relatively simple, automatic meshing was adopted. Attention was paid to further subdividing the mesh of the grinding area to ensure the accuracy of the results.

[0117] The simulation results are as follows Figure 5 As shown in (a), a point on the workpiece surface is in cyclic contact with a large number of abrasive particles. The temperature rise of the workpiece surface is the cumulative effect of the tiny cutting work of a large number of abrasive particles. Therefore, the temperature of the workpiece surface point rises slowly. When the work done by the abrasive particles is converted per unit time and the heat transferred to the workpiece is equal to the heat dissipated from the exposed part of the cutting area to the environment, the temperature in the cutting area tends to balance.

[0118] Depend on Figure 5 As shown in (b), belt grinding generates grinding heat. As the grinding time increases, the temperature of the cutting area tends to be balanced, and the heat will continue to diffuse to the surrounding area. Figure 5 (b) The length of the heat affected zone (i.e., the area with the maximum temperature change) was then extracted for the nodes on the workpiece surface, and the discrete data points were imported into Matlab for fitting. From the temperature data distribution, it can be seen that the Gaussian curve fitting effect is the best. The fitting curve is as follows Figure 5 As shown in (c), it can be concluded that the surface temperature distribution of the workpiece polished by multi-abrasive micro-belt grinding is a Gaussian distribution.

[0119] Step 3: A macroscopic temperature field finite element model is established based on the obtained area length of the heat-affected zone and the temperature distribution fitting curve of the grinding plane, and the heat source transfer process is simulated using the macroscopic temperature field finite element model to obtain the temperature field distribution of the workpiece and the temperature field distribution curve of the heat-affected zone depth;

[0120] The method for constructing the macroscopic temperature field finite element model is as follows: establishing a three-dimensional model of the workpiece and a three-dimensional model of a movable thin plate as an equivalent heat source, then setting material parameters, analysis steps, action relationships and loads of the three-dimensional model of the workpiece and the three-dimensional model of the movable thin plate, and finally performing meshing to obtain the macroscopic temperature field finite element model;

[0121] The length of the movable thin plate is the length of the heat-affected zone; the temperature distribution fitting curve of the grinding plane is applied to the three-dimensional model of the movable thin plate;

[0122] The heat-affected zone depth temperature field distribution curve is a temperature curve formed by the temperature along the depth direction starting from the contact point between the movable thin plate and the workpiece;

[0123] Establishment of finite element model of macroscopic temperature field: Figure 3 As shown in (c), the model consists of an equivalent movable thin plate and a workpiece. The length of the equivalent movable thin plate is the length of the heat affected zone, and the size of the workpiece is the same as that in step 1.

[0124] Set parameters: Set the analysis step, action relationship, load, and mesh in ABAQUS finite element software in sequence. First create an analysis step, and select the "dynamic, temperature-displacement, display" analysis step as the analysis step type. Create field output, select stress components and invariants in strain force, volume average stress components and invariants, select plastic strain components and logarithmic strain components in strain, select translation and rotation in displacement / velocity / acceleration, select reaction force and torque in action / reaction force, select contact stress in contact, select node temperature, unit temperature, heat flow vector and reaction flux in thermal, and select state in state / field / user / time. Create process output, set the scope to the entire model, and select total energy in energy as the output variable.

[0125] To set the interaction relationship, firstly, select the surface-to-surface contact, the first surface is the bottom surface of the movable thin plate, the second surface is the workpiece surface, and then set the surface heat exchange conditions for the workpiece.

[0126] Set the load. First, apply the temperature distribution fitting curve of the grinding plane in step 2 as the analytical field in the load temperature boundary condition to the surface of the movable thin plate. Secondly, set the boundary condition "symmetric / antisymmetric / completely fixed" for the movable thin plate to be completely fixed. Finally, set the boundary condition "speed / angular velocity" for the workpiece and set the speed V1 to -2. Then set the predefined field and give the sanding belt and abrasive grains an ambient temperature of 20 degrees Celsius.

[0127] The CRE4RT unit is used to mesh the movable thin plate and the workpiece. Due to the simplicity of the model, the mesh is automatically divided.

[0128] The simulation results are as follows Figure 6 As shown, the temperature field distribution of the workpiece with equivalent heat source transmitted to it can be obtained as follows: Figure 6 As shown in (a), further, the temperature data of the longitudinal nodes of the workpiece are extracted and spline curve fitting is performed to obtain the temperature field distribution curve of the heat affected zone depth as shown in Figure 6 (b) as shown.

Claims

1. A multi-scale finite element simulation method for temperature field of robot belt grinding, characterized in that: The steps include: Step 1: Establish a macroscopic abrasive belt machine workpiece contact finite element model and use the macroscopic abrasive belt machine workpiece contact finite element model to simulate the grinding process, and obtain the normal force curve of a single abrasive grain on the abrasive belt and the contact area length between the abrasive belt and the workpiece during the grinding process; Step 2: According to the normal force curve of a single abrasive grain and the length of the contact area between the abrasive belt and the workpiece during the grinding process obtained in step 1, a finite element model of a microscopic force-heat coupling cutting zone is established, and the microscopic grinding process and heat transfer process are simulated using the finite element model of the microscopic force-heat coupling cutting zone to obtain the length of the heat-affected zone and the temperature distribution fitting curve of the grinding plane; Step 3: According to the zone length of the heat affected zone obtained in step 2 and the temperature distribution fitting curve of the grinding plane, a macroscopic temperature field finite element model is established, and the heat source transfer process is simulated using the macroscopic temperature field finite element model to obtain the temperature field distribution of the workpiece and the temperature field distribution curve of the depth of the heat affected zone.

2. The multi-scale finite element simulation method for the temperature field of robot belt grinding according to claim 1 is characterized in that: The contact area in step 1 is the area where the groove exists after the abrasive belt grinding and polishing; the length of the contact area is the length of the groove on the surface of the workpiece after grinding and polishing.

3. The multi-scale finite element simulation method for temperature field of robot belt grinding according to claim 1 is characterized in that: The establishment process of the macroscopic sanding machine workpiece contact finite element model is as follows: Firstly, the three-dimensional models of the rubber contact wheel, the three-dimensional model of the abrasive belt and the three-dimensional model of the workpiece are established, and the material parameters, analysis steps, action relationships and loads of the three-dimensional models of the rubber contact wheel, the three-dimensional model of the abrasive belt and the workpiece are set. Then, the three-dimensional models of the rubber contact wheel, the three-dimensional model of the abrasive belt and the workpiece are meshed respectively into several units, each unit including several nodes, and a macroscopic abrasive belt machine workpiece contact finite element model is obtained.

4. The multi-scale finite element simulation method for the temperature field of robot belt grinding according to claim 3 is characterized in that: The rubber contact wheel comprises a rubber layer and a metal wheel, and the abrasive belt comprises a substrate and abrasive grains; The material parameters include conductivity, density, elastic modulus, Poisson's ratio, expansion coefficient, inelastic heat share, specific heat capacity, constitutive model and damage failure model.

5. The multi-scale finite element simulation method for temperature field of robot belt grinding according to claim 1, characterized in that: The heat affected zone described in step 2 is a region on the workpiece where there is a temperature difference with the ambient temperature.

6. The multi-scale finite element simulation method for temperature field of robot belt grinding according to claim 4, characterized in that: The method for establishing the finite element model of the microscopic force-heat coupling cutting zone is as follows: extract a section of the abrasive belt and the workpiece within a set range in the macroscopic abrasive belt machine workpiece contact finite element model, simplify the substrate of the contact area from a section of an arc to a straight line of equal length, the sand grains are still arranged on the substrate, and the workpiece is divided into a grinding area and a heat transfer area according to the length of the contact area, the length of the grinding area is the length of the contact area, and the rest is the heat transfer area, and the material parameters, analysis steps, action relationships and loads of the abrasive belt and the workpiece are set, and the mesh is re-divided to obtain the finite element model of the microscopic force-heat coupling cutting zone.

7. The multi-scale finite element simulation method for temperature field of robot belt grinding according to claim 1, characterized in that: The normal force curve of the single abrasive particle is used as the amplitude of the load of the finite element model of the microscopic force-heat coupling cutting zone, and is loaded onto each abrasive particle according to time.

8. The multi-scale finite element simulation method for temperature field of robot belt grinding according to claim 6, characterized in that: The set range is a range consisting of the contact area and set distance areas on both sides thereof, and the length of the set range is greater than the length of the contact area.

9. The multi-scale finite element simulation method for temperature field of robot belt grinding according to claim 1, characterized in that: The method for constructing the macroscopic temperature field finite element model is as follows: establishing a three-dimensional model of the workpiece and a three-dimensional model of a movable thin plate as an equivalent heat source, then setting the material parameters, analysis steps, action relationships and loads of the three-dimensional model of the workpiece and the three-dimensional model of the movable thin plate, and finally performing meshing to obtain the macroscopic temperature field finite element model.

10. The multi-scale finite element simulation method for temperature field of robot belt grinding according to claim 9, characterized in that: The length of the movable thin plate is the length of the heat-affected zone; the temperature distribution fitting curve of the grinding plane is applied to the three-dimensional model of the movable thin plate; The heat-affected zone depth temperature field distribution curve is a temperature curve formed by the temperature along the depth direction starting from the contact point between the movable thin plate and the workpiece.

Images