CN112926152B - Digital twin-driven thin-wall part clamping force precise control and optimization method - Google Patents
Digital twin-driven thin-wall part clamping force precise control and optimization method Download PDFInfo
- Publication number
- CN112926152B CN112926152B CN202110186148.9A CN202110186148A CN112926152B CN 112926152 B CN112926152 B CN 112926152B CN 202110186148 A CN202110186148 A CN 202110186148A CN 112926152 B CN112926152 B CN 112926152B
- Authority
- CN
- China
- Prior art keywords
- thin
- clamping
- wall part
- clamp
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The invention provides a digital twin-driven precise control and optimization method for clamping force of a thin-wall part, which comprises the steps of firstly obtaining clamping force and clamping deformation data of a manufacturing main body of the thin-wall part, and configuring various sensors for the thin-wall part, a clamp and a machine tool by adopting the Internet of things technology; on the basis of acquiring data, establishing a digital twin virtual model such as a thin-wall part geometric model, a finite element deformation simulation model, a thin-wall part clamping deformation prediction and the like, and mapping and simulating the thin-wall part clamping process in real time; and finally, establishing a thin-wall part clamping force optimization model, optimizing the clamping force of the thin-wall part, forming a thin-wall part clamping force selection rule, and improving the machining precision of the thin-wall part. The invention can solve the problem that the clamping force range meeting the clamping deformation error is difficult to determine in the process of clamping the thin-wall part.
Description
Technical Field
The invention relates to the technical field of clamping of thin-wall parts, in particular to a digital twin-driven precise control and optimization method for clamping force of the thin-wall parts.
Background
When the wall thickness or the ratio of the plate thickness to the contour dimension of the part is less than 1:20, it is called a thin-walled member. The thin-wall part has the advantages of high specific strength, relatively light weight and the like, and is widely applied to the field of aerospace. However, the thin-walled member is easily deformed during processing due to the thin wall thickness and generally weak rigidity. The clamping quality is one of the main factors influencing the deformation of the thin-wall part in the machining process, and the workpiece performance is directly influenced. However, the clamping force in the existing clamping process is determined by experience of an operator, so that the defects of high randomness, non-uniform standards, difficult optimization and the like exist, the clamping force is difficult to accurately control, and the fluctuation of the quality of the thin-wall part is aggravated. Therefore, how to accurately control the clamping force has important theoretical value for improving the processing quality of the thin-wall part.
The Chinese patent (201710964936X) discloses a method for predicting residual stress deformation of a thin-wall part based on clamping force monitoring, which comprises the steps of firstly predicting the residual stress deformation trend of the thin-wall part by a finite element simulation method, and adding clamping force sensing points in a large deformation area; then designing a sensing clamp, and monitoring the change of clamping force in the machining process at a sensing point through a pressure sensor; and finally, a finite element model of a static base of the clamping system is established, and a counter force of a clamping force change value is applied to a sensing point to obtain the residual stress deformation of the part, so that the prediction of the residual stress deformation of the thin-wall part is realized. The method has an effect in the aspect of perception and prediction of residual stress deformation of the thin-wall part, but because the mapping relation between clamping force and actual clamping deformation is not considered, the clamping force is not optimally controlled, so that the clamping force range meeting the clamping deformation error cannot be obtained, the clamping force is difficult to accurately control, and the actual use requirement cannot be met.
Disclosure of Invention
The invention provides a digital twin-driven precise control and optimization method for clamping force of a thin-wall part, which aims to solve the problem that the clamping force range meeting clamping deformation errors is difficult to determine in the process of clamping the thin-wall part.
The basic idea of the invention is:
firstly, obtaining clamping force and clamping deformation data of a thin-wall part manufacturing main body, wherein the implementation means adopts the modes of configuring various sensors for the thin-wall part, a clamp and a machine tool by adopting the technology of the Internet of things; on the basis of acquiring data, establishing a digital twin virtual model such as a thin-wall part geometric model, a finite element deformation simulation model, a thin-wall part clamping deformation prediction and the like, and mapping and simulating the thin-wall part clamping process in real time; and finally, establishing a thin-wall part clamping force optimization model, optimizing the clamping force of the thin-wall part, forming a thin-wall part clamping force selection rule, and improving the machining precision of the thin-wall part.
The technical scheme of the invention is as follows:
the digital twin-driven thin-wall part clamping force accurate control and optimization method comprises the following steps:
step 1: acquiring off-line data of a thin-wall piece and a clamp before clamping and on-line data in the process of clamping the thin-wall piece; the offline data comprises geometric parameters and material properties; the online data comprise input torque applied to the clamp, deformation displacement data of monitoring point positions arranged on the thin-wall piece and clamping environment data;
step 2: based on a digital twin technology, establishing a physical space model and a virtual space model of the thin-wall part-clamp; the virtual space model comprises a geometric information model, a finite element simulation model and a thin-wall part clamping deformation prediction model
Step 3: based on the step 1 and the step 2, executing the precise control and optimization process of the clamping force of the thin-wall piece based on digital twinning:
step 3.1: transmitting the clamping process data of the real thin-walled part in the physical space to a physical space model, calling real-time information of the physical space model by the virtual space model, and displaying the clamping process picture in real time through man-machine interaction equipment;
step 3.2: performing simulation in a virtual space model of the thin-wall part-clamp, and further predicting the deformation of the thin-wall part according to a thin-wall part clamping deformation prediction model:
step 3.3: after the clamping movement of the thin-wall part is completed, the whole flow data are stored into a thin-wall part-clamp physical space model, and on the basis of a thin-wall part clamping deformation prediction model, a group of optimal positioning points and clamping force parameters are obtained through an optimization algorithm so that the clamping deformation is minimum, and therefore the initial clamping force of the thin-wall part is determined.
Further, in step 1, the technology of the internet of things is applied to the clamping process of the thin-wall piece, torque wrenches and sensor equipment are correspondingly configured on the thin-wall piece and the clamp, offline data of the thin-wall piece and the clamp before clamping are obtained, monitoring points are set, and online data in the clamping process of the thin-wall piece are obtained.
Further, in step 1, the method specifically includes the following steps:
step 1.1: before clamping, detecting a thin-wall piece to be clamped and a clamp to obtain geometric parameters and material properties of the thin-wall piece and the clamp;
step 1.2: the method comprises the steps of acquiring input torque applied to a clamp by adopting a torque wrench, measuring the clamping force of the contact position of a thin-wall piece and the clamp by adopting a pressure sensor, setting an acquisition point on the thin-wall piece, arranging a ranging sensing device to acquire deformation displacement data of a monitoring point of the thin-wall piece, and configuring a temperature sensor and a humidity sensor in a clamping environment to acquire clamping environment data.
Further, in step 2, the method specifically includes the following steps:
step 2.1: establishing a physical space model of the thin-wall part-clamp, and converting a thin-wall part clamping system from a physical entity into a Web-identified data model for tracking and calling the thin-wall part-clamp information;
step 2.2: and constructing a thin-wall part-clamp virtual space model, wherein the model comprises a geometric information model, a finite element simulation model and a thin-wall part clamping deformation prediction model, and mapping the clamping deformation state of the thin-wall part in real time.
Further, in step 2.2, the method specifically includes the following steps:
step 2.2.1: constructing a geometric model of the thin-wall part and the clamp by adopting three-dimensional software according to the data acquired in the step 1;
step 2.2.2: constructing a finite element simulation model of the deformation of the thin-wall piece by adopting finite element software;
step 2.2.3: based on a deep neural network, a thin-wall piece clamping deformation prediction model is established, a nonlinear mapping relation between clamping force and clamping deformation is established, deformation data simulated by a finite element simulation model is used as a training sample set of the thin-wall piece clamping deformation prediction model, and the thin-wall piece clamping deformation prediction model is trained.
Further, in step 2.2.2, the method specifically includes the following steps:
a. geometry model importation: importing the thin-wall piece and the clamp geometric model established in the step 2.2.1 into finite element software, wherein the clamp is set to analyze a rigid body, and the thin-wall piece is set to be an isotropic elastomer;
b. setting material properties: reading basic information of the thin-wall part and the clamp object of the physical space model established in the step 2.1, and setting material properties of the thin-wall part and the clamp;
c. and (3) assembling: assembling the thin-wall part and the clamp in finite element software according to an actual clamping mode, setting a contact position, and defining the thin-wall part and the clamp by adopting 'surface-to-surface contact';
d. boundary conditions and loads: completely restraining six degrees of freedom of the bottom surface of the thin-wall part, and applying load on the thin-wall part and the contact position of the clamp;
e. dividing grids: the grid control attribute selects hexahedron, agate, central axis algorithm and adopts C3D20R unit to divide grids.
Further, in step 2.2.3, the method specifically includes the following steps:
step 2.2.3.1: writing a script file according to the finite element simulation model established in the step 2.2.2 by utilizing the parameterization analysis function of finite element software, and carrying out finite element calculation by changing the contact position coordinates and clamping force size parameters to obtain thin-wall piece deformation data corresponding to the measuring points;
step 2.2.3.2: determining input and output of a thin-wall piece clamping deformation prediction model; the input unit of the thin-wall part clamping deformation prediction model is a coordinate parameter of a locating point and a clamping load applied to the thin-wall part, and the coordinate parameter is expressed as x= { x k }(1≤k≤K)，x k ＝{s k ,F k }，s k Representing the coordinates of the anchor point in the kth sample, F k Representing the clamp load in the kth sample; the thin-wall piece clamping deformation prediction model output unit is clamping deformation data of a thin-wall piece measuring point, and is expressed as y= { y k }(1≤k≤K)；
Step 2.2.3.3: establishing a deep neural network with L layers, wherein the node numbers of an input layer, an i-th hidden layer and an output layer of the deep neural network are respectively r and L i C, selecting a tanh function as an activation function, wherein the tanh function is expressed as follows:
the error function is a square loss function;
step 2.2.3.4: and carrying out gradient calculation by adopting a back propagation algorithm, correcting the weight parameters and the bias parameters of the deep neural network, and ending iteration when the iteration times or errors are smaller than or equal to a preset value to obtain a trained thin-wall part clamping deformation prediction model.
Further, in step 2.2.3, the sample is normalized: assuming that the mean value of the input feature over the entire sample dataset is u and the standard deviation is σ, the normalization operation is to subtract u from each value of the input feature and then divide by the standard deviation.
Further, the specific steps of the step 3.2 are as follows:
step 3.2.1: transmitting actions and system control instructions of the thin-wall part clamping process to a thin-wall part-clamp virtual space model in real time, and simulating clamping deformation of the thin-wall part by the thin-wall part-clamp virtual space model, so as to realize three-dimensional visual simulation of clamping behaviors of the thin-wall part;
step 3.2.2: predicting the clamping deformation of the thin-wall part through a thin-wall part clamping deformation prediction model, wherein the method comprises the following specific steps of:
a. according to the formula f=km n Input moment M collected by moment spanner n And converting the thin-walled workpiece into a clamping load F applied to the surface of the thin-walled workpiece, wherein k is a conversion coefficient, and determining by the following experiment: when the torque wrench inputs torque M n Obtaining a clamping force value F ' through the pressure sensor arranged in the step 1.2, so as to obtain a quotient value k ', carrying out multiple tests to obtain a plurality of quotient values k ', and taking an average value as a conversion coefficient k;
b. inputting the converted clamping load and the positioning position coordinates into a thin-wall piece clamping deformation prediction model to obtain a clamping deformation prediction value; and comparing the real-time clamping deformation data with historical data under the running condition, and updating the thin-wall part clamping deformation prediction model when the error of the real-time clamping deformation data exceeds the allowable value, wherein the updating step is to firstly add the real-time clamping deformation data into a training set of the thin-wall part clamping deformation prediction model, and then train the thin-wall part clamping deformation prediction model to obtain an updated prediction model.
Further, the optimization algorithm in step 3.3 adopts a genetic algorithm, and the objective function of the genetic algorithm is defined as:
min f(NET,s,F)
wherein F represents a thin-wall part clamping deformation prediction model, NET represents a neural network established in the prediction model, S represents positioning point coordinates, S represents a point set where a positioning point is located, F represents the magnitude of clamping force, and F min And F max Representing the minimum and maximum values of clamping force;
randomly generating an initial population comprising p individuals; decoding individuals one by one, eliminating individuals which do not meet constraint conditions in the optimization model, and predicting clamping deformation by the individuals which meet the constraint conditions through a deep neural network;
individual fitness is defined as
In the above, U i (1.ltoreq.i.ltoreq.p) is the number of individuals in the population, and Δ is a value predetermined according to the objective function value of each generation.
Advantageous effects
Compared with the prior art, the invention has the advantages that: the technology of the Internet of things is applied to a manufacturing workshop, real-time data of clamping the thin-wall piece is collected, and active sensing of the clamping process of the thin-wall piece is realized; a digital twin technology is introduced to construct a digital twin model of the thin-wall part-clamp, and real-time simulation is carried out, so that a deformation state in the clamping of the thin-wall part is obtained, and physical data is expanded in dimensions; the deep neural network can well model the conditions that the coupling mechanism between factors is undefined and the exact relation between input and output is difficult to determine through extremely strong self-adaption and self-learning capability, so that the deformation of the thin-wall part can be accurately and rapidly predicted, the clamping force with the minimum clamping deformation can be obtained by applying a genetic algorithm, and the clamping force selection rule of the thin-wall part is formed, thereby improving the machining precision of the thin-wall part.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the invention will become apparent and may be better understood from the following description of embodiments taken in conjunction with the accompanying drawings in which:
FIG. 1 is a technical roadmap of the invention;
FIG. 2 is a workflow diagram of the method of the present invention;
FIG. 3 is a geometric model of a clamp three-jaw chuck;
FIG. 4 is a geometric model of a thin-walled member;
fig. 5 is a physical-virtual space interaction timing diagram of the method of the present invention.
Detailed Description
The following detailed description of embodiments of the invention is exemplary and intended to be illustrative of the invention and not to be construed as limiting the invention.
The invention designs a digital twin-driven thin-wall part clamping force accurate control optimization method, which aims to establish a thin-wall part geometric model, a finite element deformation simulation model, a thin-wall part clamping deformation prediction model, a clamping force optimization model and other digital twin-virtual models on the basis of acquiring clamping force and clamping deformation data of a thin-wall part manufacturing main body, map and simulate the thin-wall part clamping process in real time, optimize the thin-wall part clamping force and finally improve the machining precision of the thin-wall part. The specific implementation steps are as follows:
step 1: the technology of the Internet of things is applied to the clamping process of the thin-wall part, equipment such as a digital torque wrench, a pressure sensor and the like is configured on physical main bodies such as the thin-wall part and the clamp, off-line data such as geometric parameters, material properties and the like of the thin-wall part and the clamp before clamping are obtained, and proper acquisition points are set, so that on-line data in the clamping process of the thin-wall part are obtained, and the method specifically comprises the following steps:
step 1.1: detecting a thin-wall part to be clamped and a clamp to obtain data such as geometric parameters, material properties and the like of the thin-wall part;
step 1.2: the method comprises the steps of acquiring torque applied to a clamp by adopting a digital display torque wrench, measuring the clamping force of the contact position of a thin-wall piece and the clamp by adopting a pressure sensor, setting an acquisition point on the thin-wall piece, arranging a laser range finder range finding sensing device to acquire deformation displacement data of a monitoring point of the thin-wall piece, and finally configuring temperature and humidity sensors in a clamping environment to acquire clamping processing environment data.
Step 2: based on the digital twin technology, a physical space model and a virtual space model of the thin-wall part-clamp are established, and the method specifically comprises the following steps:
step 2.1: establishing a physical space model of the thin-wall part-clamp, converting a clamping system formed by the thin-wall part-clamp from a physical entity into a Web-identified data model, and tracking and calling the information of the thin-wall part-clamp; the physical space model of the thin-walled part-clamp is represented by a five-dimensional tuple:
TCGPM＝{TCBI,GBI,SSL,EI,TCGSI}
wherein TCBI (Thin-walled component Basic Information) represents Thin-walled workpiece object basic information, GBI (Gig Basic Information) represents fixture object basic information, SSL (Smart Sensor List) represents a sensor list configured in a Thin-walled workpiece clamping system, EI (Environmental Information) represents environmental information, and TCGSI (Thin-walled component and Gig Status Information) represents real-time status information of a Thin-walled workpiece-fixture.
Step 2.2: and constructing a thin-wall part-clamp virtual space model, wherein the model comprises a geometric model, a finite element simulation model and a thin-wall part clamping deformation prediction model, and mapping the clamping deformation state of the thin-wall part in real time.
Step 2.2.1: constructing a geometric model of the thin-wall part and the clamp by adopting SolidWorks three-dimensional software according to the data acquired in the step 1;
step 2.2.2: and constructing a finite element simulation model of the deformation of the thin-wall part by adopting Abaqus finite element software and adopting a finite element statics cutting simulation technology. The method specifically comprises the following steps:
f. geometry model importation: importing the thin-wall piece and the clamp geometric model established in the step 2.2.1 into Abaqus software, setting the clamp to analyze a rigid body, and setting the thin-wall piece to be an isotropic elastomer;
g. setting material properties: reading basic information of the thin-wall part and the clamp object of the physical space model established in the step 2.1, and setting material properties such as elastic modulus, poisson ratio, density and the like of the thin-wall part and the clamp;
h. and (3) assembling: assembling the thin-wall part and the clamp in finite element software according to an actual clamping mode, setting a contact position (positioning point), and defining the thin-wall part and the clamp by adopting 'surface-to-surface contact';
i. boundary conditions and loads: six degrees of freedom of the bottom surface of the thin-wall part are completely restrained, no movement occurs in the loading process, and load is applied to the contact position of the thin-wall part and the clamp;
j. dividing grids: the grid control attribute selects hexahedron, agate, central axis algorithm and adopts C3D20R units to divide grids;
step 2.2.3: based on the deep neural network, a thin-wall part clamping deformation prediction model is established to construct a nonlinear mapping relation between clamping force and clamping deformation. The finite element simulation is to ignore the simulation of clamping deformation under other factors except clamping force in the processing environment, namely the simulation of clamping deformation under ideal conditions, and when the actual clamping deformation data is not available at first or the actual clamping data is smaller in scale, the deformation data simulated by the finite element simulation model is used as a training set of the thin-wall part clamping deformation prediction model, and the thin-wall part clamping deformation prediction model is trained. The method specifically comprises the following steps:
step 2.2.3.1: writing a script file according to the finite element simulation model established in the step 2.2.2 by utilizing the parameterization analysis function of the Abaqus finite element software, and automatically performing finite element calculation by changing the coordinates of the contact position (locating point) and the size parameters of clamping force to obtain thin-wall piece deformation data corresponding to the measuring point;
step 2.2.3.2: determining input and output of a thin-wall piece clamping deformation prediction model; the model input unit is the coordinate parameter of the locating point and the clamping load applied on the thin-wall part, and is expressed as x= { x k }(1≤k≤K)，x k ＝{s k ,F k }，s k Representing the coordinates of the anchor point in the kth sample, F k Representing the clamp load in the kth sample; the model output unit is clamping deformation data of a thin-wall part measuring point and is expressed as y= { y k }(1≤k≤K)；
Step 2.2.3.3: in order to ensure network convergence, carrying out standardized processing on an input sample; assuming that the mean value of the input feature over the entire dataset is u and the standard deviation is σ, the normalization operation is to subtract u from each value of the input feature and then divide by the standard deviation σ. After the standardization is completed, dividing the obtained data into two groups, wherein one group is a training sample, the data size of the training sample is 80% of the original data size, and the other group is a test sample, and the data size of the testing sample is 20% of the original data size;
step 2.2.3.4: establishing a deep neural network with L layers, wherein the node numbers of an input layer, an i-th hidden layer and an output layer of the deep neural network are respectively r and L i C, selecting a tanh function as an activation function, wherein the tanh function is expressed as follows:
the error function is a square loss function;
step 2.2.3.5: and carrying out gradient calculation by adopting a back propagation algorithm, correcting the weight parameters and the bias parameters of the deep neural network, and ending iteration when the iteration times or errors are smaller than or equal to a preset value to obtain a trained thin-wall part clamping deformation prediction model.
Step 3: on the basis of the step 1 and the step 2, the precise control and optimization process of the clamping force of the thin-wall piece based on digital twinning is executed, and the method specifically comprises the following steps:
step 3.1: transmitting the clamping process data of the real thin-walled part in the physical space to a physical space model, calling real-time information of the physical space model by a virtual space model, and displaying a clamping process picture in front of a manager in real time through man-machine interaction equipment (such as a computer and the like);
step 3.2: and performing simulation in the thin-wall part-clamp virtual space model, and further predicting the deformation of the thin-wall part according to the prediction model. The method comprises the following specific steps:
step 3.2.1: transmitting actions and system control instructions of the thin-wall part clamping process to a thin-wall part-clamp virtual space model in real time, carrying out simulation of clamping deformation of the thin-wall part by the thin-wall part-clamp virtual space model, and realizing three-dimensional visual simulation of clamping behaviors of the thin-wall part;
step 3.2.2: predicting the clamping deformation of the thin-wall part through a thin-wall part clamping deformation prediction model, wherein the method comprises the following specific steps of:
c. the input torque collected by the digital display torque wrench is converted into the input torque which is applied to the surface of the thin-wall partThe conversion formula of the three-jaw chuck is as follows: f=km n ，M n For input torque, k is the conversion factor. The k value is determined by the following experiment: input torque M of digital display torque wrench n The clamping force value F 'is acquired by the pressure sensor arranged in the step 1.2, so that a quotient value k' is obtained. K is averaged after multiple trials. Because the pressure sensor can not control the clamping force to input, the clamping force can only be measured, and the clamping force measured in a complex processing environment fluctuates greatly, the clamping force value is obtained by adopting the pressure sensor only in an experiment with a single environment, so as to determine the conversion coefficient. In the actual clamping processing process, a stable digital display torque wrench is selected to control the input torque, so that the clamping force is controlled.
d. And inputting the converted clamping load and the positioning position coordinates into a prediction model to obtain a predicted value of clamping deformation. And comparing the model with historical data under the running condition to verify the accuracy and the effectiveness of the thin-wall part clamping deformation prediction model. When the error of the two errors exceeds the allowable value, updating the prediction model, wherein the updating step is to add the real-time clamping deformation data into a training set of the prediction model, and train the prediction model to obtain an updated prediction model so as to ensure that the prediction model can accurately map the thin-wall clamping deformation;
step 3.3: after the clamping movement of the thin-wall part is completed, the whole flow data are stored into a thin-wall part-clamp physical space model, and on the basis of a thin-wall part clamping deformation prediction model, a group of optimal positioning points and clamping force parameters are obtained through a genetic algorithm so that the clamping deformation is minimum, thereby determining the initial clamping force of the thin-wall part. The genetic algorithm objective function is defined as:
min f(NET,s,F)
wherein f represents a thin-wall part clamping deformation prediction model, NET represents a neural network established in the prediction model, S represents positioning point coordinates, and S represents a point set where a positioning point is locatedF represents the clamping force, F min And F max Representing the minimum and maximum clamping force.
An initial population comprising p individuals is randomly generated. And decoding the individuals one by one, wherein the individuals which do not meet the constraint conditions in the optimization model are removed, and the individuals which meet the constraint conditions can predict clamping deformation by the deep neural network.
Individual fitness is defined as
In the above, U i (1.ltoreq.i.ltoreq.p) is the number of individuals in the population, and Δ is a relatively large value predetermined based on the objective function value of each generation.
The research is supported by Beijing aerospace fresh air mechanical equipment limited responsibility company, and fund project (No: D5204200210) and Innovative seed fund (No: CX 2020100) of North-North Industrial university research students.
Although embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives, and variations may be made in the above embodiments by those skilled in the art without departing from the spirit and principles of the invention.
Claims (9)
1. A digital twin-driven thin-wall part clamping force accurate control and optimization method is characterized by comprising the following steps:
step 1: acquiring off-line data of a thin-wall piece and a clamp before clamping and on-line data in the process of clamping the thin-wall piece; the offline data comprises geometric parameters and material properties; the online data comprise input torque applied to the clamp, deformation displacement data of monitoring point positions arranged on the thin-wall piece and clamping environment data;
step 2: based on a digital twin technology, establishing a physical space model and a virtual space model of the thin-wall part-clamp; the virtual space model comprises a geometric information model, a finite element simulation model and a thin-wall part clamping deformation prediction model;
the method comprises the steps of establishing a thin-wall piece clamping deformation prediction model based on a deep neural network, constructing a nonlinear mapping relation between clamping force and clamping deformation, and training the thin-wall piece clamping deformation prediction model by taking deformation data simulated by a finite element simulation model as a training sample set of the thin-wall piece clamping deformation prediction model; the specific process is as follows:
step 2.2.3.1: writing a script file according to the finite element simulation model by utilizing the parameterization analysis function of finite element software, and carrying out finite element calculation by changing the contact position coordinates and clamping force size parameters to obtain thin-wall piece deformation data corresponding to the measuring points;
step 2.2.3.2: determining input and output of a thin-wall piece clamping deformation prediction model; the input unit of the thin-wall part clamping deformation prediction model is a coordinate parameter of a locating point and a clamping load applied to the thin-wall part, and the coordinate parameter is expressed as x= { x k },1≤k≤K，x k ＝{s k ,F k }，s k Representing the coordinates of the anchor point in the kth sample, F k Representing the clamp load in the kth sample; the thin-wall piece clamping deformation prediction model output unit is clamping deformation data of a thin-wall piece measuring point, and is expressed as y= { y k },1≤k≤K；
Step 2.2.3.3: establishing a deep neural network with L layers, wherein the node numbers of an input layer, an i-th hidden layer and an output layer of the deep neural network are respectively r and L i C, selecting a tanh function as an activation function, wherein the tanh function is expressed as follows:
the error function is a square loss function;
step 2.2.3.4: carrying out gradient calculation by adopting a back propagation algorithm, correcting weight parameters and bias parameters of the deep neural network, and ending iteration when the iteration times or errors are smaller than or equal to preset values to obtain a trained thin-wall part clamping deformation prediction model; step 3: based on the step 1 and the step 2, executing the precise control and optimization process of the clamping force of the thin-wall piece based on digital twinning:
step 3.1: transmitting the clamping process data of the real thin-walled part in the physical space to a physical space model, calling real-time information of the physical space model by the virtual space model, and displaying the clamping process picture in real time through man-machine interaction equipment;
step 3.2: performing simulation in a virtual space model of the thin-wall part-clamp, and further predicting the deformation of the thin-wall part according to a thin-wall part clamping deformation prediction model:
step 3.3: after the clamping movement of the thin-wall part is completed, the whole flow data are stored into a thin-wall part-clamp physical space model, and on the basis of a thin-wall part clamping deformation prediction model, a group of optimal positioning points and clamping force parameters are obtained through an optimization algorithm so that the clamping deformation is minimum, and therefore the initial clamping force of the thin-wall part is determined.
2. The method for precisely controlling and optimizing the clamping force of the thin-walled workpiece driven by the digital twin system according to claim 1 is characterized in that in the step 1, the technology of the Internet of things is applied to the clamping process of the thin-walled workpiece, torque wrenches and sensor equipment are correspondingly arranged on the thin-walled workpiece and the clamp, offline data of the thin-walled workpiece and the clamp before clamping are obtained, monitoring points are set, and online data in the clamping process of the thin-walled workpiece are obtained.
3. The method for precisely controlling and optimizing the clamping force of the digital twin-driven thin-wall piece according to claim 2, wherein in the step 1, the method specifically comprises the following steps:
step 1.1: before clamping, detecting a thin-wall piece to be clamped and a clamp to obtain geometric parameters and material properties of the thin-wall piece and the clamp;
step 1.2: the method comprises the steps of acquiring input torque applied to a clamp by adopting a torque wrench, measuring the clamping force of the contact position of a thin-wall piece and the clamp by adopting a pressure sensor, setting an acquisition point on the thin-wall piece, arranging a ranging sensing device to acquire deformation displacement data of a monitoring point of the thin-wall piece, and configuring a temperature sensor and a humidity sensor in a clamping environment to acquire clamping environment data.
4. The method for precisely controlling and optimizing the clamping force of the digital twin-driven thin-wall piece according to claim 3, wherein in the step 2, the method specifically comprises the following steps:
step 2.1: establishing a physical space model of the thin-wall part-clamp, and converting a thin-wall part clamping system from a physical entity into a Web-identified data model for tracking and calling the thin-wall part-clamp information;
step 2.2: and constructing a thin-wall part-clamp virtual space model, wherein the model comprises a geometric information model, a finite element simulation model and a thin-wall part clamping deformation prediction model, and mapping the clamping deformation state of the thin-wall part in real time.
5. The method for precisely controlling and optimizing the clamping force of the digital twin-driven thin-wall piece according to claim 4, wherein in the step 2.2, the method specifically comprises the following steps:
step 2.2.1: constructing a geometric model of the thin-wall part and the clamp by adopting three-dimensional software according to the data acquired in the step 1;
step 2.2.2: constructing a finite element simulation model of the deformation of the thin-wall piece by adopting finite element software;
step 2.2.3: based on a deep neural network, a thin-wall piece clamping deformation prediction model is established, a nonlinear mapping relation between clamping force and clamping deformation is established, deformation data simulated by a finite element simulation model is used as a training sample set of the thin-wall piece clamping deformation prediction model, and the thin-wall piece clamping deformation prediction model is trained.
6. The method for precisely controlling and optimizing the clamping force of the digital twin-driven thin-wall piece according to claim 5, wherein in step 2.2.2, the method specifically comprises the following steps:
step 2.2.2.1: geometry model importation: importing the thin-wall piece and the clamp geometric model established in the step 2.2.1 into finite element software, wherein the clamp is set to analyze a rigid body, and the thin-wall piece is set to be an isotropic elastomer;
step 2.2.2.2: setting material properties: reading basic information of the thin-wall part and the clamp object of the physical space model established in the step 2.1, and setting material properties of the thin-wall part and the clamp;
step 2.2.2.3: and (3) assembling: assembling the thin-wall part and the clamp in finite element software according to an actual clamping mode, setting a contact position, and defining the thin-wall part and the clamp by adopting 'surface-to-surface contact';
step 2.2.2.4: boundary conditions and loads: completely restraining six degrees of freedom of the bottom surface of the thin-wall part, and applying load on the thin-wall part and the contact position of the clamp;
step 2.2.2.5: dividing grids: the grid control attribute selects hexahedron, agate, central axis algorithm and adopts C3D20R unit to divide grids.
7. The method for precisely controlling and optimizing the clamping force of the digital twin-driven thin-walled workpiece according to claim 5, wherein in step 2.2.3, the sample is subjected to standardization processing: assuming that the mean value of the input feature over the entire sample dataset is u and the standard deviation is σ, the normalization operation is to subtract u from each value of the input feature and then divide by the standard deviation.
8. The method for precisely controlling and optimizing the clamping force of the digital twin-driven thin-wall piece according to claim 5, wherein the specific steps of step 3.2 are as follows:
step 3.2.1: transmitting actions and system control instructions of the thin-wall part clamping process to a thin-wall part-clamp virtual space model in real time, and simulating clamping deformation of the thin-wall part by the thin-wall part-clamp virtual space model, so as to realize three-dimensional visual simulation of clamping behaviors of the thin-wall part;
step 3.2.2: predicting the clamping deformation of the thin-wall part through a thin-wall part clamping deformation prediction model, wherein the method comprises the following specific steps of:
step 3.2.2.1: according to the formula f=km n Input moment M collected by moment spanner n TransformationAnd (3) forming a clamping load F applied to the surface of the thin-walled workpiece, wherein k is a conversion coefficient, and determining by the following experiment: when the torque wrench inputs torque M n Obtaining a clamping force value F ' through the pressure sensor arranged in the step 1.2, so as to obtain a quotient value k ', carrying out multiple tests to obtain a plurality of quotient values k ', and taking an average value as a conversion coefficient k;
step 3.2.2.2: inputting the converted clamping load and the positioning position coordinates into a thin-wall piece clamping deformation prediction model to obtain a clamping deformation prediction value; and comparing the real-time clamping deformation data with historical data under the running condition, and updating the thin-wall part clamping deformation prediction model when the error of the real-time clamping deformation data exceeds the allowable value, wherein the updating step is to firstly add the real-time clamping deformation data into a training set of the thin-wall part clamping deformation prediction model, and then train the thin-wall part clamping deformation prediction model to obtain an updated prediction model.
9. The method for precisely controlling and optimizing the clamping force of the thin-walled workpiece driven by digital twin according to claim 5, wherein the optimization algorithm in the step 3.3 adopts a genetic algorithm, and the objective function of the genetic algorithm is defined as:
minf(NET,s,F)
wherein F represents a thin-wall part clamping deformation prediction model, NET represents a neural network established in the prediction model, S represents positioning point coordinates, S represents a point set where a positioning point is located, F represents the magnitude of clamping force, and F min And F max Representing the minimum and maximum values of clamping force;
randomly generating an initial population comprising p individuals; decoding individuals one by one, eliminating individuals which do not meet constraint conditions in the optimization model, and predicting clamping deformation by the individuals which meet the constraint conditions through a deep neural network;
individual fitness is defined as
In the above, U i I is more than or equal to 1 and less than or equal to p, the dyed individuals in the population, and delta is a value predetermined according to the objective function value of each generation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110186148.9A CN112926152B (en) | 2021-02-15 | 2021-02-15 | Digital twin-driven thin-wall part clamping force precise control and optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110186148.9A CN112926152B (en) | 2021-02-15 | 2021-02-15 | Digital twin-driven thin-wall part clamping force precise control and optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112926152A CN112926152A (en) | 2021-06-08 |
| CN112926152B true CN112926152B (en) | 2023-04-28 |
Family
ID=76169749
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110186148.9A Active CN112926152B (en) | 2021-02-15 | 2021-02-15 | Digital twin-driven thin-wall part clamping force precise control and optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112926152B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113591333B (en) * | 2021-09-30 | 2022-02-18 | 南方电网科学研究院有限责任公司 | Construction method of GIS (gas insulated switchgear) temperature simulation model based on digital twinning |
| CN114803472B (en) * | 2022-04-22 | 2023-12-26 | 深圳航天科技创新研究院 | Clamping control method and system based on robot |
| CN114547826B (en) * | 2022-04-25 | 2022-07-12 | 长江空间信息技术工程有限公司(武汉) | Operation method of engineering deformation monitoring network optimization design system based on digital twin |
| CN117763926A (en) * | 2024-02-22 | 2024-03-26 | 大连理工大学 | digital twin information driven high-reliability structure deformation monitoring method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007285832A (en) * | 2006-04-14 | 2007-11-01 | Nippon Steel Corp | System and method for acquiring breaking point, system and method for estimating break, programs and recording media of these method |
| CN104239599A (en) * | 2014-07-07 | 2014-12-24 | 西安工业大学 | Dynamics simulated analysis method on basis of multipoint positioning flexibility tool system |
| CN106874624A (en) * | 2017-03-15 | 2017-06-20 | 中南大学 | The method evaluated the online virtual detection of the yielding cylindrical member Forming Quality of ultra-thin-wall |
| CN107657129A (en) * | 2017-10-17 | 2018-02-02 | 西北工业大学 | Thin-wall part residual stress deformation based on clamping power monitoring perceives Forecasting Methodology |
| KR20200066932A (en) * | 2018-12-03 | 2020-06-11 | 한국건설기술연구원 | Apparatus and Method for Monitoring Damage of Structure with Measuring Strain and Digital Twin |
| EP3685969A1 (en) * | 2019-01-28 | 2020-07-29 | Siemens Aktiengesellschaft | Computer-aided optimization of a numerically controlled machining of a workpiece |
| CN112327774A (en) * | 2020-11-09 | 2021-02-05 | 东北大学 | Digital twinning-based thin-wall part riveting quality control method |
-
2021
- 2021-02-15 CN CN202110186148.9A patent/CN112926152B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007285832A (en) * | 2006-04-14 | 2007-11-01 | Nippon Steel Corp | System and method for acquiring breaking point, system and method for estimating break, programs and recording media of these method |
| CN104239599A (en) * | 2014-07-07 | 2014-12-24 | 西安工业大学 | Dynamics simulated analysis method on basis of multipoint positioning flexibility tool system |
| CN106874624A (en) * | 2017-03-15 | 2017-06-20 | 中南大学 | The method evaluated the online virtual detection of the yielding cylindrical member Forming Quality of ultra-thin-wall |
| CN107657129A (en) * | 2017-10-17 | 2018-02-02 | 西北工业大学 | Thin-wall part residual stress deformation based on clamping power monitoring perceives Forecasting Methodology |
| KR20200066932A (en) * | 2018-12-03 | 2020-06-11 | 한국건설기술연구원 | Apparatus and Method for Monitoring Damage of Structure with Measuring Strain and Digital Twin |
| EP3685969A1 (en) * | 2019-01-28 | 2020-07-29 | Siemens Aktiengesellschaft | Computer-aided optimization of a numerically controlled machining of a workpiece |
| CN112327774A (en) * | 2020-11-09 | 2021-02-05 | 东北大学 | Digital twinning-based thin-wall part riveting quality control method |
Non-Patent Citations (4)
| Title |
|---|
| 2124铝合金桁梁薄壁件铣削变形仿真优化;龚智鹏等;《机械制造与自动化》;20180620(第03期);全文 * |
| Integrated Simulation of Process of Steel Casting on the Continuous Steel Casting Unit;Alexander Galkin 等;《2020 2nd International Conference on Control Systems, Mathematical Modeling, Automation and Energy Efficiency (SUMMA)》;20201217;全文 * |
| 基于ABAQUS的异型薄壁件装夹变形控制研究;郭诗瑶等;《制造技术与机床》;20180702(第07期);全文 * |
| 薄壁件铣削过程仿真与分析;焦辉等;《中国制造业信息化》;20071001(第19期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112926152A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112926152B (en) | Digital twin-driven thin-wall part clamping force precise control and optimization method | |
| WO2021238191A1 (en) | Positioning compensation method and apparatus for robot | |
| Wilhelm et al. | Task specific uncertainty in coordinate measurement | |
| CN113051831B (en) | Modeling method and thermal error control method for thermal error self-learning prediction model of machine tool | |
| CN111723440B (en) | Thin-wall part machining precision prediction hybrid modeling method | |
| CN113110367B (en) | Engine hardware in-loop test method and system | |
| CN110007645B (en) | Feeding system hybrid modeling method based on dynamics and deep neural network | |
| CN111368466B (en) | Mechanical vibration prediction method based on frequency response function parameter correction | |
| CN107063060A (en) | A kind of method and device for determining surface planarity | |
| CN110631792B (en) | Seismic hybrid test model updating method based on convolutional neural network | |
| CN107609227B (en) | Assembly process optimization method based on maximum entropy theory | |
| Chen | FE model validation for structural dynamics | |
| Hu et al. | Design, implementation, and testing of advanced virtual coordinate-measuring machines | |
| Ostrowska et al. | Comparison of accuracy of virtual articulated arm coordinate measuring machine based on different metrological models | |
| CN116611166A (en) | Aircraft component multi-level assembly deviation analysis method considering initial error | |
| CN111444619B (en) | Online analysis method and equipment for injection mold cooling system | |
| CN113570147B (en) | Method and equipment for rapidly predicting machining errors of thin-wall parts in multiple modes | |
| CN112949000B (en) | Component residual stress inversion method based on convolutional neural network model | |
| CN111210877B (en) | Method and device for deducing physical parameters | |
| CN106709167A (en) | Similarity theory-based assembling validity assessment method | |
| Naumann et al. | Optimization of characteristic diagram based thermal error compensation via load case dependent model updates | |
| Sun et al. | Analysis and inference of stream of dimensional errors in multistage machining process based on an improved semiparametric model | |
| Silvestri et al. | Experimental results of a self-learning compensation system for high precision manufacturing | |
| CN114417490A (en) | Large weak-rigidity edge beam assembling method based on finite element analysis | |
| Zhu et al. | Dynamic Error Compensation Model of Articulated Arm Coordinate Measuring Machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |