Digital twinning and stress prediction method and device for large component additive manufacturing process

Abstract

The invention discloses a digital twinning and stress prediction method and device in a large component additive manufacturing process. A time sequence image is obtained by scanning according to a time sequence through a slidable line laser head, a three-dimensional real-time model corresponding to the time sequence image is constructed, and the surface displacement difference in the process from molten pool formation to cooling solidification is obtained; inputting and loading the surface displacement load into a three-dimensional real-time model for numerical simulation, obtaining a temperature field and a stress field through numerical simulation, comparing the temperature measured by the thermal infrared imager and the stress measured by the stress sensor with the temperature and the stress obtained through numerical simulation, and inputting the obtained data into the three-dimensional real-time model to obtain the three-dimensional real-time model. A digital twin of the real-time molten pool and the deposited layer is formed. According to the method, the simulation time can be well shortened, the efficiency of numerical simulation in the printing process is improved, and meanwhile deformation and cracks generated in the large component printing process can be effectively monitored and predicted.

Description

Technical Field

[0001] The present invention belongs to the field of additive manufacturing, and specifically relates to a digital twin and stress prediction method and device for the additive manufacturing process of a large component. Background Art

[0002] During the additive manufacturing process, stress can cause deformation and cracks in printed parts, which in turn affects the quality of the printed parts. In the printing of large components (super-large components), this will not only affect the manufacturing efficiency, but also increase the manufacturing cost, which seriously affects the development of additive manufacturing technology in the manufacture of large components. By constructing a real-time 3D model digital twin, not only can the printing quality be controlled during the printing process, but the stress of large components can also be predicted, so that the deformation and cracks generated during printing can be effectively controlled.

[0003] In the process of additive manufacturing, numerical simulation is usually used to simulate the printing process, so that the printing process can be optimized, printing defects can be predicted, and trial and error costs can be reduced based on the simulation results. In the printing process of large components, the calculation amount of numerical simulation is very large due to the huge model. Conventional numerical simulation methods are not fully applicable to large components, resulting in a waste of time and energy. Summary of the invention

[0004] Purpose of the invention: In order to solve the problem that in the printing process of large components, the calculation amount of numerical simulation is very large due to the huge model, and the conventional numerical simulation method is not fully applicable to large components, resulting in a waste of time and energy, the present invention provides a digital twin and stress prediction method for the additive manufacturing process of large components. By adopting pooling and partitioning simulation for large components, the simulation time can be greatly reduced, the efficiency of numerical simulation of the printing process can be improved, and at the same time, the deformation and cracks generated in the printing process of large components can be effectively monitored and predicted.

[0005] Technical solution: To achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A digital twin method for a large component additive manufacturing process comprises the following steps:

[0007] Step 1: During the additive manufacturing process of large components, a slidable line laser head is used to scan in a time series to obtain a time-dependent image of the deposition process, which is recorded as a time series image M(t). The image records the morphological height h(x, y, z, t) of the deposit, where x, y, z are spatial coordinates and t is time.

[0008] Step 2: construct a corresponding three-dimensional real-time model based on the time series image M(t).

[0009] Step 3: Obtain the initial morphological height h of the molten pool according to the time series image M(t) initial (x, y, z) and the height h of the morphology after the molten pool cools and solidifies final (x, y, z), according to the initial morphological height h of the molten pool initial (x, y, z) and the height h of the morphology after the molten pool cools and solidifies final (x, y, z) to obtain the surface displacement difference Δh(x, y, z) from the formation of the molten pool to the cooling and solidification process.

[0010] Step 4: Use the surface displacement difference Δh(x, y, z) during the cooling and solidification process as the surface displacement load u(x, y, z), and use the surface displacement load u(x, y, z) as input to load into the three-dimensional real-time model for numerical simulation. The stress generated during the cooling and shrinkage of the molten pool is calculated through numerical simulation.

[0011] Step 5: obtain the temperature field and stress field through numerical simulation, and compare the temperature measured by the infrared thermal imager and the stress measured by the stress sensor with the temperature and stress obtained by numerical simulation. If the error between the two results is within the error range, take the simulation result and the result measured by the stress sensor for average processing, input into the three-dimensional real-time model, and form a digital twin of the real-time molten pool and the deposited layer.

[0012] A stress prediction method for monitoring a large-scale component additive manufacturing process, using a digital twin method of the large-scale component additive manufacturing process, to calculate the stress generated during the cooling and shrinkage of the molten pool through numerical simulation, comprising the following steps:

[0013] Step 41: preliminarily divide the printing area A into several parts according to the size of the established three-dimensional real-time model, and perform pooling partition processing on the three-dimensional real-time model. Each pooling partition is recorded as P i .

[0014] Step 42: Select the first pooling partition P1 according to the printing path, and simulate the stress field of the first pooling partition P1 according to the deposition process obtained by numerical simulation, and obtain the stress denoted as S 1,sim , the stress result measured by the stress sensor is recorded as S 1,meas , the stress S obtained by numerical simulation 1,sim The stress S measured by the stress sensor 1,meas Make the absolute difference, and record the absolute difference as ΔS1.

[0015] Step 43, determine the dynamic adjustment error limit ε i , i = 1, 2, 3, ..., n, n is the total number of pooling partitions. In the first pooling partition, if ΔS1 ≤ ε1, the stress S obtained by numerical simulation is1,sim Input to the corresponding part of the 3D real-time model. If ΔS1>ε1, the two data are averaged to obtain the stress result of the average processing. The stress result of the average processing is The stress result S processed by the mean 1,avg Input to the corresponding part of the 3D real-time model. Select the second pooling partition P2 according to the printing path, and input the numerical simulation stress results of the first pooling area As the boundary condition load stress S for the second pooling region bod , and then solve the second pooling partition in the same way as the first pooling partition. According to the printing path, perform the pooling partition simulation in sequence.

[0016] Preferably: if the selected pooling partition has multiple pooling partitions that have completed numerical simulation, all completed numerical simulation stress results should be considered, and all completed numerical simulation stress results should be summed up as the cumulative stress S sum As the boundary condition for the next region, S sum,i represents the boundary condition of the ith pooling partition. After completing all partition simulations, the actual total response size of the entire 3D real-time model can be obtained by summing up.

[0017] Preferred: Use the stress S obtained sum A deep learning model was trained to predict stress in unprinted areas.

[0018] Preferably: the dynamic adjustment error limit is determined as:

[0019] ε i =α+β·SD(S prev )

[0020] In the formula, ε i To determine the dynamic adjustment error limit, i represents the i-th pooling partition, α is the basic error tolerance, β is the adjustment coefficient, SD(S prev ) is the standard deviation of the stress measure of the previous pooled partition.

[0021] Another object of the present invention is to provide a large-scale component gantry printing device, which is used to implement the stress prediction method for monitoring the additive manufacturing process of the large-scale component, comprising a reversible laser head, an infrared thermal imager, a substrate table, a slidable line laser head, a gantry column, a substrate fixing shaft, a control console, a printing base, a rear beam, and a front beam, wherein:

[0022] The rear crossbeam and the front crossbeam are installed above the gantry column, the gantry column is loaded with an up and down movable shaft, the base fixed shaft is installed on the up and down movable shaft, the base table is fixed on the base fixed shaft, the printing base is arranged on the base table, and the base fixed shaft is loaded with an up and down movable shaft driving motor and a stress sensor.

[0023] The flip-over laser head is loaded on the rear crossbeam through a sliding flipping mechanism, and the infrared thermal imager is installed on the flip-over laser head.

[0024] The slidable line laser head is installed on the front crossbeam through a sliding mechanism.

[0025] The flip-over laser head, the infrared thermal imager, the slidable line laser head and the stress sensor are all connected to the console.

[0026] Preferably, the sliding mechanism includes a linear guide rail, a slider and a linear motor. The linear guide rail is installed on the front crossbeam. The linear guide rail and the slider are slidably connected to each other. The linear motor is installed on the front crossbeam, and the linear motor and the slider are drivingly connected.

[0027] Preferably, the sliding flip mechanism includes a second linear guide rail, a second slider, a second linear motor, a vertical rotating bearing, a powder conveying tube, a motor output shaft, a coupling, a reducer input shaft, a reducer, a reducer output shaft, a rotating bearing sleeve, a horizontal rotating bearing, a universal joint, a laser generator, and a second motor. The second linear guide rail is arranged on the rear crossbeam, the second linear guide rail and the second slider cooperate with each other to slide and connect, the second linear motor is installed on the rear crossbeam, and the second linear motor is drivingly connected to the second slider. The second motor is installed on the second slider, and the second motor, the coupling, the reducer, the horizontal rotating bearing, the universal joint, the vertical rotating bearing, and the laser generator are drivingly connected.

[0028] Preferably, the motor 2 transmits power to the coupling through the motor output shaft, the coupling cooperates with the reducer input shaft, the reducer output shaft cooperates with the horizontal rotating bearing through the rotating bearing sleeve, the horizontal rotating bearing cooperates with the universal joint, the universal joint is connected to the vertical rotating bearing, and the vertical rotating bearing is connected to the laser generator.

[0029] An electronic device includes: at least one processor, at least one memory and a communication interface. The processor, memory and communication interface communicate with each other. The memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the stress prediction method for monitoring the additive manufacturing process of a large component.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. The digital twin of the present invention is based on the real-time morphology obtained by real-time scanning of a slidable linear laser head, which corresponds to the actual deposition process.

[0032] 2. The digital twin proposed in the present invention calculates stress by scanning with a slidable linear laser head and calculating the height difference between the height of the molten pool when it is formed and the height of the deposited layer after cooling, which is used as the surface displacement load to calculate the stress field result. This is different from traditional thermal-mechanical coupling and greatly improves the calculation efficiency.

[0033] 3. The numerical simulation strategy proposed in the present invention divides the completed printing area into several small areas, performs pooling partition simulation, and conducts deep learning on the simulation results. This not only reduces the numerical simulation time of large components, but also can predict stress in unprinted areas.

[0034] 4. The present invention takes into account the errors between the simulation results and the measurement results of the experimental stress sensor equipment, processes the results that exceed the error, and takes the average of the two results, so that the stress assigned value of the digital twin is more accurate and better simulates the real printing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is the front view of a large-scale component gantry printing device.

[0036] Figure 2 It is a top view of a large component gantry printing device.

[0037] Figure 3 It is a rotatable laser head.

[0038] Figure 4 It is a slidable line laser head.

[0039] Figure 5 It is a side view of the front crossbeam linear guide slider.

[0040] Figure 6 It is a digital twin flow chart.

[0041] Figure 7 It is the numerical simulation flow chart. DETAILED DESCRIPTION

[0042] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these examples are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0043] A large component gantry printing device, such as Figure 1 ,2 As shown, it includes a reversible laser head 1, an infrared thermal imager 2, a substrate table 3, a slidable line laser head 4, a gantry column 6, a substrate fixing shaft 7, a control console 9, a printing base 12, a rear beam 13, and a front beam 14, wherein:

[0044] The rear crossbeam 13 and the front crossbeam 14 are installed above the gantry column 6, the gantry column 6 is loaded with an up and down movable shaft 5, the base fixed shaft 7 is installed on the up and down movable shaft 5, the base table 3 is fixed on the base fixed shaft 7, the printing base 12 is arranged on the base table 3, and the base fixed shaft 7 is loaded with an up and down movable shaft driving motor 8 and a stress sensor 15, so as to facilitate the up and down movement of the base table 3.

[0045] The flip-over laser head 1 is loaded on the rear crossbeam 13 through a sliding flip mechanism, and the infrared thermal imager 2 is installed on the flip-over laser head 1 , and the infrared thermal imager 2 measures the temperature of the deposited layer.

[0046] The slidable line laser head 4 is installed on the front crossbeam 14 through a sliding mechanism. The slidable line laser head 4 is used to measure the height of the molten pool and the height of the deposited layer after cooling, and obtain the height difference through a computer, and then obtain the height displacement difference generated in the process of cooling and solidification of the molten pool. The result is transmitted to the computer console, and the displacement difference is loaded as a surface displacement load on the three-dimensional real-time model for numerical simulation to obtain the stress generated in the cooling and shrinking process of the molten pool.

[0047] The flip-type laser head 1 , the infrared thermal imager 2 , the slidable line laser head 4 , and the stress sensor 15 are all connected to the console 9 .

[0048] In another embodiment, if Figure 4 , 5 As shown, the sliding mechanism includes a linear guide rail 261, a slider 271 and a linear motor 161. The linear guide rail 261 is installed on the front beam 14. The linear guide rail 261 and the slider 271 are slidably connected with each other. The linear motor 161 is installed on the front beam 14, and the linear motor 161 and the slider 271 are drivingly connected to realize the left and right movement of the movable line laser head 4 on the front beam 14.

[0049] In another embodiment, if Figure 3As shown, the sliding flipping mechanism includes a linear guide rail 262, a slider 272, a linear motor 2, a vertical rotating bearing 10, a powder conveying tube 11, a motor output shaft 17, a coupling 18, a reducer input shaft 19, a reducer 20, a reducer output shaft 21, a rotating bearing sleeve 22, a horizontal rotating bearing 23, a universal joint 24, a laser generator 25, and a motor 28. The linear guide rail 262 is arranged on the rear beam 13, and the linear guide rail 262 and the slider 272 are slidably connected with each other. The linear motor 162 is installed on the rear beam 13, and the linear motor 162 is drivingly connected with the slider 272 to realize the left and right movement of the flippable laser head 1 on the rear beam 13. In another embodiment, the second motor 28 is installed on the second slider 272, and the second motor 28, the coupling 18, the reducer 20, the horizontal rotation bearing 23, the universal joint 24, the vertical rotation bearing 10, and the laser generator 25 are connected in transmission. The second motor 28 transmits power to the coupling 18 through the motor output shaft 17, the coupling 18 cooperates with the reducer input shaft 19, the reducer output shaft 21 cooperates with the horizontal rotation bearing 23 through the rotating bearing sleeve 22, the horizontal rotation bearing 23 cooperates with the universal joint 24, the universal joint 24 is connected to the vertical rotation bearing 10, and the vertical rotation bearing 10 is connected to the laser generator 25. The powder conveying pipe 11 conveys powder to the laser head to realize the rotation transformation of the flip-type laser head 1 at any angle.

[0050] The slidable linear laser head 4 is used to measure the height of the molten pool and the height of the deposited layer after cooling and obtain the height difference through a computer, and then obtain the height displacement difference generated in the process of cooling and solidification of the molten pool. The result is transmitted to the computer console 9, and the displacement difference is loaded as a surface displacement load onto the three-dimensional real-time model for numerical simulation to obtain the stress generated in the cooling and shrinkage process of the molten pool.

[0051] A digital twin method for the additive manufacturing process of a large component generates a three-dimensional real-time model based on the real-time measurement of the size and shape of the molten pool and the size and shape of the deposited layer by a slidable line laser head 4, performs numerical simulation on the three-dimensional real-time model, obtains the temperature field and the stress field, and makes a difference between the temperature measured by the infrared thermal imager 2 and the stress measured by the stress sensor 15 and the temperature and stress obtained by the numerical simulation. If the error between the two results is within the relative error range, the simulation result is input into the three-dimensional real-time model; if it is not within the range error, the simulation result and the result measured by the stress sensor 13 are averaged and input into the three-dimensional real-time model, thereby forming a digital twin of the real-time deposited layer, such as Figure 6 As shown, the specific steps include:

[0052] Step 1: During the additive manufacturing process of large components, a slidable line laser head is used to scan in a time series to obtain a time-dependent image of the deposition process, which is recorded as a time series image M(t). The image records the morphological height h(x, y, z, t) of the deposit, where x, y, z are spatial coordinates and t is time.

[0053] Step 2: construct a corresponding three-dimensional real-time model based on the time series image M(t).

[0054] Step 3: In order to simulate the stress distribution during the deposition process, the morphological height at the initial stage of the molten pool formation is defined as h initial (x, y, z), and the height of the morphology after the molten pool cools and solidifies is h final (x, y, z), the morphological height h at the initial stage of molten pool formation is obtained according to the time series image M(t) initial (x, y, z) and the height h of the morphology after the molten pool cools and solidifies final (x, y, z), according to the initial morphological height h of the molten pool initial (x, y, z) and the height h of the morphology after the molten pool cools and solidifies final (x, y, z) to obtain the surface displacement difference Δh(x, y, z) from the formation of the molten pool to the cooling and solidification process.

[0055] Step 4: The surface displacement difference Δh(x, y, z) during the cooling and solidification process is used as the surface displacement load u(x, y, z), that is, u(x, y, z) = Δh(x, y, z). The surface displacement load u(x, y, z) is used as input and loaded into the three-dimensional real-time model for numerical simulation. The stress generated during the cooling and shrinkage of the molten pool is calculated through numerical simulation.

[0056] Numerical simulation process: According to the laser power, numerical simulation is performed to obtain the temperature field of the deposited layer. According to the surface displacement difference scanned and calculated by the slidable line laser head, the surface displacement difference is used as the displacement load, and the displacement load is loaded on the model as the boundary condition for numerical simulation to obtain the stress field results.

[0057] In the numerical simulation of large components, a numerical simulation strategy is as follows: according to the real-time deposition layer morphology obtained by scanning with a slidable line laser head, a three-dimensional real-time model is established, the printing area is preliminarily divided into several parts according to the size of the model, and the three-dimensional real-time model is pooled and partitioned. The first pooled partition is selected according to the printing path, and the stress field of the first pooled partition is simulated according to the above numerical simulation process. The obtained result is compared with the stress result measured by the stress sensor. If the two results are within the error range, the simulation result is imported into the real-time model; if not within the error range, the pooled partition is optimized, the optimized simulation result and the stress sensor measurement result are averaged, and the averaged result is imported into the real-time model. Then, the second pooled partition is selected according to the printing path, and the numerical simulation result of the first pooled area is used as the boundary condition of the second pooled area for numerical simulation. Similarly, consider whether the obtained stress result and the stress sensor measurement result are within the error range. If they are within the error range, the simulation result is imported into the real-time model. If they are not within the error range, the simulation result and the stress sensor measurement result are averaged, and the averaged result is imported into the real-time model. According to the printing path, the pooling partition simulation is carried out in sequence. If there are multiple pooling partitions that have completed the simulation in the selected pooling area, all the completed simulation results should be considered and simulated as boundary conditions. The numerical simulation results can be deeply learned to predict the stress size of large components.

[0058] Step 5: obtain the temperature field and stress field through numerical simulation, and compare the temperature measured by the infrared thermal imager and the stress measured by the stress sensor with the temperature and stress obtained by numerical simulation. If the error between the two results is within the error range, take the simulation result and the result measured by the stress sensor for average processing, input into the three-dimensional real-time model, and form a digital twin of the real-time molten pool and the deposited layer.

[0059] A stress prediction method for monitoring the additive manufacturing process of a large component, using a digital twin method of the additive manufacturing process of the large component, and a method for calculating the stress generated during the cooling and shrinkage of the molten pool through numerical simulation, such as Figure 7 As shown, the following steps are included:

[0060] Step 41: preliminarily divide the printing area A into several parts according to the size of the established three-dimensional real-time model, and perform pooling partition processing on the three-dimensional real-time model. Each pooling partition is recorded as P i .

[0061] Step 42: Select the first pooling partition P1 according to the printing path, and simulate the stress field of the first pooling partition P1 according to the deposition process obtained by numerical simulation, and obtain the stress denoted as S 1,sim, the stress result measured by the stress sensor 15 is recorded as S 1,meas , the stress S obtained by numerical simulation 1,sim The stress S measured by the stress sensor 15 1,meas Make an absolute difference, and record the absolute difference as ΔS1, ΔS1=|S 1,sim -S 1,meas |.

[0062] Step 43, determine the dynamic adjustment error limit ε i , i = 1, 2, 3, …, n, where n is the total number of pooling partitions.

[0063] Determine the dynamic adjustment error limit as:

[0064] ε i =α+β·SD(S prev )

[0065] In the formula, ε i To determine the dynamic adjustment error limit, i represents the i-th pooling partition, α is the basic error tolerance, β is the adjustment coefficient, SD(S prev ) is the standard deviation of the stress measure of the previous pooled partition.

[0066] In the first pooling partition, if ΔS1≤ε1, the stress S obtained by numerical simulation is 1,sim Input to the corresponding part of the 3D real-time model. If ΔS1>ε1, the two data are averaged to obtain the stress result of the average processing. The stress result of the average processing is The stress result S processed by the mean 1,avg Input to the corresponding part of the 3D real-time model. Select the second pooling partition P2 according to the printing path, and input the numerical simulation stress results of the first pooling area As the boundary condition load stress S for the second pooling region bod , and then solve the second pooling partition in the same way as the first pooling partition. According to the printing path, perform the pooling partition simulation in sequence.

[0067] It should be noted that if the selected pooling partition contains multiple pooling partitions that have completed numerical simulation, all completed numerical simulation stress results should be considered and the sum of all completed numerical simulation stress results should be recorded as the cumulative stress S sum As the boundary condition for the next region, S sum,i represents the boundary condition of the ith pooling partition. After completing all partition simulations, the actual total response size of the entire 3D real-time model can be obtained by summing up. Using the obtained stress S sum A deep learning model was trained to predict stress in unprinted areas.

[0068] In another embodiment, an electronic device is provided, comprising: at least one processor, at least one memory and a communication interface. The processor, the memory and the communication interface communicate with each other. The memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the stress prediction method for monitoring the additive manufacturing process of a large component.

[0069] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A digital twin method for additive manufacturing of large components, characterized in that: The following steps are involved: Step 1, in the process of additive manufacturing of large components, a slidable line laser head (4) is used to scan in time series to obtain a time-dependent image of the deposition process, which is recorded as a time series image M(t), and the image records the topography height h(x, y, z, t) of the deposit, where x, y, z are spatial coordinates and t is time; Step 2, constructing a corresponding three-dimensional real-time model based on the time series image M(t); Step 3: Obtain the initial morphological height h of the molten pool according to the time series image M(t) initial (x, y, z) and the height h of the morphology after the molten pool cools and solidifies final (x, y, z), according to the initial morphological height h of the molten pool initial (x, y, z) and the height h of the morphology after the molten pool cools and solidifies final (x, y, z) to obtain the surface displacement difference Δh(x, y, z) from the formation of the molten pool to the cooling and solidification process; Step 4, using the surface displacement difference Δh(x, y, z) during the cooling and solidification process as the surface displacement load u(x, y, z), and using the surface displacement load u(x, y, z) as input to load into the three-dimensional real-time model for numerical simulation, and calculating the stress generated during the cooling and shrinkage of the molten pool through numerical simulation; Step 5, obtain the temperature field and stress field through numerical simulation, and compare the temperature measured by the infrared thermal imager and the stress measured by the stress sensor with the temperature and stress obtained by numerical simulation. If the error between the two results is within the error range, take the simulation result and the result measured by the stress sensor, perform average processing, and input them into the three-dimensional real-time model to form a digital twin of the real-time molten pool and deposition layer.

2. A stress prediction method for monitoring the additive manufacturing process of large components, characterized in that: A method for calculating the stress generated during the cooling and shrinkage of the molten pool by numerical simulation using the digital twin method of the additive manufacturing process of a large component as described in claim 1, comprising the following steps: Step 41: preliminarily divide the printing area A into several parts according to the size of the established three-dimensional real-time model, and perform pooling partition processing on the three-dimensional real-time model. Each pooling partition is recorded as P i ; Step 42: Select the first pooling partition P1 according to the printing path, and simulate the stress field of the first pooling partition P1 according to the deposition process obtained by numerical simulation, and obtain the stress denoted as S 1,sim , the stress result measured by the stress sensor (15) is recorded as S 1,meas , the stress S obtained by numerical simulation 1,sim The stress S measured by the stress sensor (15) 1,meas Make an absolute difference, and record the absolute difference as ΔS1; Step 43, determine the dynamic adjustment error limit ε i , i = 1, 2, 3, ..., n, n is the total number of pooling partitions; in the first pooling partition, if ΔS1 ≤ ε1, the stress S obtained by numerical simulation is 1,sim Input to the corresponding part of the 3D real-time model; if ΔS1>ε1, the two data are averaged to obtain the stress result of the average processing, and the stress result of the average processing is The stress result S processed by the mean 1,avg Input to the corresponding part of the 3D real-time model; select the second pooling partition P2 according to the printing path, and put the numerical simulation stress results of the first pooling area As the boundary condition load stress S for the second pooling region bod , and then solve the second pooling partition in the same way as the first pooling partition; and perform pooling partition simulations in sequence according to the printing path.

3. The stress prediction method for monitoring the additive manufacturing process of large components according to claim 2, characterized in that: If the selected pooling partition has multiple pooling partitions that have completed numerical simulation, all completed numerical simulation stress results should be considered and the sum of all completed numerical simulation stress results is recorded as the cumulative stress S sum As the boundary condition of the next region, S sum,i Represents the boundary condition of the ith pooling partition; after completing all partition simulations, the actual total response size of the entire 3D real-time model can be obtained by summing up 4. The stress prediction method for monitoring the additive manufacturing process of large components according to claim 3 is characterized in that: Using the obtained stress S sum A deep learning model was trained to predict stress in unprinted areas.

5. The stress prediction method for monitoring the additive manufacturing process of large components according to claim 4, characterized in that: Determine the dynamic adjustment error limit as: e i =α+β·SD(S prev ) In the formula, ε i To determine the dynamic adjustment error limit, i represents the i-th pooling partition, α is the basic error tolerance, β is the adjustment coefficient, SD(S prev ) is the standard deviation of the stress measure of the previous pooled partition.

6. A large component gantry printing device, characterized in that: The stress prediction method for realizing the monitoring of the additive manufacturing process of large components as claimed in claim 2 comprises a reversible laser head (1), an infrared thermal imager (2), a substrate table (3), a slidable line laser head (4), a gantry column (6), a substrate fixing shaft (7), a control console (9), a printing base (12), a rear beam (13), and a front beam (14), wherein: The rear cross beam (13) and the front cross beam (14) are installed above the gantry column (6), the gantry column (6) is equipped with an up-and-down movable shaft (5), the base fixed shaft (7) is installed on the up-and-down movable shaft (5), the base platform (3) is fixed on the base fixed shaft (7), the printing base (12) is arranged on the base platform (3), and the base fixed shaft (7) is equipped with an up-and-down movable shaft driving motor (8) and a stress sensor (15); The flip-over laser head (1) is loaded on the rear crossbeam (13) via a sliding flip mechanism, and the infrared thermal imager (2) is mounted on the flip-over laser head (1); The slidable line laser head (4) is mounted on the front crossbeam (14) via a sliding mechanism; The flippable laser head (1), the infrared thermal imager (2), the slidable line laser head (4), and the stress sensor (15) are all connected to the console (9).

7. The large-scale component gantry printing device according to claim 6, characterized in that: The sliding mechanism comprises a linear guide rail (261), a slider (271) and a linear motor (161); the linear guide rail (261) is mounted on a front crossbeam (14); the linear guide rail (261) and the slider (271) are slidably connected to each other; the linear motor (161) is mounted on the front crossbeam (14); and the linear motor (161) and the slider (271) are drivingly connected.

8. The large-scale component gantry printing device according to claim 7, characterized in that: The sliding flip mechanism comprises a second linear guide rail (262), a second slider (272), a second linear motor, a vertical rotary bearing (10), a powder conveying pipe (11), a motor output shaft (17), a coupling (18), a reducer input shaft (19), a reducer (20), a reducer output shaft (21), a rotary bearing sleeve (22), a horizontal rotary bearing (23), a universal joint (24), a laser generator (25), and a second motor (28). The second linear guide rail (262) is arranged on the rear crossbeam (13). The linear guide rail 2 (262) and the slider 2 (272) are slidably connected with each other, the linear motor 2 (162) is installed on the rear crossbeam (13), and the linear motor 2 (162) and the slider 2 (272) are drivingly connected; the motor 2 (28) is installed on the slider 2 (272), and the motor 2 (28), the coupling (18), the reducer (20), the horizontal rotating bearing (23), the universal joint (24), the vertical rotating bearing (10), and the laser generator (25) are drivingly connected.

9. The large-scale component gantry printing device according to claim 8, characterized in that: The motor 2 (28) transmits power to the coupling (18) through the motor output shaft (17); the coupling (18) cooperates with the reducer input shaft (19); the reducer output shaft (21) cooperates with the horizontal rotating bearing (23) through the rotating bearing sleeve (22); the horizontal rotating bearing (23) cooperates with the universal joint (24); the universal joint (24) is connected to the vertical rotating bearing (10); and the vertical rotating bearing (10) is connected to the laser generator (25).

10. An electronic device, characterized in that: include: At least one processor, at least one memory and a communication interface; the processor, memory and communication interface communicate with each other; The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the stress prediction method for monitoring the additive manufacturing process of large components as described in any one of claims 2-5.

Images