A method for laser additive manufacturing heat history database establishment and screening

Abstract

The application discloses a kind of methods for laser additive manufacturing heat history database establishment and screening, comprising the following steps: one, determine the composition of alloy sample to be laser additive manufactured;Two, construct the laser in-situ thermal cycling sample heat history database of alloy sample composition to be laser additive manufactured;Three, prepare laser in-situ thermal cycling sample under corresponding laser action process scheme;Four, the laser in-situ thermal cycling sample corresponding to different heat history conditions is carried out microstructure characterization and performance test;Five, train BP neural network model;Six, heat history data group screening of alloy sample to be laser additive manufactured.The application takes the composition of alloy sample to be laser additive manufactured, microstructure feature, hardness and elastic modulus performance as the input of BP neural network model, and corresponding laser action process scheme and its heat history curve are taken as the output of BP neural network model, and corresponding heat history curve can be screened out, to guide the rapid design of actual additive manufacturing process parameters.

Description

Technical Field

[0001] This invention belongs to the field of thermal history data screening technology, specifically relating to a method for establishing and screening thermal history databases for laser additive manufacturing. Background Technology

[0002] Metal additive manufacturing (AM) offers numerous advantages over traditional manufacturing methods, such as greater design freedom and near-net-shape forming, the ability to manufacture complex shapes without extensive post-processing steps, and superior mechanical properties compared to traditional processes. In actual production, the microstructure and properties of laser additively manufactured samples depend primarily on different manufacturing dimensions, structures, and process conditions, essentially reflecting the sample's experience of varying complex cyclic thermal histories. The laser's point-to-point, layer-by-layer action provides a wide range of thermal history control, thus offering high design freedom for the microstructure.

[0003] However, existing research has limited process windows, and the efficiency of exploring optimal process conditions by preparing large-sized deposited samples is low. This results in an incomplete understanding of the relationship between the cyclic thermal history, microstructure, and properties of laser additive manufacturing samples, particularly regarding the unclear patterns of alloy composition-process-microstructure-properties under unconventional or even extreme process conditions, leading to incomplete sample coverage. Furthermore, the current screening of additive manufacturing alloy preparation processes is largely arbitrary, often considering only formability while neglecting the active control of thermal history behavior, making it difficult to achieve optimal microstructure selection and limiting the full potential of additive manufacturing technology. Therefore, providing a method for establishing a thermal history database for laser additive manufacturing, rapidly and efficiently acquiring the relationship between alloy composition, thermal history, microstructure, and properties, and accurately exploring appropriate additive manufacturing processes for specific alloys as needed, thereby achieving active control of thermal history behavior, is a crucial step in advancing additive manufacturing technology. Summary of the Invention

[0004] The technical problem to be solved by this invention is to address the shortcomings of the prior art by providing a method for establishing and screening a thermal history database for laser additive manufacturing. This method involves constructing a thermal history database of laser in-situ thermal cycling samples with the same composition as the alloy to be laser-added, obtaining thermal history curves under different laser processing schemes, preparing laser in-situ thermal cycling samples according to the laser processing scheme, and performing microstructure characterization and micro-area performance testing on the obtained samples. Using the composition, microstructure characteristics, hardness, and elastic modulus of the alloy sample to be laser-added as the input layer nodes of a BP neural network model, and the laser processing scheme and corresponding thermal history curves as the output layer nodes, the BP neural network model is trained. This allows for the screening of the laser processing scheme and corresponding thermal history curves for the desired samples, guiding the rapid design of actual additive manufacturing process parameters. This method is suitable for obtaining unconventional processes, unconventional structures, and unconventional properties in additive manufacturing, maximizing the advantages of specific alloy compositions in the additive manufacturing process, and facilitating widespread application.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for establishing and screening a thermal history database for laser additive manufacturing, characterized in that the method includes the following steps:

[0006] Step 1: Determine the composition of the alloy sample to be laser additively manufactured;

[0007] Step 2: Construct a thermal history database of the laser in-situ thermal cycling sample composition of the alloy sample to be laser additively manufactured. The specific process is as follows:

[0008] Step 201: Use a computer and finite element calculation software to establish a geometric model of the powder bed of the laser in-situ thermal cycling sample composition;

[0009] Step 202: Calculate the thermal properties of the laser in-situ thermal cycling sample in step 201, and assign the thermal properties to the geometric model of the powder bed of the alloy composition to be laser additively manufactured.

[0010] Step 203: Assign boundary conditions, initial temperature, and laser absorption rate to the geometric model of the laser in-situ thermal cycling sample powder bed;

[0011] Step 204: Set up multiple different laser action process schemes, wherein the laser action process scheme includes laser power P, laser action time t, and laser action interval time T;

[0012] Step 205: Use finite element calculation software to perform mesh generation, obtain thermal history curves under different laser treatment process schemes, and construct a thermal history database of the laser in-situ thermal cycling sample from Step 201. The thermal history database includes multiple sets of thermal history data, and each thermal history data set includes thermal history curves and corresponding laser treatment process schemes.

[0013] Step 3: Prepare laser in-situ thermal cycling samples under the corresponding laser action process scheme. The specific process for preparing laser in-situ thermal cycling samples under any laser action process scheme is as follows: according to the laser power P, laser action time t and laser action interval time T under the corresponding laser action process scheme, obtain laser in-situ thermal cycling samples corresponding to different thermal behavior conditions.

[0014] Step 4: Perform microstructure characterization and performance testing on laser in-situ thermal cycling samples under different thermal history conditions: Use optical microscopy, scanning electron microscopy and X-ray diffraction to characterize each laser in-situ thermal cycling sample and obtain the microstructure characteristics and parameters of each laser in-situ thermal cycling sample.

[0015] Micro-nano indentation tests were performed on laser-induced in-situ thermal cycling samples to obtain the hardness and elastic modulus of alloy samples under different thermal history conditions.

[0016] Step 5: Train the BP neural network model: Use the composition, microstructure characteristics, hardness and elastic modulus properties of the laser in-situ thermal cycling sample as the input layer nodes of the BP neural network model, and use the laser treatment process scheme and the corresponding thermal history curve as the output layer nodes of the BP neural network model to train the BP neural network model.

[0017] Step 6: Screening of thermal history data sets for alloy samples to be laser-added manufacturing: Input the composition of the alloy samples to be laser-added manufacturing, as well as the desired microstructure characteristics and properties, into the trained BP neural network model to screen out laser treatment process schemes. The BP neural network model then uses the laser power P, laser treatment time t, and laser treatment interval T in the laser treatment process scheme to fit the thermal history curve, thereby guiding the actual laser additive manufacturing process design.

[0018] The above-mentioned method for establishing and filtering a thermal history database for laser additive manufacturing is characterized in that: the finite element calculation software includes Comsol or Ansys finite element calculation software.

[0019] The above-mentioned method for establishing and screening a thermal history database for laser additive manufacturing is characterized in that: in step 202, Jmat-Pro or Thermo-Calc software is used to calculate the thermal property parameters of the laser in-situ thermal cycling sample in step 201, wherein the thermal property parameters include the density, specific heat capacity, thermal conductivity and enthalpy of the laser in-situ thermal cycling sample.

[0020] The above-mentioned method for establishing and screening a thermal history database for laser additive manufacturing is characterized in that the boundary conditions of the geometric model of the laser in-situ thermal cycling sample powder bed include the heat transfer coefficient and the radiation coefficient.

[0021] The above-mentioned method for establishing and screening a thermal history database for laser additive manufacturing is characterized in that: the microstructure parameters include multiple phase morphology parameters, and each phase morphology parameter includes phase shape, phase size and phase volume fraction.

[0022] Compared with the prior art, the present invention has the following advantages:

[0023] 1. This invention constructs a thermal history database of laser in-situ thermal cycling samples to obtain thermal history curves under different laser treatment process schemes. Laser in-situ thermal cycling samples are prepared according to the laser treatment process schemes, and the obtained samples are characterized by microstructure and tested for micro-area properties. Using the composition, microstructure characteristics, hardness, and elastic modulus properties of the laser in-situ thermal cycling samples as input layer nodes of a BP neural network model, and the thermal history curves and corresponding laser treatment process schemes as output layer nodes, the BP neural network model is trained. This allows for the selection of the desired laser treatment process scheme and corresponding thermal history curves for the samples, guiding the rapid design of actual additive manufacturing process parameters.

[0024] 2. This invention establishes a thermal history database based on the composition of laser in-situ thermal cycling samples with the same composition as the alloy sample to be laser-added manufacturing. This database can be used to guide the actual additive manufacturing process. Based on the selected thermal history and process conditions of the laser in-situ thermal cycling samples, the thermal behavior of the actual additive manufacturing is made relatively consistent with it. This can guide the rapid design of actual additive manufacturing process parameters. In addition to guiding conventional additive manufacturing technology, it can also guide the acquisition of unconventional processes, unconventional microstructures, and unconventional properties in the additive manufacturing process. It maximizes the advantages of specific alloy compositions in the additive manufacturing process. The process is precise and controllable. At the same time, it makes full use of metal powder, with advantages such as high material utilization and short overall sample preparation cycle. The method is efficient, energy-saving, and environmentally friendly, and is easy to promote and use.

[0025] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0026] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0027] like Figure 1 As shown, the present invention provides a method for establishing and filtering a thermal history database for laser additive manufacturing, comprising the following steps:

[0028] Step 1: Determine the composition of the alloy sample to be laser additively manufactured;

[0029] Step 2: Construct a thermal history database of the laser in-situ thermal cycling sample composition of the alloy sample to be laser additively manufactured. The specific process is as follows:

[0030] Step 201: Use a computer and finite element calculation software to establish a geometric model of the powder bed of the laser in-situ thermal cycling sample composition;

[0031] Step 202: Calculate the thermal properties of the laser in-situ thermal cycling sample in step 201, and assign the thermal properties to the geometric model of the powder bed of the alloy composition to be laser additively manufactured.

[0032] Step 203: Assign boundary conditions, initial temperature, and laser absorption rate to the geometric model of the laser in-situ thermal cycling sample powder bed;

[0033] Step 204: Set up multiple different laser action process schemes, wherein the laser action process scheme includes laser power P, laser action time t, and laser action interval time T;

[0034] Step 205: Use finite element calculation software to perform mesh generation, obtain thermal history curves under different laser treatment process schemes, and construct a thermal history database of the laser in-situ thermal cycling sample from Step 201. The thermal history database includes multiple sets of thermal history data, and each thermal history data set includes thermal history curves and corresponding laser treatment process schemes.

[0035] Step 3: Prepare laser in-situ thermal cycling samples under the corresponding laser action process scheme. The specific process for preparing laser in-situ thermal cycling samples under any laser action process scheme is as follows: according to the laser power P, laser action time t and laser action interval time T under the corresponding laser action process scheme, obtain laser in-situ thermal cycling samples corresponding to different thermal history conditions.

[0036] Step 4: Perform microstructure characterization and performance testing on laser in-situ thermal cycling samples under different thermal history conditions: Use optical microscopy, scanning electron microscopy and X-ray diffraction to characterize each laser in-situ thermal cycling sample and obtain the microstructure characteristics and parameters of each laser in-situ thermal cycling sample.

[0037] Micro-nano indentation tests were performed on laser-induced in-situ thermal cycling samples to obtain the hardness and elastic modulus of alloy samples under different thermal history conditions.

[0038] Step 5: Train the BP neural network model: Use the composition, microstructure characteristics, hardness and elastic modulus properties of the laser in-situ thermal cycling sample as the input layer nodes of the BP neural network model, and use the laser treatment process scheme and the corresponding thermal history curve as the output layer nodes of the BP neural network model to train the BP neural network model.

[0039] Step 6: Screening of thermal history data sets for alloy samples to be laser-added manufacturing: Input the composition of the alloy samples to be laser-added manufacturing, as well as the desired microstructure characteristics and properties, into the trained BP neural network model to screen out laser treatment process schemes. The BP neural network model then uses the laser power P, laser treatment time t, and laser treatment interval T in the laser treatment process scheme to fit the thermal history curve, thereby guiding the actual laser additive manufacturing process design.

[0040] It should be noted that the spherical sample rapidly formed on a rectangular powder bed using laser action is a laser in-situ thermal cycling sample. In the actual preparation process, the laser action process scheme is slightly different, but the thermal history curve is the same. Therefore, the BP neural network model can use the laser power P, laser action time t and laser action interval T in the laser action process scheme to fit the thermal history curve and guide the design of actual process parameters for additive manufacturing alloys.

[0041] In this embodiment, the finite element calculation software includes Comsol or Ansys finite element calculation software.

[0042] In this embodiment, in step 202, the thermal properties of the laser in-situ thermal cycling sample in step 201 are calculated using Jmat-Pro or Thermo-Calc software. The thermal properties include the density, specific heat capacity, thermal conductivity and enthalpy of the laser in-situ thermal cycling sample.

[0043] In this embodiment, the boundary conditions of the geometric model of the laser in-situ thermal cycling sample powder bed include the heat transfer coefficient and the radiation coefficient.

[0044] In this embodiment, the microstructure parameters include multiple phase morphology parameters, each of which includes phase shape, phase size, and phase volume fraction.

[0045] In use, this invention constructs a thermal history database of laser-induced in-situ thermal cycling samples to obtain thermal history curves under different laser-assisted processing schemes. Based on the laser-assisted processing scheme, samples are prepared for laser-induced in-situ thermal cycling, and their microstructure characterization and performance testing are performed. The composition, microstructure characteristics, hardness, and elastic modulus of the laser-induced in-situ thermal cycling samples are used as input layer nodes of a BP neural network model, while the laser-assisted processing scheme and corresponding thermal history curves are used as output layer nodes. Training the BP neural network model allows for the selection of the desired laser-assisted processing scheme and corresponding thermal history curves for the samples, guiding the rapid design of actual additive manufacturing process parameters. Based on the laser-assisted additive manufacturing process... A thermal history database established by creating laser in-situ thermal cycling samples with the same alloy composition can be used to guide the actual additive manufacturing process. By designing thermal history and process conditions based on the selected laser in-situ thermal cycling samples, the thermal behavior of actual additive manufacturing can be made relatively consistent with them. This can guide the rapid design of actual additive manufacturing process parameters. In addition to guiding conventional additive manufacturing technology, it can also guide the acquisition of unconventional processes, unconventional microstructures, and unconventional properties in additive manufacturing. It maximizes the advantages of specific alloy compositions in additive manufacturing, making the process precise and controllable. At the same time, it makes full use of metal powder, with advantages such as high material utilization and short overall sample preparation cycle. The method is efficient, energy-saving, and environmentally friendly.

[0046] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for laser additive manufacturing heat history database creation and screening, characterized in that, The method comprises the following steps: Step one, determining the composition of the alloy sample to be laser additive manufactured; Step two, constructing a thermal history database of the laser in-situ thermal cycling sample with the composition of the alloy sample to be laser additive manufactured, and the specific process is as follows: Step 201, using a computer to establish a powder bed geometric model of the laser in-situ thermal cycling sample by using a finite element calculation software; Step 202, calculating the thermal physical parameters of the laser in-situ thermal cycling sample in step 201, and assigning the thermal physical parameters to the powder bed geometric model of the alloy composition to be laser additive manufactured; Step 203, assigning the boundary conditions, initial temperature and absorption rate of laser to the powder bed geometric model of the laser in-situ thermal cycling sample; Step 204, setting multiple groups of different laser action process schemes, wherein the laser action process scheme comprises laser power P, laser action time t and laser action interval time T; Step 205, performing mesh division by using the finite element calculation software to obtain the thermal history curve under different laser action process schemes, and constructing the thermal history database of the laser in-situ thermal cycling sample in step 201, wherein the thermal history database comprises multiple groups of thermal history data, and each group of thermal history data comprises a thermal history curve and a corresponding laser action process scheme; Step three, preparing the laser in-situ thermal cycling sample under the corresponding laser action process scheme, wherein the preparation process of the laser in-situ thermal cycling sample under any laser action process scheme is as follows: obtaining the laser in-situ thermal cycling sample corresponding to different thermal history conditions according to the laser power P, laser action time t and laser action interval time T under the corresponding laser action process scheme; Step four, performing microstructure characterization and performance testing on the laser in-situ thermal cycling sample corresponding to different thermal history conditions: performing characterization on each laser in-situ thermal cycling sample by using an optical microscope, a scanning electron microscope and an X-ray diffractometer to obtain the microstructure characteristics and parameters of each laser in-situ thermal cycling sample; Performing micro-nano indentation testing on the laser in-situ thermal cycling sample to obtain the hardness and elastic modulus of the alloy sample corresponding to different thermal history conditions; Step five, training a BP neural network model: taking the composition of the laser in-situ thermal cycling sample, the microstructure characteristics, the hardness and the elastic modulus as the input layer nodes of the BP neural network model, and taking the laser action process scheme and the corresponding thermal history curve as the output layer nodes of the BP neural network model, and training the BP neural network model; Step six, screening the thermal history data group of the alloy sample to be laser additive manufactured: inputting the composition of the alloy sample to be laser additive manufactured and the desired microstructure characteristics and performance into the trained BP neural network model to screen out the laser action process scheme, and the BP neural network model fits the thermal history curve by using the laser power P, the laser action time t and the laser action interval time T in the laser action process scheme, thereby guiding the actual laser additive manufacturing process design.

2. A method for laser additive manufacturing heat history database creation and screening as claimed in claim 1, characterized in that: The finite element calculation software comprises Comsol or Ansys finite element calculation software.

3. A method for laser additive manufacturing heat history database creation and screening as claimed in claim 1, wherein: In step 202, the thermal physical parameters of the laser in-situ heat treatment sample in step 201 are calculated by using Jmat-Pro or Thermo-Calc software, and the thermal physical parameters include the density, specific heat capacity, thermal conductivity and enthalpy of the composition of the laser in-situ heat treatment sample.

4. A method for laser additive manufacturing heat history database creation and screening as claimed in claim 1, wherein: The boundary conditions of the powder bed geometry model of the composition of the laser in-situ heat treatment sample include a heat transfer coefficient and a radiation coefficient.

5. A method for laser additive manufacturing heat history database creation and screening as claimed in claim 1, wherein: The microstructure parameters include a plurality of phase morphology parameters, and each phase morphology parameter includes a phase shape, a phase size and a phase volume fraction.

Images