A molten bath oscillation frequency and radius based online molten bath depth detection system

Abstract

The application provides a molten pool depth online detection system based on molten pool oscillation frequency and radius, which realizes online real-time monitoring of molten pool photoelectric signals and molten pool temperature field through a photoelectric signal acquisition module and a molten pool temperature field measurement module, and then can realize real-time calculation of molten pool equivalent radius and molten pool intrinsic oscillation frequency; then a relationship model between molten pool depth and molten pool intrinsic oscillation frequency and molten pool equivalent radius in a metal additive manufacturing process is established, so that real-time online detection of molten pool depth in the manufacturing process is realized.

Description

Technical Field

[0001] This invention belongs to the field of infrared optical mechanics technology, and particularly relates to an online melt depth detection system based on the oscillation frequency and radius of the molten pool. Background Technology

[0002] Melt depth refers to the distance between the deepest point of the molten pool and its surface, that is, the depth of the molten metal in the printed component or substrate during additive manufacturing or welding. Like the morphology and contour of the molten pool, the depth of the molten pool directly affects the quality and performance of the printed component. Therefore, the detection of melt depth is of great significance in additive manufacturing processes.

[0003] However, there are not many methods for detecting melt depth. This is mainly because the opaque molten pool makes melt depth, unlike surface features such as width, length, area, and temperature, impossible to detect directly using process monitoring tools. Currently, melt depth detection can be mainly divided into two categories: internal perspective measurement methods and hybrid methods combining surface feature measurement with correlation models.

[0004] Regarding internal spectral imaging, some researchers have primarily used high-speed synchronous X-ray imaging to observe the fluid dynamics of the molten pool under simulated processes such as WAAM, LPBF, and DED, providing intuitive and abundant melt depth data. However, due to limitations in economic cost, technical difficulty, and application scope, it is unlikely that this internal spectral imaging method, based on high-speed synchronous X-ray imaging technology, will be widely used in industry as an online melt depth detection tool.

[0005] Hybrid methods combining surface feature measurement and theoretical calculation are widely used. Some researchers have established correlation models between melt depth and molten pool oscillation frequency or geometric dimensions under LPBF process through experimental results or simulation calculations. However, due to limitations in experimental conditions or the completeness of theoretical derivations, the main parameters that determine the melt depth have not been fully considered, resulting in limitations in the established correlation models. Some researchers have established correlation models between melt depth and two-dimensional images of the molten pool through deep learning algorithms, in order to obtain melt depth information by inputting the acquired two-dimensional images into the model. However, the training of the model requires a large amount of data, which results in a large workload, and the final model is not very universal.

[0006] In other words, current methods for detecting melt depth can be mainly divided into two categories: internal perspective measurement and hybrid methods that combine surface feature measurement with correlation models.

[0007] Regarding internal spectral imaging, some researchers have primarily used high-speed synchronous X-ray imaging to observe the fluid dynamics of the molten pool under simulated processes such as WAAM, LPBF, and DED, providing intuitive and abundant melt depth data. However, due to limitations in economic cost, technical difficulty, and application scope, it is unlikely that this internal spectral imaging method, based on high-speed synchronous X-ray imaging technology, will be widely used in industry as an online melt depth detection tool.

[0008] Hybrid methods combining surface feature measurement and theoretical calculation are widely used. Some researchers have established correlation models between melt depth and molten pool oscillation frequency or geometric dimensions under LPBF process through experimental results or simulation calculations. However, due to limitations in experimental conditions or the completeness of theoretical derivations, the main parameters that determine the melt depth have not been fully considered, resulting in limitations in the established correlation models. Some researchers have established correlation models between melt depth and two-dimensional images of the molten pool through deep learning algorithms, in order to obtain melt depth information by inputting the acquired two-dimensional images into the model. However, the training of the model requires a large amount of data, which results in a large workload, and the final model is not very universal.

[0009] Therefore, in order to address the above problems, developing an efficient, accurate, and widely applicable online melt depth detection system and method is an urgent issue to be solved in the field of melt depth measurement technology for metal additive manufacturing. Summary of the Invention

[0010] To address the aforementioned issues, this invention provides an online melting depth detection system based on the oscillation frequency and radius of the molten pool. A model is established to show the relationship between the melting depth of the molten pool, the intrinsic oscillation frequency of the molten pool, and the equivalent radius of the molten pool during the metal additive manufacturing process. This system enables real-time online detection of the melting depth of the molten pool during the manufacturing process.

[0011] An online melt depth detection system based on molten pool oscillation frequency and radius includes a photoelectric signal acquisition module, a molten pool temperature field measurement module, and a processing module;

[0012] The photoelectric signal acquisition module is used to acquire the light intensity signal of the molten pool radiation light, and to acquire the intrinsic oscillation frequency f of the molten pool based on the time-frequency information of the light intensity signal.

[0013] The molten pool temperature field measurement module is used to obtain the equivalent radius R of the molten pool based on the colorimetric thermometry principle;

[0014] The processing module is used to obtain the melt depth d based on the intrinsic oscillation frequency f and the equivalent radius R of the molten pool, wherein the formula for calculating the melt depth d is as follows:

[0015]

[0016] Where g is the acceleration due to gravity, ρ is the density of the molten pool, and γ is the surface tension of the molten pool.

[0017] Furthermore, the photoelectric signal acquisition module includes a photodetector, a data acquisition card, a narrowband filter, and a long working distance microscope lens;

[0018] The long working distance microscope lens is used to focus the light from the molten pool scene onto the narrowband filter.

[0019] The narrowband filter is used to filter the background light in the molten pool scene light to obtain the molten pool radiation light;

[0020] The photodetector is used to convert the molten pool radiation light emitted from the narrowband filter into a molten pool radiation electrical signal, wherein the light intensity of the molten pool radiation light is proportional to the amplitude of the molten pool radiation electrical signal.

[0021] The data acquisition card is used to estimate the time-frequency information of the radiated electrical signal of the molten pool by wavelet transform or short-time Fourier transform, and the frequency in the time-frequency information is used as the intrinsic oscillation frequency f of the molten pool.

[0022] Furthermore, an online melt depth detection system based on molten pool oscillation frequency and radius also includes an aluminum profile structure;

[0023] The aluminum profile frame is used to fix the long focal length microscope lens, the dual-channel spectral filtering device, the CMOS camera, and the long working distance microscope lens.

[0024] Furthermore, an online melt depth detection system based on the oscillation frequency and radius of the molten pool also includes a three-dimensional fine-tuning translation stage;

[0025] The three-dimensional fine-tuning translation stage is used to adjust the distance between the long working distance microscope head and the molten pool, so that the molten pool is located in the middle imaging area of ​​the long working distance microscope head.

[0026] Furthermore, the molten pool temperature field measurement module includes a calculation module, a CMOS camera, a long-focal-length microscope lens, and a dual-channel spectroscopic filtering device;

[0027] The dual-channel beam splitter is used to filter the molten pool radiation light to obtain two specific near-infrared band images in the molten pool radiation light.

[0028] The long-focal-length microscope lens is used to magnify and focus the two specific near-infrared band images onto the CMOS camera;

[0029] The CMOS camera is used to image the two specific near-infrared bands to obtain the molten pool image;

[0030] The calculation module is used to extract the molten pool outline from the molten pool image based on the colorimetric thermometry principle, determine the molten pool area based on the number of pixels in the molten pool outline, and then solve the equivalent radius of the molten pool based on the molten pool area.

[0031] Beneficial effects:

[0032] This invention provides an online melt depth detection system based on molten pool oscillation frequency and radius. It uses a photoelectric signal acquisition module and a molten pool temperature field measurement module to monitor the molten pool photoelectric signal and temperature field in real time, thereby calculating the equivalent radius and intrinsic oscillation frequency of the molten pool in real time. Furthermore, it establishes a relationship model between the melt depth, intrinsic oscillation frequency, and equivalent radius of the molten pool in the metal additive manufacturing process, thus achieving real-time online detection of the melt depth during manufacturing. Attached Figure Description

[0033] Figure 1 This invention provides an online melt depth detection process based on the oscillation frequency and radius of the molten pool;

[0034] Figure 2 The present invention provides a principle block diagram of an online melt depth detection system based on the oscillation frequency and radius of the molten pool. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0036] An online melt depth detection system based on molten pool oscillation frequency and radius includes a photoelectric signal acquisition module, a molten pool temperature field measurement module, and a processing module;

[0037] The photoelectric signal acquisition module is used to acquire the light intensity signal of the molten pool radiation light, and to acquire the intrinsic oscillation frequency f of the molten pool based on the time-frequency information of the light intensity signal.

[0038] The molten pool temperature field measurement module is used to obtain the equivalent radius R of the molten pool based on the colorimetric thermometry principle;

[0039] The processing module is used to obtain the melt depth d based on the intrinsic oscillation frequency f and the equivalent radius R of the molten pool, wherein the formula for calculating the melt depth d is as follows:

[0040]

[0041] Where g is the acceleration due to gravity, ρ is the density of the molten pool, and γ is the surface tension of the molten pool.

[0042] Therefore, this invention mainly measures key parameters of the molten pool through a photoelectric signal acquisition module and a high-speed molten pool temperature field measurement module. Based on the accurate detection of the molten pool's intrinsic oscillation frequency and equivalent radius, combined with real-time measurement of the molten pool temperature field, it achieves online detection of the melt depth. Figure 1 As shown, it includes the following steps:

[0043] Step 1: Establish the relationship between melt depth and intrinsic oscillation frequency and equivalent radius of the molten pool through formula derivation;

[0044] Step 2: Design an online melt depth detection platform. Use a photoelectric signal acquisition module with a determinable detection area to collect the radiation light signal of the molten pool and determine the intrinsic oscillation frequency of the molten pool.

[0045] Step 3: Use the high-speed measurement module for the molten pool temperature field to quickly acquire the molten pool temperature field and obtain the main two-dimensional dimensional information such as the molten pool area and equivalent radius;

[0046] Step 4: Experimentally verify the effectiveness of the intrinsic oscillation frequency of the molten pool measured by radiation light by building a system for detecting the intrinsic oscillation frequency of the molten pool based on specular reflection light and external excitation.

[0047] Step 5: Input the main parameters such as the intrinsic oscillation frequency and equivalent radius of the molten pool obtained in the above steps into the calculation module to obtain the molten depth result, and compare the metallographic section measurement results to verify the accuracy and effectiveness of the correlation model and development method.

[0048] The derivation process of the formula for calculating the melting depth d is described in detail below.

[0049] By establishing an oscillation model of the molten pool using the classical free surface wave theory of liquids, the formula for the intrinsic oscillation frequency of the molten pool can be obtained. Where g is the gravitational acceleration, λ is the wavelength, γ is the surface tension, ρ is the molten pool density, and d is the molten depth. Since there exists a relationship between wavelength λ and wavenumber k, k = 2π / λ. When the molten pool oscillation is in the sputtering oscillation mode, kR = 7.66, and the formula can be obtained. A formula for calculating melt depth based on intrinsic oscillation frequency and equivalent radius can be derived. During the stable phase of additive manufacturing, the surface tension is approximately constant. Therefore, the melt depth d is related to the intrinsic oscillation frequency f of the molten pool and the equivalent radius R of the molten pool.

[0050] It should be noted that the equivalent radius R of the molten pool in the melt depth calculation formula derived in this invention is calculated based on the molten pool area measured by the high-speed molten pool temperature field measurement module. The molten pool outline can be obtained through a gradient algorithm using the melting point and temperature field, and the molten pool outline area can be calculated from the number of pixels in the molten pool temperature field image. Treating the molten pool as a circle with radius R, the molten pool temperature field area is equal to πR.2 Calculate the equivalent radius of the molten pool.

[0051] Furthermore, such as Figure 2 As shown, the online melt depth detection system of the present invention includes an aluminum profile structure and a three-dimensional fine-tuning translation stage; the photoelectric signal acquisition module includes a photodetector, a data acquisition card, a 905nm narrowband filter, and a long working distance microscope lens; the melt pool temperature field measurement module includes a calculation module, a CMOS camera, a long focal length microscope lens, and a dual-channel beam splitting and filtering device; the photoelectric signal acquisition module and the melt pool temperature field measurement module are respectively connected to the processing module for signal storage and analysis.

[0052] The long working distance microscope lens is used to focus the light from the molten pool scene onto a narrowband filter; the narrowband filter is used to filter the background light from the molten pool scene light to obtain the molten pool radiation light; the photodetector is used to convert the molten pool radiation light emitted from the narrowband filter into a molten pool radiation electrical signal, wherein the intensity of the molten pool radiation light is proportional to the amplitude of the molten pool radiation electrical signal; the data acquisition card is used to estimate the time-frequency information of the molten pool radiation electrical signal using wavelet transform or short-time Fourier transform, and uses the frequency in the time-frequency information as the intrinsic oscillation frequency f of the molten pool.

[0053] The dual-channel beam splitter is used to filter the molten pool radiation light to obtain two specific near-infrared band images in the molten pool radiation light; the long-focal-length microscope lens is used to magnify and focus the two specific near-infrared band images onto the CMOS camera; the CMOS camera is used to image the two specific near-infrared band images to obtain the molten pool image; the calculation module is used to extract the molten pool contour from the molten pool image according to the colorimetric thermometry principle, determine the molten pool area according to the number of pixels in the molten pool contour, and then solve for the equivalent radius of the molten pool based on the molten pool area.

[0054] The aluminum profile frame is used to fix the long focal length microscope lens, the dual-channel spectral filtering device, the CMOS camera, and the long working distance microscope lens.

[0055] The three-dimensional fine-tuning translation stage is used to adjust the distance between the long working distance microscope head and the molten pool, so that the molten pool is located in the middle imaging area of ​​the long working distance microscope head.

[0056] Optionally, the angle between the photoelectric signal acquisition module and the horizontal direction of the working platform is 45°, and the distance between the long working distance microscope head and the substrate is 0.65m. A three-dimensional fine-tuning translation stage is added between the aluminum profile structure and the long working distance microscope head. This translation stage can be finely adjusted to ensure that the molten pool is in the middle region of the field of view of the long working distance microscope head. The photoelectric signal acquisition module acquires the light intensity signal of the 905nm band radiation light of the molten pool, and estimates the spectrum information of the molten pool radiation light through Fourier transform or Thomson multi-cone method to obtain the intrinsic oscillation frequency of the molten pool. The high-speed measurement module of the molten pool temperature field uses the colorimetric thermometry principle to process the dual-channel molten pool grayscale image into the molten pool temperature field. After quickly acquiring the molten pool temperature field, the main parameters such as the area and equivalent radius of the molten pool are analyzed.

[0057] In summary, this invention provides an online melt depth detection system based on the oscillation frequency and radius of the molten pool. It uses a photoelectric signal acquisition module and a molten pool temperature field measurement module to monitor the photoelectric signal and temperature field of the molten pool in real time, thereby calculating the equivalent radius and intrinsic oscillation frequency of the molten pool in real time. Furthermore, a relationship model between the melt depth, intrinsic oscillation frequency, and equivalent radius of the molten pool in the metal additive manufacturing process is established, thus enabling real-time online detection of the melt depth during the manufacturing process.

[0058] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. An online melt depth detection system based on molten pool oscillation frequency and radius, characterized in that, It includes a photoelectric signal acquisition module, a molten pool temperature field measurement module, and a processing module; The photoelectric signal acquisition module is used to acquire the light intensity signal of the molten pool radiation, and to obtain the intrinsic oscillation frequency of the molten pool based on the time-frequency information of the light intensity signal. ; The molten pool temperature field measurement module is used to obtain the equivalent radius of the molten pool based on the colorimetric thermometry principle. ; The processing module is used to determine the intrinsic oscillation frequency of the molten pool. and equivalent radius Obtaining the depth of penetration Among them, melting depth The calculation formula is as follows: in, It is the acceleration due to gravity. The density of the molten pool, The surface tension of the molten pool; The molten pool temperature field measurement module includes a calculation module, a CMOS camera, a long-focal-length microscope lens, and a dual-channel spectroscopic filter device. The dual-channel beam splitter is used to filter the molten pool radiation light to obtain two specific near-infrared band images in the molten pool radiation light. The long-focal-length microscope lens is used to magnify and focus the two specific near-infrared band images onto the CMOS camera; The CMOS camera is used to image the two specific near-infrared bands to obtain the molten pool image; The calculation module is used to extract the molten pool outline from the molten pool image based on the colorimetric thermometry principle, determine the molten pool area based on the number of pixels in the molten pool outline, and then solve the equivalent radius of the molten pool based on the molten pool area.

2. The online melt depth detection system based on molten pool oscillation frequency and radius as described in claim 1, characterized in that, The photoelectric signal acquisition module includes a photodetector, a data acquisition card, a narrowband filter, and a long working distance microscope lens; The long working distance microscope lens is used to focus the light from the molten pool scene onto the narrowband filter. The narrowband filter is used to filter the background light in the molten pool scene light to obtain the molten pool radiation light; The photodetector is used to convert the molten pool radiation light emitted from the narrowband filter into a molten pool radiation electrical signal, wherein the light intensity of the molten pool radiation light is proportional to the amplitude of the molten pool radiation electrical signal. The data acquisition card is used to estimate the time-frequency information of the molten pool's radiated electrical signal using wavelet transform or short-time Fourier transform, and uses the frequency in the time-frequency information as the intrinsic oscillation frequency of the molten pool. .

3. The online melt depth detection system based on molten pool oscillation frequency and radius as described in claim 2, characterized in that, It also includes aluminum profile structures; The aluminum profile frame is used to fix the long focal length microscope lens, the dual-channel spectral filtering device, the CMOS camera, and the long working distance microscope lens.

4. The online melt depth detection system based on molten pool oscillation frequency and radius as described in claim 2, characterized in that, It also includes a three-dimensional fine-tuning translation stage; The three-dimensional fine-tuning translation stage is used to adjust the distance between the long working distance microscope head and the molten pool, so that the molten pool is located in the middle imaging area of ​​the long working distance microscope head.

Images