Titanium alloy smelting method and system based on smelting image analysis

Abstract

The invention relates to the technical field of smelting image analysis, in particular to a titanium alloy smelting method and system based on smelting image analysis, and the method comprises the steps that a multispectral real-time imaging system is built, titanium alloy, a vacuum arc furnace and smelting parameters are obtained, three sections of smelting parameters are obtained, and a first section of smelting parameter is sequentially extracted from the three sections of smelting parameters; carrying out the following operations on the extracted first-section smelting parameters: carrying out first-time smelting operation on the titanium alloy, recording an initial molten pool liquid level in real time, shooting a molten pool to obtain a near-infrared image set, a visible light image set and an X-ray image set, obtaining a temperature field, a liquid drop velocity vector and a binary mask, and obtaining three-section molten pool liquid levels; and titanium alloy smelting based on smelting image analysis is completed based on the three-section molten pool liquid level, the temperature field set, the liquid drop velocity vector set and the binary mask set. According to the invention, image analysis can be rapidly and accurately carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of smelting image analysis, and in particular to a titanium alloy smelting method and system based on smelting image analysis. Background Art

[0002] Melting image analysis is a method that uses advanced image processing and analysis technology to conduct in-depth research on images collected during the titanium alloy smelting process to obtain relevant information about the smelting process. Titanium alloy smelting refers to the process of mixing titanium and other alloy elements in a certain proportion, heating them to a molten state, and refining and solidifying them under specific conditions to obtain titanium alloy materials with the required properties.

[0003] The existing image analysis systems mainly focus on the image acquisition and analysis of a single spectrum or a few spectra, and cannot fully reflect the key information such as temperature distribution, material transfer, chemical reaction, etc. in the molten pool. At the same time, the existing image analysis methods are difficult to meet the needs of real-time monitoring and control of titanium alloy smelting in terms of processing speed and analysis accuracy. Therefore, how to achieve fast and accurate image analysis. Summary of the invention

[0004] The present invention provides a titanium alloy smelting method based on smelting image analysis and a computer-readable storage medium, the main purpose of which is to achieve rapid and accurate image analysis.

[0005] To achieve the above-mentioned purpose, the present invention provides a titanium alloy smelting method based on smelting image analysis, comprising: building a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system comprises: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system; obtaining titanium alloy, a vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace comprises: a smelting observation window and a molten pool; obtaining three-stage smelting parameters according to the smelting parameters, wherein the three-stage smelting parameters comprise: a first-stage smelting parameter, a second-stage smelting parameter and a third-stage smelting parameter, wherein the smelting parameters comprise: an electrical range, a gas range and a temperature range; extracting the first-stage smelting parameters from the three-stage smelting parameters in turn, and performing the following operations on the extracted first-stage smelting parameters: using the vacuum arc furnace and the first-stage smelting parameters to obtain the titanium alloy, the vacuum arc furnace and the first-stage smelting parameters; obtaining the titanium alloy, the vacuum arc furnace and the first-stage smelting parameters; obtaining the titanium alloy, the vacuum arc furnace and the first-stage smelting parameters; wherein the vacuum arc furnace and the first-stage smelting parameters are used to obtain ... Perform the first melting operation on several pairs of titanium alloys, and record the initial molten pool level in real time; use the multi-spectral real-time imaging system and the melting observation window to photograph the molten pool to obtain a near-infrared image set, a visible light image set and an X-ray image set; receive a melting image analysis instruction, and obtain a temperature field, a droplet velocity vector and a binary mask according to the melting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set; summarize the temperature field, the droplet velocity vector and the binary mask respectively to obtain a temperature field set, a droplet velocity vector set and a binary mask set corresponding to the three-stage melting parameters; obtain the three-stage molten pool level based on the three-stage melting parameters and the initial molten pool level; complete the titanium alloy melting based on the melting image analysis based on the three-stage molten pool level, the temperature field set, the droplet velocity vector set and the binary mask set.

[0006] Optionally, the three-stage smelting parameters are obtained according to the smelting parameters, including: obtaining the upper limit of the first section interval and the upper limit of the second section interval; using the upper limit of the first section interval and the upper limit of the second section interval to perform three-stage operation on the electrical interval, gas interval and temperature interval in the smelting parameters, and obtain an electrical segmented interval group, a gas segmented interval group and a temperature segmented interval group, wherein the electrical segmented interval group includes: a first electrical interval, a second electrical interval and a third electrical interval, the gas segmented interval group includes: a first gas interval, a second gas interval and a third gas interval, and the temperature segmented interval group includes: a first temperature interval, a second temperature interval and a third temperature interval; using the first electrical interval, the first gas interval and the first temperature interval as the first-stage smelting parameters, using the second electrical interval, the second gas interval and the second temperature interval as the second-stage smelting parameters, and using the third electrical interval, the third gas interval and the third temperature interval as the third-stage smelting parameters; confirming the first-stage smelting parameters, the second-stage smelting parameters and the third-stage smelting parameters as three-stage smelting parameters.

[0007] Optionally, obtaining the upper limit of the first interval and the upper limit of the second interval includes: calculating the upper limit of the first interval and the upper limit of the second interval using a pre-constructed interval segmentation formula, wherein the interval segmentation formula is as follows: , ,in, Indicates the upper limit of the first interval, Indicates the upper limit of the second interval, Indicates the minimum value of the preset interval. Indicates the maximum value of the preset interval. Indicates the preset time adjustment parameters. Indicates the time value in the preset time interval. Indicates the minimum value of the preset time interval. Indicates the maximum value of the preset time interval.

[0008] Optionally, the method of acquiring the temperature field, the droplet velocity vector and the binary mask according to the melting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set includes: performing a normalization operation on all near-infrared images in the near-infrared image set using the melting image analysis instruction to obtain a near-infrared normalized atlas; performing the following operations on each visible light image in the visible light image set: acquiring a pixel value set of the visible light image, dynamically compressing each pixel value in the pixel value set using a pre-built logarithmic compression method to obtain a logarithmic compressed pixel value set, wherein the logarithmic compressed pixel value set includes multiple logarithmic compressed pixel values; extracting the maximum pixel value and the minimum pixel value from the logarithmic compressed pixel value set, and using the maximum pixel value, the minimum pixel value A pixel value, a preset pixel value interval and a pre-constructed linear stretching method are used to perform a linear mapping operation on each logarithmic compressed pixel value in the logarithmic compressed pixel value set to obtain a mapped pixel value set; a data type conversion operation is performed on each mapped pixel value in the mapped pixel value set to obtain a visible light compression map; the visible light compression map is summarized to obtain a visible light compression atlas corresponding to the visible light image set, wherein the visible light compression map is an image corresponding to a frame in the visible light image set; a correction and suppression operation is performed on all X-ray images in the X-ray image set to obtain an X-ray correction atlas; the temperature field is obtained based on the near-infrared normalized atlas, the droplet velocity vector is obtained based on the visible light compression atlas, and the binary mask is obtained based on the X-ray correction atlas.

[0009] Optionally, the method of acquiring the temperature field based on the near-infrared normalized atlas includes: extracting near-infrared normalized images from the near-infrared normalized atlas in sequence, and performing the following operations on the extracted near-infrared normalized images: gridding the near-infrared normalized images using a preset pixel specification to obtain a grid area set, obtaining a molten pool surface fluctuation model, and using the molten pool surface fluctuation model to calculate the molten pool surface state corresponding to each grid area in the grid area set to obtain a molten pool surface state set; using the molten pool surface state set to adjust the pre-constructed emissivity value of each grid area in the grid area set to obtain an optimized grid area set, and calculating the optimized The temperature value of each optimized grid area in the grid area set is calculated, and the temperature values ​​are collected to obtain an accurate temperature value set; the number of grids in the optimized grid area set and the number of temperature values ​​in the accurate temperature value set are obtained, and it is determined whether the number of grids is greater than the number of temperature values; if the number of grids is greater than the number of temperature values, the optimized grid area is located to obtain an abnormal grid area, and spatial interpolation is performed on the abnormal grid area to obtain an interpolated temperature value, wherein the abnormal grid area includes: one abnormal grid area or multiple abnormal grid areas; the accurate temperature value set and the interpolated temperature value are respectively summarized to obtain a temperature field corresponding to the near-infrared normalized atlas.

[0010] Optionally, the spatial interpolation of the abnormal grid area to obtain the interpolated temperature value includes: obtaining a window grid set according to a preset grid window and the abnormal grid area, calculating the distance between the abnormal grid area and each window grid in the window grid set, and obtaining a grid distance set; using a pre-constructed radial basis function to calculate the weight of each grid distance in the grid distance set, and obtaining a distance weight set; calculating the average temperature value according to the distance weight set and the precise temperature value set corresponding to the window grid set, wherein the calculation formula of the average temperature value is as follows: ,in, represents the average temperature value, Indicates the number of window grid sets, represents the index of the window grid set, Indicates the preset The measurement error correction factor of the window grid, Indicates The exact temperature value of the window grid, Indicates The distance weight of the window grid, Indicates the preset The uncertainty weight factor of the window grid; the average temperature value is used as the interpolated temperature value.

[0011] Optionally, the method of obtaining a droplet velocity vector based on a visible light compression atlas includes: extracting a current frame from the visible light compression atlas in sequence, and performing the following operations on each of the extracted current frames: obtaining a current feature point set based on the extracted current frame, extracting the next current frame adjacent to the extracted current frame to obtain a post-frame, and obtaining a post-feature point set of a visible light compression map corresponding to the post-frame; performing feature point matching on the current feature point set and the post-feature point set to obtain an initial feature point pair group; performing error matching screening on the initial feature point pair group to obtain a feature point pair group, and performing the following operations on each feature point pair in the feature point pair group: calculating the coordinate difference of the feature point pair, obtaining the time interval between the current frame and the post-frame, and obtaining a velocity vector based on the coordinate difference and the time interval; summarizing the velocity vectors to obtain a velocity vector set corresponding to the feature point pair group, and summarizing the velocity vector set corresponding to the visible light compression atlas to obtain multiple velocity vector sets; and calculating the average value of the multiple velocity vector sets to obtain a droplet velocity vector.

[0012] Optionally, performing error matching screening on the initial feature point pair group to obtain the feature point pair group includes: obtaining the relative displacement of each pair of initial feature point pairs in the initial feature point pair group to obtain a relative displacement set, and obtaining the displacement standard deviation according to the relative displacement set; judging whether there is a relative displacement greater than the displacement standard deviation in the relative displacement set; if it is confirmed that there is a relative displacement greater than the displacement standard deviation in the relative displacement set, then eliminating the initial feature point pair corresponding to the relative displacement to obtain an updated feature point pair group; extracting an updated feature point pair from the updated feature point pair group in turn, and performing the following operations on the extracted updated feature point pairs: obtaining the first local area window of the updated feature point pair using a preset local area window; area and a second local area; obtaining a first local texture feature and a second local texture feature according to the first local area and the second local area; comparing the first local texture feature with the second local texture feature; if the first local texture feature is the same as the second local texture feature, obtaining a feature point pair; if the first local texture feature is different from the second local texture feature, eliminating the updated feature point pair, and summarizing the retained updated feature point pairs to obtain an optimized feature point pair group, taking the optimized feature point pair group as the updated feature point pair group, returning to the step of extracting an updated feature point pair from the updated feature point pair group in sequence; summarizing the feature point pairs to obtain a feature point pair group corresponding to the updated feature point pair group.

[0013] Optionally, obtaining the three-stage molten pool liquid level based on the three-stage smelting parameters and the initial molten pool liquid level includes: calculating the three-stage molten pool liquid level according to the three-stage smelting parameters and the initial molten pool liquid level, wherein the calculation formula of the three-stage molten pool liquid level is as follows: , , ,in, The molten pool level represents the first stage smelting parameters, represents the initial molten pool level, Indicates the preset current, Indicates the preset gas flow rate, Indicates the preset impurity content, Indicates the preset melting temperature, The molten pool level represents the second stage smelting parameters, The molten pool level represents the third stage smelting parameters. They represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level in the first stage of smelting parameters. , , , They represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level in the second stage smelting parameters. , , , They respectively represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level within the third stage smelting parameters. Represents the natural logarithm.

[0014] To achieve the above-mentioned purpose, the present invention also provides a titanium alloy smelting system based on smelting image analysis, including: a smelting parameter preparation module, used to build a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system, to obtain titanium alloy, vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace includes: a smelting observation window and a molten pool, and three-stage smelting parameters are obtained according to the smelting parameters, wherein the three-stage smelting parameters include: first-stage smelting parameters, second-stage smelting parameters and third-stage smelting parameters, wherein the smelting parameters include: an electrical range, a gas range and a temperature range; a smelting image acquisition module, used to extract the first-stage smelting parameters from the three-stage smelting parameters in turn, and perform the following operations on the extracted first-stage smelting parameters: using the vacuum arc furnace and the first-stage smelting The first smelting operation is performed on the titanium alloy according to the parameters, and the initial molten pool liquid level is recorded in real time, and the molten pool is photographed by using the multi-spectral real-time imaging system and the smelting observation window to obtain a near-infrared image set, a visible light image set and an X-ray image set; a smelting image analysis module is used to receive a smelting image analysis instruction, and obtain a temperature field, a droplet velocity vector and a binary mask according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set, and respectively summarize the temperature field, the droplet velocity vector and the binary mask to obtain a temperature field set, a droplet velocity vector set and a binary mask set corresponding to the three-stage smelting parameters; a smelting completion module is used to obtain a three-stage molten pool liquid level based on the three-stage smelting parameters and the initial molten pool liquid level, and complete the titanium alloy smelting based on the smelting image analysis based on the three-stage molten pool liquid level, the temperature field set, the droplet velocity vector set and the binary mask set.

[0015] In order to solve the above problems, the present invention also provides an electronic device, which includes: a memory storing at least one instruction; and a processor executing the instructions stored in the memory to implement the above-mentioned titanium alloy smelting method based on smelting image analysis.

[0016] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned titanium alloy smelting method based on smelting image analysis.

[0017] The present invention solves the problems described in the background technology. The present invention builds a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system. The present invention can obtain information about the molten pool from multiple dimensions through imaging systems of different spectra. The near-infrared pyrometer array can measure the temperature distribution on the surface of the molten pool. The visible light high dynamic range imaging system can clearly capture the appearance of the molten pool, liquid level fluctuations, etc. The X-ray real-time imaging system can penetrate the molten pool to detect internal structures and defects. This multi-spectral real-time imaging system can provide more comprehensive and detailed molten pool information, which is helpful for accurately analyzing the smelting process and obtaining titanium alloys, vacuum arc furnaces and smelting parameters. The electric arc furnace comprises: a smelting observation window and a molten pool. The present invention can reduce the reaction between the titanium alloy and impurities in the air during the smelting process through the use of a vacuum electric arc furnace, thereby improving the smelting quality. Three-stage smelting parameters are obtained according to the smelting parameters, wherein the three-stage smelting parameters include: first-stage smelting parameters, second-stage smelting parameters and third-stage smelting parameters, wherein the smelting parameters include: an electrical range, a gas range and a temperature range. The present invention divides the entire smelting process into three sections, and sets different smelting parameters respectively. The smelting process can be controlled more accurately according to the characteristics and requirements of the titanium alloy at different smelting stages, which is helpful to improve the purity, uniformity and performance stability of the titanium alloy. The first-stage smelting parameters are extracted from the three-stage smelting parameters in turn, and the extracted first-stage smelting parameters are processed. The following operations are performed: a first smelting operation is performed on a titanium alloy using a vacuum arc furnace and the first stage smelting parameters, and the initial molten pool liquid level is recorded in real time. The present invention provides a benchmark for subsequent monitoring of changes in the molten pool liquid level by recording the initial molten pool liquid level in real time, thereby ensuring that the liquid in the molten pool does not overflow; the molten pool is photographed using the multi-spectral real-time imaging system and the smelting observation window to obtain a near-infrared image set, a visible light image set and an X-ray image set. The present invention photographs the molten pool image using the multi-spectral real-time imaging system, and can obtain various information about the molten pool at this smelting stage, so that operators can observe the smelting process in real time and discover potential problems in a timely manner; a smelting image analysis instruction is received, and the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set are detected according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set are detected. The temperature field, droplet velocity vector and binary mask are obtained from the line image set. The present invention can help understand the temperature distribution inside the molten pool through the temperature field, and provide a basis for adjusting the temperature range. The droplet velocity vector can reflect the flow of materials in the molten pool, which helps to optimize the convection and mixing during the smelting process. The binary mask can be used to identify different materials or defects in the molten pool, which is convenient for quality control. The temperature field, droplet velocity vector and binary mask are summarized respectively to obtain the temperature field set, droplet velocity vector set and binary mask set corresponding to the three-stage smelting parameters. The present invention summarizes the temperature field, droplet velocity vector and binary mask obtained in each smelting process to form a complete data set, so that the entire smelting process can be comprehensively analyzed and evaluated.Understand the changes and mutual relationships in different stages, obtain the three-stage molten pool level based on the three-stage smelting parameters and the initial molten pool level. The present invention can reflect the material balance and reaction conditions in the smelting process through the changes in the molten pool level, which is helpful to judge whether the smelting process is stable. The smelting of titanium alloy based on smelting image analysis is completed based on the three-stage molten pool level, temperature field set, droplet velocity vector set and binary mask set. The present invention can improve the purity, uniformity and performance stability of titanium alloys through precise control based on image analysis, reduce the generation of defects and impurities, and thus improve the quality and reliability of titanium alloy products. Therefore, the present invention can realize rapid and accurate image analysis. , BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of a process of a titanium alloy smelting method based on smelting image analysis provided by one embodiment of the present invention; Figure 2 A functional module diagram of a titanium alloy smelting system based on smelting image analysis provided by one embodiment of the present invention; Figure 3 A schematic structural diagram of an electronic device for implementing the titanium alloy smelting method based on smelting image analysis provided in one embodiment of the present invention.

[0019] Explanation of the reference numerals: 1. electronic device; 10. processor; 11. memory; 12. bus.

[0020] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0022] The embodiment of the present application provides a titanium alloy smelting method based on smelting image analysis. The execution subject of the titanium alloy smelting method based on smelting image analysis includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the titanium alloy smelting method based on smelting image analysis can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0023] Reference Figure 1 As shown, it is a schematic diagram of a process of a titanium alloy smelting method based on smelting image analysis provided by an embodiment of the present invention. In this embodiment, the titanium alloy smelting method based on smelting image analysis includes: S1, building a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system.

[0024] It should be explained that the construction of a multi-spectral real-time imaging system refers to the integration of a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system to form a system. The near-infrared pyrometer array is an array composed of multiple near-infrared pyrometers arranged and combined according to certain rules, which is used to measure the temperature distribution on the surface of the molten pool. A near-infrared pyrometer refers to a device that measures temperature based on the principle of infrared radiation. The visible light high dynamic range is used to obtain the spectral range of high-contrast images of the molten pool in the visible light band. The X-ray real-time imaging system refers to a system used to obtain structural information inside the molten pool.

[0025] S2. Acquire titanium alloy, vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace includes: a smelting observation window and a molten pool.

[0026] It should be explained that titanium alloy refers to an alloy composed of titanium and other alloying elements. For example, other alloying elements are aluminum, vanadium, molybdenum, etc. A vacuum arc furnace refers to a device that uses the high temperature generated by an electric arc to melt metal in a vacuum environment. Melting parameters refer to the parameters that need to be controlled and adjusted during the melting process of titanium alloys, which have a vital impact on the stability of the melting process, the quality and properties of the metal. The melting observation window refers to a window on a vacuum arc furnace for taking images of the molten pool. The molten pool refers to the liquid metal area formed after the titanium alloy is heated and melted in a vacuum arc furnace.

[0027] S3. Acquire three-stage smelting parameters according to the smelting parameters, wherein the three-stage smelting parameters include: first-stage smelting parameters, second-stage smelting parameters and third-stage smelting parameters, wherein the smelting parameters include: electrical range, gas range and temperature range.

[0028] In detail, the three-stage smelting parameters are obtained according to the smelting parameters, including: obtaining the upper limit of the first section interval and the upper limit of the second section interval; using the upper limit of the first section interval and the upper limit of the second section interval to perform three-stage operation on the electrical interval, gas interval and temperature interval in the smelting parameters, and obtain an electrical segmented interval group, a gas segmented interval group and a temperature segmented interval group, wherein the electrical segmented interval group includes: a first electrical interval, a second electrical interval and a third electrical interval, the gas segmented interval group includes: a first gas interval, a second gas interval and a third gas interval, and the temperature segmented interval group includes: a first temperature interval, a second temperature interval and a third temperature interval; the first electrical interval, the first gas interval and the first temperature interval are used as the first-stage smelting parameters, the second electrical interval, the second gas interval and the second temperature interval are used as the second-stage smelting parameters, and the third electrical interval, the third gas interval and the third temperature interval are used as the third-stage smelting parameters; the first-stage smelting parameters, the second-stage smelting parameters and the third-stage smelting parameters are confirmed as three-stage smelting parameters.

[0029] It should be explained that the upper limit of the first interval is the limit of the first interval division point used to divide the electrical interval, gas interval and temperature interval in the smelting parameters. The upper limit of the second interval is the limit of the second interval division point used to divide the electrical interval, gas interval and temperature interval in the smelting parameters. The electrical interval refers to the value range of electrical-related parameters during the vacuum arc furnace smelting process. For example, the electrical-related parameters are: current. The gas interval refers to the value range of the volume of the combustion gas introduced during the vacuum arc furnace smelting process. The temperature interval refers to the value range of the furnace temperature during the smelting process.

[0030] It should be explained that the three-stage interval operation refers to an operation that divides the interval into three sections. The electrical segmented interval group refers to a group of sub-intervals obtained by dividing the electrical interval after the three-stage interval operation. The gas segmented interval group refers to a group of sub-intervals obtained after the gas interval is subjected to the three-stage interval operation. The temperature segmented interval group refers to a group of sub-intervals obtained after the temperature interval is subjected to the three-stage interval operation. The first-stage melting parameters refer to the parameters composed of the first electrical interval, the first gas interval and the first temperature interval. The second-stage melting parameters refer to the parameters composed of the second electrical interval, the second gas interval and the second temperature interval. The third-stage melting parameters refer to the parameters composed of the third electrical interval, the third gas interval and the third temperature interval.

[0031] In detail, obtaining the upper limit of the first interval and the upper limit of the second interval includes: using a pre-constructed interval segmentation formula to calculate the upper limit of the first interval and the upper limit of the second interval, wherein the interval segmentation formula is as follows: , ,in, Indicates the upper limit of the first interval, Indicates the upper limit of the second interval, Indicates the minimum value of the preset interval. Indicates the maximum value of the preset interval. Indicates the preset time adjustment parameters. Indicates the time value in the preset time interval. Indicates the minimum value of the preset time interval. Indicates the maximum value of the preset time interval.

[0032] It should be explained that the minimum value of the interval refers to the lower limit value of the electrical interval, gas interval or temperature interval. The maximum value of the interval refers to the upper limit value of the electrical interval, gas interval or temperature interval. The time interval refers to the time range of the electrical interval, gas interval or temperature interval. The time value refers to a specific time value in the time interval.

[0033] S4. Extract the first-stage melting parameters from the three-stage melting parameters in sequence, and perform the following operations on the extracted first-stage melting parameters: perform the first melting operation on the titanium alloy using the vacuum arc furnace and the first-stage melting parameters, and record the initial molten pool level in real time.

[0034] It should be explained that the first smelting operation refers to the process of smelting titanium alloy using three-stage smelting parameters, in which the first stage smelting parameters are extracted from the three-stage smelting parameters in order, and then the smelting operation of titanium alloy is carried out by means of a vacuum arc furnace according to the first electrical range, the first gas range and the first temperature range specified by the first stage smelting parameters. The initial molten pool level refers to the molten pool level at the first moment obtained by real-time recording of the molten pool level during the first smelting operation.

[0035] S5. Use the multi-spectral real-time imaging system and the smelting observation window to photograph the molten pool to obtain a near-infrared image set, a visible light image set and an X-ray image set.

[0036] It should be explained that the near-infrared image set refers to a set consisting of all near-infrared images, the visible light image set refers to a set consisting of visible light images, and the X-ray image set refers to a set consisting of X-ray images.

[0037] S6. Receive a melting image analysis instruction, and acquire a temperature field, a droplet velocity vector, and a binary mask according to the melting image analysis instruction, a near-infrared image set, a visible light image set, and an X-ray image set.

[0038] It should be explained that the smelting image analysis instruction is an instruction for triggering and guiding the analysis operation of the smelting related images.

[0039] For example, during the melting of titanium alloy in a vacuum arc furnace, the operator found that the temperature distribution was uneven or the droplet splashing was abnormal. In order to further analyze these problems, the operator sent a melting image analysis instruction to the system, requiring the system to obtain the temperature field, droplet velocity vector and binary mask from the previously collected near-infrared image set, visible light image set and X-ray image set, so as to have a deeper understanding of the physical phenomena in the melting process, find out possible problems and make adjustments.

[0040] In detail, the method of acquiring the temperature field, the droplet velocity vector and the binary mask according to the melting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set includes: using the melting image analysis instruction to perform a normalization operation on the near-infrared images in the near-infrared image set to obtain a near-infrared normalized atlas; performing the following operations on each visible light image in the visible light image set: obtaining a pixel value set of the visible light image, dynamically compressing each pixel value in the pixel value set using a pre-built logarithmic compression method to obtain a logarithmic compressed pixel value set, wherein the logarithmic compressed pixel value set includes multiple logarithmic compressed pixel values; extracting the maximum pixel value and the minimum pixel value from the logarithmic compressed pixel value set, and using the maximum pixel value and the minimum pixel value to obtain a logarithmic compressed pixel value set; A pixel value, a preset pixel value interval and a pre-constructed linear stretching method are used to perform a linear mapping operation on each logarithmic compressed pixel value in the logarithmic compressed pixel value set to obtain a mapped pixel value set; a data type conversion operation is performed on each mapped pixel value in the mapped pixel value set to obtain a visible light compression map; the visible light compression map is summarized to obtain a visible light compression atlas corresponding to the visible light image set, wherein the visible light compression map is an image corresponding to a frame in the visible light image set; a correction and suppression operation is performed on all X-ray images in the X-ray image set to obtain an X-ray correction atlas; the temperature field is obtained based on the near-infrared normalized atlas, the droplet velocity vector is obtained based on the visible light compression atlas, and the binary mask is obtained based on the X-ray correction atlas.

[0041] It should be explained that the normalization operation refers to an operation that maps data to a specific range. The use of the smelting image analysis instruction to perform a normalization operation on all near-infrared images in the near-infrared image set as described in the embodiment of the present invention is a prior art and will not be repeated here. The near-infrared normalized atlas refers to a set of near-infrared normalized atlases in which each near-infrared image in the near-infrared image set is converted to a uniform pixel value range after the normalization operation. The pixel value set for obtaining the visible light image refers to the visible light image, in which each visible light image is composed of multiple pixels, each pixel has its corresponding pixel value, and the pixel values ​​are summarized to obtain a pixel value set.

[0042] Importantly, the logarithmic compression method refers to a nonlinear data compression method. The step of dynamically compressing each pixel value in the pixel value set using the pre-constructed logarithmic compression method described in the embodiment of the present invention is a prior art and will not be repeated here. The logarithmic compression pixel value set refers to a new set of pixel values ​​obtained after the pixel value set of the visible light image is dynamically compressed using the logarithmic compression method. The pixel value interval refers to a pre-set standard pixel value range, which is used to determine the target range of linear mapping and map the pixel values ​​in the logarithmic compression pixel value set to this preset interval. The linear stretching method refers to a linear data transformation method, an algorithm for linearly mapping data from one range to another range. The step of performing a linear mapping operation on each logarithmic compression pixel value in the logarithmic compression pixel value set using the maximum pixel value, the minimum pixel value, the preset pixel value interval and the pre-constructed linear stretching method is: performing a linear mapping operation on each logarithmic compression pixel value in the logarithmic compression pixel value set using the following formula to obtain a mapped pixel value set: ,in, Represents the mapped pixel value, represents logarithmically compressed pixel values, represents the minimum pixel value, represents the maximum pixel value, Represents the maximum mapped pixel value in the mapped pixel value set, Represents the minimum mapped pixel value in the mapped pixel value set.

[0043] It should be explained that the mapped pixel value refers to the value obtained by mapping the logarithmic compressed pixel value using the linear stretching method. The mapped pixel value set refers to a set consisting of all mapped pixel values. The data type conversion operation refers to the operation of converting the mapped pixel values ​​in the mapped pixel value set from one data type to another data type. For example, converting the floating-point pixel value to an unsigned 8-bit integer. The visible light compression map refers to the process of recombining these processed pixel values ​​into an image after performing the data type conversion operation on the mapped pixel value set. The visible light compression atlas refers to a set consisting of all visible light compression maps. The steps of performing correction and suppression operations on all X-ray images in the X-ray image set described in the embodiment of the present invention are prior art and will not be repeated here. The X-ray correction map refers to the image obtained after performing the correction and suppression operation on the X-ray image. The X-ray correction atlas is a set consisting of X-ray correction maps. The detailed steps of obtaining the temperature field based on the near-infrared normalized atlas and obtaining the droplet velocity vector based on the visible light compression atlas will be given later. Acquiring a binary mask based on the X-ray correction atlas refers to performing a binarization operation on each X-ray correction atlas in the X-ray correction atlas, and using the X-ray correction atlas after the binarization operation as a binary mask.

[0044] In detail, the method of acquiring the temperature field based on the near-infrared normalized atlas includes: extracting near-infrared normalized images from the near-infrared normalized atlas in sequence, and performing the following operations on the extracted near-infrared normalized images: gridding the near-infrared normalized images using a preset pixel specification to obtain a grid area set, obtaining a molten pool surface fluctuation model, and calculating the molten pool surface state corresponding to each grid area in the grid area set using the molten pool surface fluctuation model to obtain a molten pool surface state set; adjusting the pre-constructed emissivity value of each grid area in the grid area set using the molten pool surface state set to obtain an optimized grid area set, and calculating the optimized The temperature value of each optimized grid area in the grid area set is calculated, and the temperature values ​​are collected to obtain an accurate temperature value set; the number of grids in the optimized grid area set and the number of temperature values ​​in the accurate temperature value set are obtained, and it is determined whether the number of grids is greater than the number of temperature values; if the number of grids is greater than the number of temperature values, the optimized grid area is located to obtain an abnormal grid area, and spatial interpolation is performed on the abnormal grid area to obtain an interpolated temperature value, wherein the abnormal grid area includes: one abnormal grid area or multiple abnormal grid areas; the accurate temperature value set and the interpolated temperature value are respectively summarized to obtain a temperature field corresponding to the near-infrared normalized atlas.

[0045] It should be explained that the pixel size refers to the size of the basic unit used to grid the near-infrared normalized image. For example, each grid area includes pixels. The grid area set refers to the set of all grid areas obtained after the near-infrared normalized image is gridded according to a preset pixel specification. The acquisition of the molten pool surface fluctuation model refers to taking a set of pictures of changes in the metal solution in the molten pool in real time during the smelting process, performing picture preprocessing on the change picture set to obtain a pre-adjusted picture set, and constructing a molten pool surface fluctuation model based on the pre-adjusted picture set. The change picture set refers to a set composed of change pictures. Picture preprocessing refers to the operation of denoising and filtering the change pictures. The pre-adjusted picture set refers to a set composed of pre-adjusted pictures obtained after the change picture set is preprocessed.

[0046] It should be explained that the step of using the molten pool surface fluctuation model to calculate the molten pool surface state corresponding to each grid area in the grid area set is: mapping the pixel coordinates of each grid area in the grid area set to the actual physical space of the molten pool surface, determining the relevant input parameters of the grid area on the molten pool surface (such as height, speed, etc.) according to the mapped pixel coordinates, substituting the determined input parameters into the molten pool surface fluctuation model, and obtaining the molten pool surface state corresponding to the grid area (such as the degree of surface undulation, flow rate, etc.) through the calculation of the model. The molten pool surface state set refers to the set of molten pool surface states corresponding to each grid area in the grid area set calculated by the molten pool surface fluctuation model. The emissivity value refers to the ability value of the molten pool surface to emit radiation. The optimized grid area set refers to the new grid area set obtained after adjusting the emissivity value of each grid area in the grid area set.

[0047] Importantly, the calculation of the temperature value of each optimized grid area in the optimized grid area set refers to calculating the temperature value of each optimized grid area in the optimized grid area set using a two-color temperature measurement formula, wherein the two-color temperature measurement formula is a prior art and will not be repeated here. The precise temperature value refers to the set of all temperature values ​​obtained after calculating the temperature value of each optimized grid area in the optimized grid area set. The positioning of the optimized grid area refers to finding the pixel coordinates of the optimized grid area that does not have a corresponding temperature value by traversing the optimized grid area set and the precise temperature value set. An abnormal grid area refers to a grid area in the optimized grid area set that does not have a corresponding precise temperature value.

[0048] In detail, the spatial interpolation of the abnormal grid area to obtain the interpolated temperature value includes: obtaining a window grid set according to a preset grid window and the abnormal grid area, calculating the distance between the abnormal grid area and each window grid in the window grid set, and obtaining a grid distance set; using a pre-constructed radial basis function to calculate the weight of each grid distance in the grid distance set, and obtaining a distance weight set; calculating the average temperature value according to the distance weight set and the precise temperature value set corresponding to the window grid set, wherein the calculation formula of the average temperature value is as follows: ,in, represents the average temperature value, Indicates the number of window grid sets, represents the index of the window grid set, Indicates the preset The measurement error correction factor of the window grid, Indicates The exact temperature value of the window grid, represents the distance weight of the th window grid, Indicates the preset The uncertainty weight factor of the window grid; the average temperature value is used as the interpolated temperature value.

[0049] It should be explained that the grid window refers to a pre-set square window. The acquisition of the window grid set according to the preset grid window and the abnormal grid area refers to taking the abnormal grid area as the center position of the grid window and acquiring the window grid set around the abnormal grid area. The window grid set refers to a set of grids selected around the abnormal grid area with the abnormal grid area as the center. The calculation of the distance between the abnormal grid area and each window grid in the window grid set refers to the use of the Euclidean distance formula to calculate the distance between the abnormal grid area and each window grid in the window grid set. The grid distance set refers to a set consisting of all grid distances. The radial basis function is a function used to determine the weight according to the distance between points. For example, Gaussian function, thin plate spline function, etc. The step of using the pre-constructed radial basis function to calculate the weight of each grid distance in the grid distance set and obtain the distance weight set is a prior art and will not be repeated here. The distance weight reflects the degree of influence of the distance between the window grid and the abnormal grid area on the interpolation temperature calculation. When we need to estimate the temperature value of an abnormal grid area, we cannot directly measure the temperature of the abnormal grid area, so we use the known precise temperature values ​​of the window grids around it to perform interpolation calculations. However, the contributions of different window grids to the temperature of the abnormal grid area are not the same. The closer the window grid is to the abnormal grid area, the greater the impact of its temperature information on the temperature of the abnormal grid area. Conversely, the farther the window grid is from the abnormal grid area, the smaller the impact of its temperature information on the temperature of the abnormal grid area.

[0050] It is understandable that the distance weight set refers to a set consisting of all distance weights. The measurement error correction coefficient refers to a preset one used to The uncertainty weight factor is the factor used to take into account the exact temperature value of the window grid. The uncertainty weight factor is a factor that weights the uncertainty of the exact temperature value of the window grid. The larger the uncertainty weight factor, the greater the uncertainty of the exact temperature value of the window grid.

[0051] In detail, the method of obtaining a droplet velocity vector based on a visible light compression atlas includes: extracting a current frame from the visible light compression atlas in sequence, and performing the following operations on each of the extracted current frames: obtaining a current feature point set based on the extracted current frame, extracting the next current frame adjacent to the extracted current frame to obtain a post-frame, and obtaining a post-feature point set of a visible light compression atlas corresponding to the post-frame; performing feature point matching on the current feature point set and the post-feature point set to obtain an initial feature point pair group; performing error matching screening on the initial feature point pair group to obtain a feature point pair group, and performing the following operations on each feature point pair in the feature point pair group: calculating the coordinate difference of the feature point pair, obtaining the time interval between the current frame and the post-frame, and obtaining a velocity vector based on the coordinate difference and the time interval; summarizing the velocity vectors to obtain a velocity vector set corresponding to the feature point pair group, and summarizing the velocity vector set corresponding to the visible light compression atlas to obtain multiple velocity vector sets; calculating the average value of the multiple velocity vector sets to obtain a droplet velocity vector.

[0052] It should be explained that the current frame refers to the frame of visible light compression image currently being processed, in which each frame of visible light compression image is extracted from the visible light compression atlas in sequence when processing the visible light compression atlas. The current feature point refers to the set of feature points extracted from the current frame image. For example, feature points are: corner points, edge points, etc. The post-frame refers to the visible light compression image of the next frame adjacent to the current frame in the process of sequentially processing the visible light compression atlas. The post-feature point set refers to the set of feature points extracted from the visible light compression image corresponding to the post-feature frame. The feature point matching of the current feature point set with the post-feature point set to obtain the initial feature point pair group refers to using the feature point matching algorithm to find the feature point in the post-feature point set that is most similar to each feature point in the current feature point set, thereby forming a pair of feature point combinations. For example, the feature point matching algorithm is brute force matching.

[0053] Importantly, the initial feature point pair group refers to a set of all feature point pairs obtained by feature point matching. The coordinate difference of the feature point pair is calculated by using the coordinates of the feature point pair to subtract the horizontal coordinate from the horizontal coordinate and the vertical coordinate from the vertical coordinate to obtain the coordinate difference. The coordinate difference reflects the position change of the feature point between the two frames of images. For example, the coordinate difference of the known feature point pair is And the time interval between the current frame and the subsequent frame , the velocity vector Quantity , Quantity , so the velocity vector is expressed as .

[0054] It should be explained that the velocity vector set refers to a set consisting of all velocity vectors, and the droplet velocity vector refers to a vector obtained by calculating the average value of all velocity vectors in multiple velocity vector sets.

[0055] In detail, the initial feature point pair group is subjected to error matching screening to obtain the feature point pair group, including: obtaining the relative displacement of each pair of initial feature point pairs in the initial feature point pair group to obtain a relative displacement set, and obtaining the displacement standard deviation according to the relative displacement set; judging whether there is a relative displacement greater than the displacement standard deviation in the relative displacement set; if it is confirmed that there is a relative displacement greater than the displacement standard deviation in the relative displacement set, the initial feature point pair corresponding to the relative displacement is eliminated to obtain an updated feature point pair group; extracting an updated feature point pair from the updated feature point pair group in turn, and performing the following operations on the extracted updated feature point pairs: obtaining the first local area of ​​the updated feature point pair using a preset local area window; area and a second local area; obtaining a first local texture feature and a second local texture feature according to the first local area and the second local area; comparing the first local texture feature with the second local texture feature; if the first local texture feature is the same as the second local texture feature, obtaining a feature point pair; if the first local texture feature is different from the second local texture feature, eliminating the updated feature point pair, and summarizing the retained updated feature point pairs to obtain an optimized feature point pair group, taking the optimized feature point pair group as the updated feature point pair group, returning to the step of extracting an updated feature point pair from the updated feature point pair group in sequence; summarizing the feature point pairs to obtain a feature point pair group corresponding to the updated feature point pair group.

[0056] It should be explained that the method of obtaining the relative displacement of each pair of initial feature point pairs in the initial feature point pair group is the same as the method of calculating the coordinate difference of the feature point pair, which will not be repeated here. The step of obtaining the displacement standard deviation based on the relative displacement set is a prior art and will not be repeated here. The updated feature point pair group refers to the set of initial feature point pairs remaining after the relative displacement greater than the displacement standard deviation is eliminated in the initial feature point pair group. The method of obtaining the first local area and the second local area of ​​the updated feature point pair using a preset local area window refers to extracting two updated feature points from the updated feature point pair, taking the updated feature point as the center position of the local area window, and using the local area window to obtain the local areas corresponding to the two updated feature points to obtain the first local area and the second local area.

[0057] Importantly, the obtaining of the first local texture feature and the second local texture feature based on the first local area and the second local area refers to extracting texture features from the first local area and the second local area using a texture extraction feature method to obtain the first local texture feature and the second local texture feature. For example, the texture extraction feature method is a grayscale co-occurrence matrix, a local binary pattern, etc. A feature point pair is a feature point pair that has consistency in the local texture and meets the characteristics of a correct match. An optimized feature point pair group refers to a set of feature point pairs obtained by eliminating updated feature point pairs with different local texture features when comparing texture features of updated feature point pairs. A feature point pair group is a set consisting of feature point pairs.

[0058] S7. Summarize the temperature field, droplet velocity vector and binary mask respectively to obtain the temperature field set, droplet velocity vector set and binary mask set corresponding to the three-stage smelting parameters.

[0059] It should be explained that the temperature field is an image that reflects the temperature distribution of the molten pool surface corresponding to the near-infrared normalized atlas within the entire image range. The temperature field can intuitively observe the temperature distribution characteristics of the molten pool surface.

[0060] S8. Obtaining three-stage molten pool liquid levels based on the three-stage smelting parameters and the initial molten pool liquid level.

[0061] In detail, the method of obtaining the three-stage molten pool level based on the three-stage smelting parameters and the initial molten pool level includes: calculating the three-stage molten pool level according to the three-stage smelting parameters and the initial molten pool level, wherein the calculation formula of the three-stage molten pool level is as follows: , , ,in, The molten pool level represents the first stage smelting parameters, represents the initial molten pool level, Indicates the preset current, Indicates the preset gas flow rate, Indicates the preset impurity content, Indicates the preset melting temperature, The molten pool level represents the second stage smelting parameters, The molten pool level represents the third stage smelting parameters. They represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level in the first stage of smelting parameters. , , , They represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level in the second stage smelting parameters. , , , They respectively represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level within the third stage smelting parameters. Represents the natural logarithm.

[0062] It should be explained that the gas flow rate refers to the flow rate of gas introduced into the furnace during the smelting process. The impurity content refers to the pre-set impurity content in the molten pool. The molten pool level refers to the height of the liquid in the molten pool. The influence coefficients of the current, gas flow rate, impurity content and smelting temperature on the molten pool level respectively refer to the higher the current, gas flow rate, impurity content and smelting temperature, the higher the molten pool level. The presence of impurity content may change the physical and chemical properties of the molten pool, such as density, viscosity, etc., thereby affecting the molten pool level.

[0063] S9. Complete titanium alloy smelting based on smelting image analysis based on the three-stage molten pool liquid level, temperature field set, droplet velocity vector set and binary mask set.

[0064] It should be explained that the temperature field set refers to the set consisting of all temperature fields, the droplet velocity vector set refers to the set consisting of all droplet velocity vectors, and the binary mask set refers to the set consisting of all binary masks.

[0065] The present invention solves the problems described in the background technology. The present invention builds a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system. The present invention can obtain information about the molten pool from multiple dimensions through imaging systems of different spectra. The near-infrared pyrometer array can measure the temperature distribution on the surface of the molten pool. The visible light high dynamic range imaging system can clearly capture the appearance of the molten pool, liquid level fluctuations, etc. The X-ray real-time imaging system can penetrate the molten pool to detect internal structures and defects. This multi-spectral real-time imaging system can provide more comprehensive and detailed molten pool information, which is helpful for accurately analyzing the smelting process and obtaining titanium alloys, vacuum arc furnaces and smelting parameters. The electric arc furnace comprises: a smelting observation window and a molten pool. The present invention can reduce the reaction between the titanium alloy and impurities in the air during the smelting process through the use of a vacuum electric arc furnace, thereby improving the smelting quality. Three-stage smelting parameters are obtained according to the smelting parameters, wherein the three-stage smelting parameters include: first-stage smelting parameters, second-stage smelting parameters and third-stage smelting parameters, wherein the smelting parameters include: an electrical range, a gas range and a temperature range. The present invention divides the entire smelting process into three sections, and sets different smelting parameters respectively. The smelting process can be controlled more accurately according to the characteristics and requirements of the titanium alloy at different smelting stages, which is helpful to improve the purity, uniformity and performance stability of the titanium alloy. The first-stage smelting parameters are extracted from the three-stage smelting parameters in turn, and the extracted first-stage smelting parameters are processed. The following operations are performed: a first smelting operation is performed on a titanium alloy using a vacuum arc furnace and the first stage smelting parameters, and the initial molten pool liquid level is recorded in real time. The present invention provides a benchmark for subsequent monitoring of changes in the molten pool liquid level by recording the initial molten pool liquid level in real time, thereby ensuring that the liquid in the molten pool does not overflow; the molten pool is photographed using the multi-spectral real-time imaging system and the smelting observation window to obtain a near-infrared image set, a visible light image set and an X-ray image set. The present invention photographs the molten pool image using the multi-spectral real-time imaging system, and can obtain various information about the molten pool at this smelting stage, so that operators can observe the smelting process in real time and discover potential problems in a timely manner; a smelting image analysis instruction is received, and the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set are detected according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set are detected. The temperature field, droplet velocity vector and binary mask are obtained from the line image set. The present invention can help understand the temperature distribution inside the molten pool through the temperature field, and provide a basis for adjusting the temperature range. The droplet velocity vector can reflect the flow of materials in the molten pool, which helps to optimize the convection and mixing during the smelting process. The binary mask can be used to identify different materials or defects in the molten pool, which is convenient for quality control. The temperature field, droplet velocity vector and binary mask are summarized respectively to obtain the temperature field set, droplet velocity vector set and binary mask set corresponding to the three-stage smelting parameters. The present invention summarizes the temperature field, droplet velocity vector and binary mask obtained in each smelting process to form a complete data set, so that the entire smelting process can be comprehensively analyzed and evaluated.Understand the changes and mutual relationships in different stages, obtain the three-stage molten pool level based on the three-stage smelting parameters and the initial molten pool level. The present invention can reflect the material balance and reaction conditions in the smelting process through the changes in the molten pool level, which is helpful to judge whether the smelting process is stable. The smelting of titanium alloy based on smelting image analysis is completed based on the three-stage molten pool level, temperature field set, droplet velocity vector set and binary mask set. The present invention can improve the purity, uniformity and performance stability of titanium alloys through precise control based on image analysis, reduce the generation of defects and impurities, and thus improve the quality and reliability of titanium alloy products. Therefore, the present invention can realize rapid and accurate image analysis. ,

[0066] like Figure 2 , which is a functional module diagram of a titanium alloy smelting system based on smelting image analysis provided by one embodiment of the present invention.

[0067] The titanium alloy smelting system 100 based on smelting image analysis of the present invention can be installed in an electronic device. According to the functions implemented, the titanium alloy smelting system 100 based on smelting image analysis can include a smelting parameter preparation module 101, a smelting image acquisition module 102, a smelting image analysis module 103 and a smelting completion module 104. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and is stored in the memory of the electronic device; the smelting parameter preparation module 101 is used to build a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system, to obtain titanium alloy, vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace includes: a smelting observation window and a molten pool, and three-stage smelting parameters are obtained according to the smelting parameters, wherein the three-stage smelting parameters include: a first-stage smelting parameter, a second-stage smelting parameter and a third-stage smelting parameter, wherein the smelting parameters include: an electrical range, a gas range and a temperature range; the smelting image acquisition module 102 is used to extract the first-stage smelting parameters from the three-stage smelting parameters in sequence, and perform the following on the extracted first-stage smelting parameters Operation: Perform the first smelting operation on the titanium alloy using the vacuum arc furnace and the first stage smelting parameters, and record the initial molten pool level in real time, and use the multi-spectral real-time imaging system and the smelting observation window to photograph the molten pool to obtain a near-infrared image set, a visible light image set and an X-ray image set; the smelting image analysis module 103 is used to receive a smelting image analysis instruction, obtain the temperature field, the droplet velocity vector and the binary mask according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set, respectively summarize the temperature field, the droplet velocity vector and the binary mask to obtain the temperature field set, the droplet velocity vector set and the binary mask set corresponding to the three-stage smelting parameters; the smelting completion module 104 is used to obtain the three-stage molten pool level based on the three-stage smelting parameters and the initial molten pool level, and complete the titanium alloy smelting based on the smelting image analysis based on the three-stage smelting pool level, the temperature field set, the droplet velocity vector set and the binary mask set.

[0068] In detail, each module in the titanium alloy smelting system 100 based on smelting image analysis in the embodiment of the present invention is used in the same manner as described above. Figure 1 The same technical means as the titanium alloy smelting method based on smelting image analysis described in the text and can produce the same technical effects will not be repeated here.

[0069] like Figure 3 , which is a schematic diagram of the structure of an electronic device for implementing a titanium alloy smelting method based on smelting image analysis provided by one embodiment of the present invention.

[0070] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a titanium alloy smelting method program based on smelting image analysis.

[0071] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the titanium alloy smelting method program based on smelting image analysis, but also can be used to temporarily store data that has been output or is to be output.

[0072] In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (such as a titanium alloy smelting method program based on smelting image analysis, etc.), and calls data stored in the memory 11, so as to execute various functions of the electronic device 1 and process data.

[0073] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0074] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0075] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0076] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0077] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0078] The titanium alloy smelting method program based on smelting image analysis stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve: building a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system; obtaining titanium alloy, vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace includes: a smelting observation window and a molten pool; obtaining three-stage smelting parameters according to the smelting parameters, wherein the three-stage smelting parameters include: a first-stage smelting parameter, a second-stage smelting parameter and a third-stage smelting parameter, wherein the smelting parameters include: an electrical range, a gas range and a temperature range; extracting the first-stage smelting parameters from the three-stage smelting parameters in turn, and performing the following operations on the extracted first-stage smelting parameters: A first smelting operation is performed on a titanium alloy using a vacuum arc furnace and first-stage smelting parameters, and the initial molten pool level is recorded in real time; the molten pool is photographed using the multi-spectral real-time imaging system and the smelting observation window to obtain a near-infrared image set, a visible light image set and an X-ray image set; a smelting image analysis instruction is received, and a temperature field, a droplet velocity vector and a binary mask are acquired according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set; the temperature field, the droplet velocity vector and the binary mask are respectively summarized to obtain a temperature field set, a droplet velocity vector set and a binary mask set corresponding to the three-stage smelting parameters; the three-stage molten pool levels are acquired based on the three-stage smelting parameters and the initial molten pool level; and the titanium alloy smelting based on the smelting image analysis is completed based on the three-stage molten pool levels, the temperature field set, the droplet velocity vector set and the binary mask set.

[0079] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0080] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0081] The present invention also provides a computer-readable storage medium, wherein the computer program is stored in the computer program. When the computer program is executed by the processor of the electronic device, the computer program can achieve: building a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system; obtaining titanium alloy, vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace includes: a smelting observation window and a molten pool; obtaining three-stage smelting parameters according to the smelting parameters, wherein the three-stage smelting parameters include: a first-stage smelting parameter, a second-stage smelting parameter and a third-stage smelting parameter, wherein the smelting parameters include: an electrical range, a gas range and a temperature range; extracting the first-stage smelting parameters from the three-stage smelting parameters in turn, and performing the following operations on the extracted first-stage smelting parameters: using A first smelting operation is performed on the titanium alloy using a vacuum arc furnace and the first stage smelting parameters, and the initial molten pool level is recorded in real time; the molten pool is photographed using the multi-spectral real-time imaging system and the smelting observation window to obtain a near-infrared image set, a visible light image set and an X-ray image set; a smelting image analysis instruction is received, and a temperature field, a droplet velocity vector and a binary mask are acquired according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set; the temperature field, the droplet velocity vector and the binary mask are respectively summarized to obtain a temperature field set, a droplet velocity vector set and a binary mask set corresponding to the three-stage smelting parameters; the three-stage molten pool levels are acquired based on the three-stage smelting parameters and the initial molten pool level; and the titanium alloy smelting based on the smelting image analysis is completed based on the three-stage molten pool levels, the temperature field set, the droplet velocity vector set and the binary mask set.

[0082] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0083] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0085] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A titanium alloy smelting method based on smelting image analysis, characterized in that: The method comprises: building a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system comprises: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system; obtaining titanium alloy, vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace comprises: a smelting observation window and a molten pool; obtaining three-stage smelting parameters according to the smelting parameters, wherein the three-stage smelting parameters comprise: a first-stage smelting parameter, a second-stage smelting parameter and a third-stage smelting parameter, wherein the smelting parameters comprise: an electrical range, a gas range and a temperature range; extracting the first-stage smelting parameters from the three-stage smelting parameters in sequence, and performing the following operations on the extracted first-stage smelting parameters: performing a first smelting operation on the titanium alloy using the vacuum arc furnace and the first-stage smelting parameters, and record the initial molten pool level in real time; photograph the molten pool using the multi-spectral real-time imaging system and the smelting observation window to obtain a near-infrared image set, a visible light image set and an X-ray image set; receive a smelting image analysis instruction, and obtain a temperature field, a droplet velocity vector and a binary mask according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set; respectively summarize the temperature field, the droplet velocity vector and the binary mask to obtain a temperature field set, a droplet velocity vector set and a binary mask set corresponding to the three-stage smelting parameters; obtain the three-stage molten pool level based on the three-stage smelting parameters and the initial molten pool level; and complete the titanium alloy smelting based on the smelting image analysis based on the three-stage molten pool level, the temperature field set, the droplet velocity vector set and the binary mask set.

2. The titanium alloy smelting method based on smelting image analysis according to claim 1, characterized in that: The method of obtaining three-stage smelting parameters according to the smelting parameters includes: obtaining the upper limit of the first section interval and the upper limit of the second section interval; using the upper limit of the first section interval and the upper limit of the second section interval to perform three-stage operation on the electrical interval, gas interval and temperature interval in the smelting parameters to obtain an electrical segmented interval group, a gas segmented interval group and a temperature segmented interval group, wherein the electrical segmented interval group includes: a first electrical interval, a second electrical interval and a third electrical interval, the gas segmented interval group includes: a first gas interval, a second gas interval and a third gas interval, and the temperature segmented interval group includes: a first temperature interval, a second temperature interval and a third temperature interval; using the first electrical interval, the first gas interval and the first temperature interval as the first-stage smelting parameters, using the second electrical interval, the second gas interval and the second temperature interval as the second-stage smelting parameters, and using the third electrical interval, the third gas interval and the third temperature interval as the third-stage smelting parameters; confirming the first-stage smelting parameters, the second-stage smelting parameters and the third-stage smelting parameters as the three-stage smelting parameters.

3. The titanium alloy smelting method based on smelting image analysis according to claim 2, characterized in that: The obtaining of the upper limit of the first interval and the upper limit of the second interval includes: using a pre-constructed interval segmentation formula to calculate the upper limit of the first interval and the upper limit of the second interval, wherein the interval segmentation formula is as follows: , ,in, Indicates the upper limit of the first interval, Indicates the upper limit of the second interval, Indicates the minimum value of the preset interval. Indicates the maximum value of the preset interval. Indicates the preset time adjustment parameters. Indicates the time value in the preset time interval. Indicates the minimum value of the preset time interval. Indicates the maximum value of the preset time interval.

4. The titanium alloy smelting method based on smelting image analysis according to claim 3, characterized in that: The method of acquiring the temperature field, the droplet velocity vector and the binary mask according to the melting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set includes: using the melting image analysis instruction to perform a normalization operation on the near-infrared images in the near-infrared image set to obtain a near-infrared normalized atlas; performing the following operations on each visible light image in the visible light image set: acquiring a pixel value set of the visible light image, dynamically compressing each pixel value in the pixel value set using a pre-built logarithmic compression method to obtain a logarithmic compressed pixel value set, wherein the logarithmic compressed pixel value set includes multiple logarithmic compressed pixel values; extracting the maximum pixel value and the minimum pixel value from the logarithmic compressed pixel value set, and using the maximum pixel value and the minimum pixel value to obtain a logarithmic compressed pixel value set; A linear mapping operation is performed on each logarithmically compressed pixel value in the logarithmically compressed pixel value set according to a preset pixel value interval and a pre-built linear stretching method to obtain a mapped pixel value set; a data type conversion operation is performed on each mapped pixel value in the mapped pixel value set to obtain a visible light compression map; the visible light compression maps are summarized to obtain a visible light compression atlas corresponding to the visible light image set, wherein the visible light compression map is an image corresponding to a frame in the visible light image set; a correction and suppression operation is performed on all X-ray images in the X-ray image set to obtain an X-ray correction atlas; the temperature field is obtained based on the near-infrared normalized atlas, the droplet velocity vector is obtained based on the visible light compression atlas, and the binary mask is obtained based on the X-ray correction atlas.

5. The titanium alloy smelting method based on smelting image analysis according to claim 4, characterized in that: The method of acquiring the temperature field based on the near-infrared normalized atlas includes: extracting near-infrared normalized images from the near-infrared normalized atlas in sequence, and performing the following operations on the extracted near-infrared normalized images: gridding the near-infrared normalized images using a preset pixel specification to obtain a grid area set, obtaining a molten pool surface fluctuation model, calculating the molten pool surface state corresponding to each grid area in the grid area set using the molten pool surface fluctuation model, and obtaining a molten pool surface state set; adjusting the pre-constructed emissivity value of each grid area in the grid area set using the molten pool surface state set, obtaining an optimized grid area set, and calculating the optimized grid area set. The temperature values ​​of each optimized grid area in the grid area set are collected to obtain an accurate temperature value set; the number of grids in the optimized grid area set and the number of temperature values ​​in the accurate temperature value set are obtained to determine whether the number of grids is greater than the number of temperature values; if the number of grids is greater than the number of temperature values, the optimized grid area is located to obtain an abnormal grid area, and spatial interpolation is performed on the abnormal grid area to obtain an interpolated temperature value, wherein the abnormal grid area includes: one abnormal grid area or multiple abnormal grid areas; the accurate temperature value set and the interpolated temperature value are respectively summarized to obtain the temperature field corresponding to the near-infrared normalized atlas.

6. The titanium alloy smelting method based on smelting image analysis according to claim 5, characterized in that: The method of performing spatial interpolation on the abnormal grid area to obtain an interpolated temperature value includes: obtaining a window grid set according to a preset grid window and the abnormal grid area, calculating the distance between the abnormal grid area and each window grid in the window grid set, and obtaining a grid distance set; using a pre-constructed radial basis function to calculate the weight of each grid distance in the grid distance set, and obtaining a distance weight set; and calculating an average temperature value according to the distance weight set and the precise temperature value set corresponding to the window grid set, wherein the calculation formula of the average temperature value is as follows: ,in, represents the average temperature value, Indicates the number of window grid sets, represents the index of the window grid set, Indicates the preset The measurement error correction factor of the window grid is, Indicates The exact temperature value of the window grid, Indicates The distance weight of the window grid, Indicates the preset The uncertainty weight factor of the window grid; the average temperature value is used as the interpolated temperature value.

7. The titanium alloy smelting method based on smelting image analysis according to claim 6, characterized in that: The method of obtaining a droplet velocity vector based on a visible light compression atlas comprises: extracting a current frame from the visible light compression atlas in sequence, and performing the following operations on each of the extracted current frames: obtaining a current feature point set based on the extracted current frame, extracting a next current frame adjacent to the extracted current frame to obtain a post-frame, and obtaining a post-feature point set of a visible light compression atlas corresponding to the post-frame; performing feature point matching on the current feature point set and the post-feature point set to obtain an initial feature point pair group; performing error matching screening on the initial feature point pair group to obtain a feature point pair group, and performing the following operations on each feature point pair in the feature point pair group: calculating a coordinate difference of the feature point pair, obtaining a time interval between the current frame and the post-frame, and obtaining a velocity vector based on the coordinate difference and the time interval; summarizing the velocity vectors to obtain a velocity vector set corresponding to the feature point pair group, and summarizing the velocity vector set corresponding to the visible light compression atlas to obtain multiple velocity vector sets; and calculating an average value of the multiple velocity vector sets to obtain a droplet velocity vector.

8. The titanium alloy smelting method based on smelting image analysis according to claim 7, characterized in that: The method performs error matching screening on the initial feature point pair group to obtain the feature point pair group, including: obtaining the relative displacement of each pair of initial feature point pairs in the initial feature point pair group to obtain a relative displacement set, and obtaining the displacement standard deviation according to the relative displacement set; judging whether there is a relative displacement greater than the displacement standard deviation in the relative displacement set; if it is confirmed that there is a relative displacement greater than the displacement standard deviation in the relative displacement set, the initial feature point pair corresponding to the relative displacement is eliminated to obtain an updated feature point pair group; extracting an updated feature point pair from the updated feature point pair group in turn, and performing the following operations on the extracted updated feature point pairs: obtaining a first local area of ​​the updated feature point pair using a preset local area window and a second local area; obtaining a first local texture feature and a second local texture feature according to the first local area and the second local area; comparing the first local texture feature with the second local texture feature; if the first local texture feature is the same as the second local texture feature, obtaining a feature point pair; if the first local texture feature is different from the second local texture feature, eliminating the updated feature point pairs, and summarizing the retained updated feature point pairs to obtain an optimized feature point pair group, taking the optimized feature point pair group as the updated feature point pair group, returning to the step of extracting an updated feature point pair from the updated feature point pair group in turn; summarizing the feature point pairs to obtain a feature point pair group corresponding to the updated feature point pair group.

9. The titanium alloy smelting method based on smelting image analysis according to claim 8, characterized in that: The method of obtaining the three-stage molten pool liquid level based on the three-stage smelting parameters and the initial molten pool liquid level includes: calculating the three-stage molten pool liquid level according to the three-stage smelting parameters and the initial molten pool liquid level, wherein the calculation formula of the three-stage molten pool liquid level is as follows: , , ,in, The molten pool level represents the first stage smelting parameters, represents the initial molten pool level, Indicates the preset current, Indicates the preset gas flow rate, Indicates the preset impurity content, Indicates the preset melting temperature, The molten pool level represents the second stage smelting parameters, The molten pool level represents the third stage smelting parameters. They represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level in the first stage of smelting parameters. , , , They represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level in the second stage smelting parameters. , , , They respectively represent the influence coefficients of current, gas flow, impurity content and smelting temperature on the molten pool level within the third stage smelting parameters. Represents the natural logarithm.

10. A titanium alloy smelting system based on smelting image analysis, characterized in that: The system includes: a smelting parameter preparation module, which is used to build a multi-spectral real-time imaging system, wherein the multi-spectral real-time imaging system includes: a near-infrared pyrometer array, a visible light high dynamic range and an X-ray real-time imaging system, to obtain titanium alloy, vacuum arc furnace and smelting parameters, wherein the vacuum arc furnace includes: a smelting observation window and a molten pool, and three-stage smelting parameters are obtained according to the smelting parameters, wherein the three-stage smelting parameters include: first-stage smelting parameters, second-stage smelting parameters and third-stage smelting parameters, wherein the smelting parameters include: an electrical range, a gas range and a temperature range; a smelting image acquisition module, which is used to extract the first-stage smelting parameters from the three-stage smelting parameters in turn, and perform the following operations on the extracted first-stage smelting parameters: perform a first smelting operation on the titanium alloy using the vacuum arc furnace and the first-stage smelting parameters , and record the initial molten pool level in real time, use the multi-spectral real-time imaging system and the smelting observation window to shoot the molten pool, and obtain a near-infrared image set, a visible light image set and an X-ray image set; a smelting image analysis module, which is used to receive a smelting image analysis instruction, obtain a temperature field, a droplet velocity vector and a binary mask according to the smelting image analysis instruction, the near-infrared image set, the visible light image set and the X-ray image set, respectively summarize the temperature field, the droplet velocity vector and the binary mask, and obtain the temperature field set, the droplet velocity vector set and the binary mask set corresponding to the three-stage smelting parameters; a smelting completion module, which is used to obtain the three-stage molten pool level based on the three-stage smelting parameters and the initial molten pool level, and complete the titanium alloy smelting based on the smelting image analysis based on the three-stage smelting pool level, the temperature field set, the droplet velocity vector set and the binary mask set.

Images