System for performance regulation of silver-containing martensitic antibacterial stainless steel for kitchen knives

Abstract

This invention relates to the field of smelting and control technology for martensitic antibacterial steel, specifically a performance control system for silver-containing martensitic antibacterial stainless steel for kitchen knives. The system includes modules for molten pool monitoring, thermodynamic analysis, flow analysis, distribution prediction, and process optimization. The molten pool monitoring module collects instantaneous temperature values, melt surface tension distribution values, and silver concentration data for each region of the molten pool. The thermodynamic analysis module generates a thermodynamic state cloud map. The flow analysis module integrates surface tension and flow resistance models to generate a viscous convection resistance surface, quantifying the mass transfer resistance of silver. The distribution prediction module couples the thermodynamic cloud map and the resistance surface, predicting the initial distribution morphology of silver through solidification kinetics simulation. The process optimization module identifies compositional fluctuation patterns, predicts the location of silver phase aggregation and segregation nuclei, maps distribution control targets, and generates smelting parameter adjustment schemes with risk labels. This system can achieve precise control of silver distribution during the smelting process of silver-containing martensitic antibacterial stainless steel.

Description

Technical Field

[0001] This invention relates to the field of martensitic antibacterial steel smelting and control technology, and in particular to a performance control system for silver-containing martensitic antibacterial stainless steel for kitchen knives. Background Technology

[0002] Current methods for smelting and preparing silver-containing martensitic antibacterial stainless steel primarily rely on conventional temperature monitoring, point-to-point composition detection, and simple melt state observation for process control. These methods depend on recording basic thermodynamic parameters and adjusting for compositional fluctuations to achieve the melting and casting of antibacterial stainless steel for kitchen knives. However, these technologies only discretely collect data on molten pool temperature and silver concentration, and rely on a single surface tension value to determine the melt state. They lack a targeted mechanism for quantifying and simulating mass transfer resistance. Existing technologies cannot spatially quantify the degree of convective mass transfer resistance of silver within the molten pool, making it difficult to accurately identify regions where silver mass transfer is restricted. During the ingot solidification stage, they rely solely on empirically set compositional uniformity targets, failing to combine melt flow resistance and thermodynamic state to dynamically simulate silver phase precipitation, and thus cannot predict the silver distribution morphology in advance.

[0003] In the smelting process of silver-containing martensitic antibacterial stainless steel, it is impossible to generate a viscous convection resistance surface that quantifies the mass transfer obstacle of silver by integrating the melt surface tension and flow resistance model. It is also difficult to carry out simulation of the silver phase precipitation process driven by solidification kinetics based on the coupling relationship between the thermodynamic state cloud map of the molten pool and the viscous convection resistance surface. It is impossible to accurately predict the initial distribution morphology of silver in the matrix, and it is also impossible to identify the silver phase aggregation probability and segregation nucleus position based on composition fluctuations and map them to the distribution control target. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and propose a performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives.

[0005] To achieve the above objectives, the present invention employs the following technical solution: a performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives, comprising: The molten pool monitoring module continuously records the instantaneous temperature, surface tension distribution, and silver concentration of each region of the molten pool during the smelting process of silver-containing martensitic antibacterial stainless steel. The thermal analysis module performs spatial and temporal correlation analysis on the instantaneous temperature values ​​of each region of the molten pool to generate a thermodynamic state cloud map of the molten pool. The flow analysis module calculates the viscous convection resistance surface of the molten pool by fusing the surface tension distribution value of the melt with a preset melt flow resistance model. The viscous convection resistance surface is used to quantify the obstruction effect of different molten pool regions on the convective mass transfer of silver elements. The distribution prediction module determines the uniform distribution control target of silver elements during the solidification process of the ingot based on the coupling relationship between the thermodynamic state cloud map and the viscous convection resistance surface, and performs a simulation of the silver phase precipitation process based on solidification kinetics on the uniform distribution control target to predict the initial distribution morphology of silver elements in the matrix. The process optimization module identifies the fluctuation pattern of the molten pool composition based on the continuous monitoring records of the silver concentration, predicts the probability of silver phase aggregation and the location of segregation nuclei in different parts of the ingot, maps the probability of silver phase aggregation and the location of segregation nuclei to the uniform distribution control target, and generates an alloy smelting process parameter adjustment scheme with process risk labeling.

[0006] As a further aspect of the present invention, the step of performing spatial and temporal correlation analysis on the instantaneous temperature values ​​of each region of the molten pool to generate a thermodynamic state cloud map of the molten pool includes: The thermodynamic state cloud map includes the core location of the high-energy hot zone, the location of the steep change line of the temperature gradient, and the expected segregation path of silver in the melt. The instantaneous temperature values ​​from all synchronous molten pool monitoring devices are time-synchronized to form a snapshot of the spatial temperature distribution at the same moment; Heat transfer path tracing is performed on a series of consecutive temperature spatial distribution snapshots to extract high-temperature regions that are stable in the time dimension and identify the core location of the high-energy thermal zone. Around the core of the high-energy thermal zone, dense bands of temperature isopleths and split points of temperature diffusion direction are detected, and the spatial coordinates of the dense bands and split points are recorded to form the position of the steep change line of the temperature gradient. Along the main heat output path of the high-energy hot zone, the average temperature decay rate is calculated at fixed step intervals, and the calculated average temperature decay rates are connected in sequence along the path to form the expected segregation path of the silver element in the melt. The core location of the high-energy hot zone, the location of the steep change line of the temperature gradient, and the expected segregation path of the silver element in the melt are integrated and encoded into the thermodynamic state cloud map.

[0007] As a further aspect of the present invention, the viscous convection resistance surface of the molten pool is generated by fusing the melt surface tension distribution value with a preset melt flow resistance model, including: Gradient calculations are performed on the surface tension distribution values ​​of the melt to obtain the force direction and force field strength values ​​of each micro-element region of the molten pool. By combining a pre-set database of alloying elements and a database of physical properties of furnace lining, corresponding rheological parameters are matched for each micro-element region of the molten pool. Based on the force direction, force field strength value, and rheological property parameters, the effective viscosity coefficient and convection inhibition factor of the melt corresponding to each micro-element region of the molten pool are obtained by looking up the table. The effective viscosity coefficient of the melt and the convection inhibition factor are used to parameterize the reference fluid dynamics model, and the theoretical convection mass resistance value of each micro-element region of the molten pool is calculated. By fitting the theoretical convection mass resistance values ​​of all micro-regions within the entire molten pool to a spatial surface, a viscous convection resistance surface characterizing the degree to which the melt flow hinders the diffusion of silver is constructed.

[0008] As a further aspect of the present invention, based on the continuous monitoring records of the silver concentration, the compositional fluctuation pattern of the molten pool is identified to predict the probability of silver phase aggregation and the location of segregation nuclei in different parts of the ingot, including: The continuous monitoring records of silver concentration are divided into segments according to the preset pouring stages, and the variance of silver concentration fluctuation within each pouring stage is calculated and the correlation coefficient between the two samples is calculated. The molten pool regions where the variance and correlation coefficient exceed the process stability threshold are marked as silver concentration abnormal fluctuation regions; By analyzing the spatial evolution morphology and temporal variation sequence of the abnormal silver concentration fluctuation zone, abrupt changes in the silver concentration gradient that conform to the microsegregation characteristics in alloy solidification theory are identified, and the abrupt changes in the silver concentration gradient are determined to be potential high-probability areas for silver phase aggregation. Within the abnormal fluctuation region of silver concentration, the high-frequency oscillation signal of silver concentration is further extracted, and the location of the segregation nucleus generation is located using signal power spectral density identification technology. The location of the segregation nucleus generation corresponds to the spatial core point where the high-frequency oscillation power of silver concentration is strongest. Each identified high-probability silver phase aggregation region and the location of segregation nucleus generation is assigned an aggregation intensity level label.

[0009] As a further aspect of the present invention, the silver phase aggregation probability and the segregation nucleus generation location are mapped to the uniform distribution control target to generate an alloy smelting process parameter adjustment scheme with process risk labeling, including: A spatial framework for the uniformly distributed control target is established, and the high-probability silver phase aggregation area labeled with aggregation intensity level is spatially superimposed with the segregation nucleus generation location for analysis. For uniformly distributed control target units that fall into any of the high probability zones of silver phase aggregation or whose distance from any of the segregation nuclei generation locations is less than the size of the critical solidification unit, the corresponding segregation risk type and risk level are recorded in their attributes. Based on the risk suppression strategy, one or more process parameter adjustment suggestions are generated for each uniformly distributed control target unit with the aforementioned segregation risk type and risk level. The process parameter adjustment suggestions involve the local electromagnetic stirring intensity, cooling rate, or holding time of the target unit. The uniformly distributed control target units that do not fall into the risk zone, along with all the generated process parameter adjustment suggestions, are integrated into the alloy melting process parameter adjustment scheme with process risk labels. Each control unit in the scheme includes position coordinates, segregation risk labels, and corresponding process parameter suggestions.

[0010] As a further aspect of the present invention, it also includes: The iterative optimization module is used to simulate and iteratively optimize the solidification process of the alloy melting process parameter adjustment scheme with process risk labeling. From the alloy melting process parameter adjustment schemes marked with process risks, the control unit with the highest segregation risk level is selected as the first batch of process intervention units. Based on the solidification sequence in the thermodynamic state cloud diagram, the main solidification conditions for tissue evolution are set; A dendrite growth simulator based on a phase field model was used to calculate the solute redistribution and silver phase precipitation images generated when the first batch of process intervention units solidified under the main solidification conditions. The solute redistribution and silver precipitation images are superimposed on the uniform distribution control target to evaluate the degree of influence of microsegregation on the uniformity of silver element distribution in the final tool matrix. Based on the assessment results, the expected silver distribution uniformity of the significantly affected downstream control units was reduced, and all control units to be intervened were reordered based on the reduced uniformity index and segregation risk level. From the reordered control units to be intervened, the optimal intervention unit is selected and added to the execution sequence. The steps from dendrite growth simulation to unit rearrangement are repeated until all control units to be intervened complete the process decision or reach the preset process control limit, and finally the optimized performance control process scheme is generated.

[0011] As a further aspect of the present invention, the dendrite growth simulator based on a phase-field model is used to calculate the solute redistribution and silver phase precipitation images generated when the first batch of process intervention units solidifies under the main solidification conditions, including: For each virtual material element of the first batch of process intervention units, the phase field variables and solute concentration field are initialized under the main solidification conditions; Based on the main solidification conditions, the thermodynamic parameters of the current material volume element, and the diffusion coefficient of silver in the iron matrix, the phase field variables and solute concentration field are driven to evolve and be calculated in the spatiotemporal domain. The solid-liquid interface propagation and solute repulsion processes of each material volume element during phase field evolution were tracked, and the concentration distribution of silver in the solid and liquid phases was recorded. The simulation region is discretized into a micro-grid, and the final concentration of silver in each micro-grid unit at the moment of solidification is calculated. The concentration deviation value is obtained by comparing it with the initial average concentration. The concentration deviation values ​​of all micro-grid units are integrated to generate a solute redistribution and silver phase precipitation image covering the entire simulation area, and the boundary of the severely segregated micro-segregation area is determined based on a preset concentration deviation threshold.

[0012] As a further aspect of the present invention, it also includes: The dynamic fine-tuning module is used for the dynamic fine-tuning steps of the process for small-scale sample test results. After small-scale stainless steel samples were prepared according to the preliminary process plan, the actual composition distribution data, microhardness data and silver phase morphology image data of the samples were collected. The actual component distribution data is compared with the predicted component distribution of the uniform distribution control target in the corresponding region to generate a local component distribution correction field. Analyze the silver phase morphology image data, statistically analyze the size distribution, shape factor and dispersion of the silver phase in the matrix, and identify the areas where the actual silver phase precipitation behavior deviates from the theoretical prediction; By combining the information of the local component distribution correction field and the actual silver precipitation behavior deviation region, the expected silver distribution of the remaining process control units that have not yet been executed or can be adjusted is re-evaluated; Based on the results of the reassessment, the process parameters or intervention priorities of the remaining process control units are adjusted to generate a dynamically updated performance regulation process scheme.

[0013] As a further aspect of the present invention, the actual component distribution data is compared with the predicted component distribution of the uniform distribution control target in the corresponding region to generate a local component distribution correction field, including: For each sample detection area that provides the actual component distribution data, the average concentration of silver element within a specific detection surface is extracted. From the corresponding region of the uniformly distributed control target, read the predicted value of the average silver concentration in the sample detection area within the same specific detection surface; The ratio of the measured average concentration of silver to the predicted average concentration of silver is calculated and used as the component prediction correction coefficient for the sample detection area. Using spatial interpolation technology, the component prediction correction coefficient of each sample detection area is extended to the simulated three-dimensional space range of the entire tool material, forming a continuous local component distribution correction field; The local component distribution correction field is multiplied point by point with the original uniform distribution control target to obtain the updated distribution map of the performance regulation target of silver-containing martensitic antibacterial stainless steel.

[0014] As a further aspect of the present invention, the steps for constructing the preset melt flow resistance model are as follows: The historical smelting process data of various silver-containing stainless steels with different compositions under preset smelting parameters were collected. The monitoring data included historical melt surface tension distribution, historical temperature field distribution, historical silver concentration field distribution, and historical ingot final silver element distribution evaluation. For each different composition, a correspondence between the alloy composition and its standard melt viscosity coefficient and surface tension temperature coefficient at multiple preset temperatures is established, forming an alloy composition-physical property parameter mapping table; For each historical smelting process, a dynamic spatiotemporal field of melt viscosity coefficient is generated based on the historical temperature field distribution and the alloy composition-physical property parameter mapping table. Based on the historical melt surface tension distribution and the historical silver concentration field distribution, the theoretical value of the convection velocity field driven by the surface tension gradient and concentration gradient inside the melt is calculated. The final silver element distribution evaluation of the historical ingots is quantified as an index of the uniformity of silver element distribution; Using the dynamic spatiotemporal field of the melt viscosity coefficient, the theoretical value of the convection velocity field, and the parameters in the melt flow resistance model as model inputs, and the uniformity index of silver element distribution as the optimization objective, the model parameters are fitted to obtain a preset melt flow resistance model that predicts the convection mass transfer resistance value.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: By fusing the surface tension distribution of the melt with a pre-defined melt flow resistance model, a viscous convection resistance surface of the molten pool is generated. This quantifies the obstruction effect of different regions of the molten pool on the convective mass transfer of silver, fully reflects the constraint state of each spatial location within the molten pool on the silver transport process, accurately presents the correspondence between melt flow characteristics and silver mass transfer behavior, refines the spatial distribution characteristics of mass transfer resistance within the molten pool, and realizes the numerical and spatial characterization of the restricted convective transport of silver in the melt. Simultaneously, a direct calculation correlation between surface tension parameters and mass transfer obstruction effects is established, forming a quantitative carrier that can directly reflect the distribution of silver mass transfer resistance, and refines the degree of influence of different regions within the molten pool on the silver migration process.

[0016] Based on the coupling relationship between the thermodynamic state cloud map of the molten pool and the viscous convection resistance surface, the control target for uniform silver distribution is determined. A simulation of the silver phase precipitation process based on solidification kinetics is then performed on this control target. This simulation outputs the initial distribution morphology of silver in the matrix, fully reproducing the dynamic changes in silver phase precipitation during the ingot solidification stage, and conforming to the physical evolution law of martensitic stainless steel matrix solidification. Combined with continuous monitoring records of silver concentration, the fluctuation pattern of molten pool composition is identified, determining the probability of silver phase aggregation and the location of segregation nuclei in different parts of the ingot. These parameters are mapped to the control target for uniform silver distribution, generating an alloy smelting process parameter adjustment scheme with process risk annotations. This achieves real-time matching between the silver distribution control target and the actual smelting composition fluctuation state, forming a parameter adjustment basis suitable for the entire ingot forming process. Attached Figure Description

[0017] Figure 1 This is a timing diagram of the performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives described in this invention. Figure 2 A flowchart for generating viscous convection resistance surfaces; Figure 3 A graph showing the characteristic analysis of the variance of silver concentration fluctuation and the time-series correlation coefficient at each pouring stage; Figure 4 A calibration diagram of the micro-segregation regions for silver phase precipitation and solute redistribution in phase-field simulation; Figure 5 A cloud map showing the predicted distribution of silver element concentration in the molten pool. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] See Figure 1 The overall implementation scheme of the silver-containing martensitic antibacterial stainless steel performance regulation system for kitchen knives provided by the present invention is as follows: During the smelting of silver-containing martensitic antibacterial stainless steel, the molten pool monitoring module is responsible for continuously monitoring and recording the instantaneous temperature values, melt surface tension distribution values, and silver concentration in various regions of the molten pool in real time. The thermodynamic analysis module receives the instantaneous temperature values ​​from the molten pool monitoring module and generates a thermodynamic state cloud map reflecting the overall thermodynamic state of the molten pool through spatial and temporal correlation analysis. The flow analysis module uses the acquired melt surface tension distribution values ​​and a preset melt flow resistance model to calculate and generate a viscous convection resistance surface that quantifies the convective mass transfer resistance of silver elements in different regions. The distribution prediction module couples the thermodynamic state cloud map with the viscous convection resistance surface to determine the target for uniform silver distribution during ingot solidification and simulates the silver phase precipitation process based on solidification kinetics to predict the initial distribution morphology of silver elements in the matrix. The process optimization module identifies the fluctuation pattern of the molten pool composition based on continuous monitoring records of silver concentration, predicts the probability of silver phase aggregation and the location of segregation nuclei in different parts of the ingot, and then maps these predicted risk locations to the previously determined uniform distribution control target, thereby generating an alloy smelting process parameter adjustment scheme with specific process risk labels.

[0021] In one embodiment of the invention, the instantaneous temperature values ​​of each region of the molten pool are subjected to spatial and temporal correlation analysis to generate a thermodynamic state cloud map. This cloud map includes the core location of the high-energy hot zone, the location of the steep temperature gradient line, and the expected segregation path of silver in the melt. The instantaneous temperature values ​​from all synchronous molten pool monitoring devices are first processed for temporal synchronization to form a snapshot of the spatial temperature distribution at the same moment. Heat transfer path tracking is performed on a series of consecutive temperature spatial distribution snapshots to extract high-temperature regions that are stable in the time dimension; these regions are identified as the core locations of the high-energy hot zone. Around the core location of the high-energy hot zone, dense bands of temperature isopleths and branching points in the temperature diffusion direction are detected, and the spatial coordinates of these dense bands and branching points are recorded; they together constitute the location of the steep temperature gradient line. Along the main heat output path of the high-energy hot zone, the average temperature decay rate is calculated at fixed step intervals, and the calculated average temperature decay rates are connected sequentially along the path to form the expected segregation path of silver in the melt. Ultimately, the core location of the high-energy hot zone, the location of the steep temperature gradient line, and the expected segregation path of silver in the melt were merged and encoded into a thermodynamic state cloud map.

[0022] In practical implementation, the instantaneous temperature values ​​output by the molten pool monitoring module come from sensors located at multiple physical locations. After entering the thermal analysis module, they first need to be synchronized in time. This synchronization is achieved by adding a high-precision unified time stamp to all sensor data streams. The synchronized data is integrated at each identical moment to form a temperature spatial distribution snapshot reflecting the three-dimensional temperature distribution of the molten pool at that moment. In some embodiments, the generation frequency of the temperature spatial distribution snapshot matches the step size of the smelting process control, for example, ten temperature spatial distribution snapshots are generated per second. A series of consecutive temperature spatial distribution snapshots constitute a temperature field evolution sequence. The heat transfer path tracking process is applied to this temperature field evolution sequence. The process compares the centroid movement trajectory and energy diffusion direction of the high-temperature region in the temperature spatial distribution snapshots at adjacent moments to extract high-temperature regions with stable spatial positions and temperature fluctuations below a set threshold over multiple consecutive moments. These extracted high-temperature regions are identified as the core locations of high-energy thermal zones. The core locations of high-energy thermal zones typically correspond to the Joule heat concentration area of ​​induction heating or the direct action area of ​​the electric arc within the molten pool.

[0023] In practical implementation, after identifying the core location of the high-energy thermal zone, the system performs neighborhood spatial analysis around the core location of each high-energy thermal zone. This neighborhood spatial analysis performs a radial scan outward from the core location of the high-energy thermal zone, calculating the temperature change rate in each direction. When the temperature change value per unit distance exceeds a preset gradient threshold in a certain direction, the spatial point in that direction is marked as a temperature gradient abrupt change point. Connecting adjacent temperature gradient abrupt change points forms the location of the temperature gradient abrupt change line. The location of the temperature gradient abrupt change line characterizes the boundary of abrupt change in heat transfer resistance within the molten pool. It can be understood that the branching points in the temperature diffusion direction are determined by analyzing the topological structure of the temperature field contour lines. Spatial coordinates are recorded at singular points where a single temperature contour line splits into multiple lines or diverges from convergence. These branching point coordinates, together with the aforementioned temperature gradient abrupt change point coordinates, constitute complete information on the location of the temperature gradient abrupt change line.

[0024] In practical implementation, the generation of the expected segregation path of silver in the melt relies on tracking the main heat output path, which is a virtual trajectory extending from the core of the high-energy thermal zone along the direction of most significant temperature decay. Along the main heat output path, the system calculates the temperature difference between each path point and its upstream adjacent path point at set spatial step intervals, and divides this temperature difference by the spatial distance between the two points to obtain the average temperature decay rate of that path segment. Optionally, the calculation of the average temperature decay rate can incorporate a time factor, expressed by the following formula: in: This represents the average temperature decay rate calculated over the specified path segment. It is the number of spatial steps within the path segment. and These represent the instantaneous temperature values ​​at the path point and its downstream adjacent path points, respectively. and These are the spatial coordinate vectors of the corresponding path points. Connecting the average temperature decay rates of all continuous path segments along the main heat output path in spatial order forms a curve describing the heat loss rate, defined as the expected segregation path of silver in the melt. In some embodiments, the core location of the high-energy thermal zone, the location of the steep temperature gradient line, and the expected segregation path of silver in the melt are assigned different data layer identifiers. Finally, through three-dimensional spatial data fusion and visualization encoding technology, a digital thermodynamic state cloud map containing multi-dimensional thermodynamic property information is generated. The thermodynamic state cloud map is stored in a grid data format, with each grid cell containing its spatial coordinates and its corresponding category attribute and quantization parameter value.

[0025] In one embodiment of the present invention, the viscous convection resistance surface of the molten pool is generated by fusing the melt surface tension distribution value with a preset melt flow resistance model. See also... Figure 2The gradient calculation of the surface tension distribution of the melt is performed to obtain the force direction and force field intensity value of each micro-region of the molten pool. Combined with a pre-set alloy element database and furnace lining physical property database, corresponding rheological parameters are matched for each micro-region of the molten pool. Based on the force direction, force field intensity value, and rheological parameters, the effective viscosity coefficient and convection inhibition factor of the melt for each micro-region of the molten pool are obtained from a table. The effective viscosity coefficient and convection inhibition factor of the melt are used to parametrically adjust the benchmark fluid dynamics model, and the theoretical convective mass transfer resistance value of each micro-region of the molten pool is calculated. The theoretical convective mass transfer resistance values ​​of all micro-regions of the molten pool within the entire molten pool range are fitted with a spatial surface to construct a viscous convection resistance surface characterizing the degree of hindrance of melt flow to silver diffusion. The construction steps of the pre-set melt flow resistance model are as follows: collecting monitoring data of historical smelting processes of various silver-containing stainless steels with different compositions under pre-set smelting parameters, including historical melt surface tension distribution, historical temperature field distribution, historical silver concentration field distribution, and historical evaluation of the final silver element distribution of ingots. For each different composition, a correspondence between the alloy composition and its standard melt viscosity coefficient and surface tension temperature coefficient at multiple preset temperatures is established, forming an alloy composition-physical property parameter mapping table. For each historical smelting process, based on the historical temperature field distribution and the alloy composition-physical property parameter mapping table, a dynamic spatiotemporal field of the melt viscosity coefficient during the historical smelting process is generated. Based on the historical melt surface tension distribution and historical silver concentration field distribution, the theoretical value of the convection velocity field driven by the surface tension gradient and concentration gradient inside the melt is calculated. The final silver element distribution evaluation of the historical ingot is quantified into a uniformity index of silver element distribution. Using the dynamic spatiotemporal field of the melt viscosity coefficient, the theoretical value of the convection velocity field, and the parameters in the melt flow resistance model as model inputs, and the uniformity index of silver element distribution as the optimization objective, the model parameters are fitted to obtain a preset melt flow resistance model for predicting the convection mass transfer resistance value.

[0026] In specific implementation, the surface tension distribution value of the melt provided by the molten pool monitoring module is processed. The surface tension distribution value is a scalar field defined on the three-dimensional spatial grid of the molten pool. Gradient calculation is performed on the surface tension distribution value of the melt. The gradient calculation uses the central difference method to calculate the partial derivatives of each grid point along three spatial directions, thereby solving for the force direction and force field intensity value of each molten pool micro-element region. The force direction is determined by the gradient direction, and the force field intensity value is the magnitude of the gradient vector. In some embodiments, the division of the molten pool micro-element region is consistent with the temperature monitoring grid to ensure data spatial alignment. Each molten pool micro-element region corresponds to a unique spatial coordinate and a set of calculated force direction and force field intensity values. Combined with a pre-set alloy element database and furnace lining physical property database, corresponding rheological parameter matching is performed for each molten pool micro-element region. The alloy element database stores the basic physical properties such as viscosity and density of different alloy compositions at a reference temperature, while the furnace lining physical property database provides the thermophysical properties of the furnace lining material in contact with the molten pool. Based on the direction of the force, the strength of the force field, and the rheological parameters, the effective viscosity coefficient and convection inhibition factor of the melt corresponding to each micro-element region of the molten pool are obtained by looking up a table. The table lookup operation is based on the real-time temperature, alloy composition ratio, and surface tension gradient of the micro-element region of the molten pool for multi-dimensional indexing.

[0027] In practical implementation, the baseline fluid dynamics model is parameterized using the effective melt viscosity coefficient and convection inhibition factor. The baseline fluid dynamics model employs a modified Navier-Stokes equation to describe the melt flow. The theoretical convective mass transfer resistance value for each micro-element region of the molten pool is calculated. This calculation considers the influence of the effective melt viscosity coefficient on momentum dissipation within the flow field and the attenuation effect of the convection inhibition factor on the diffusion flux of silver. The formula for calculating the theoretical convective mass transfer resistance value can be understood as follows: in: This represents the theoretical convective mass resistance value. Indicates the effective viscosity coefficient of the melt. Indicates convection inhibition factor, The velocity gradient tensor represents the velocity field of the melt. The theoretical convective mass transfer resistance values ​​of all infinitesimal regions of the molten pool are fitted with a spatial surface using three-dimensional spline interpolation or Kriging interpolation methods to construct a continuous and smooth viscous convective resistance surface. The viscous convective resistance surface is presented as a two-dimensional surface covering the three-dimensional spatial projection of the molten pool, and the height value of each point on the surface represents the intensity of convective mass transfer resistance at that location.

[0028] In some embodiments, the construction step of the preset melt flow resistance model involves historical data collection, collecting monitoring data of historical smelting processes of silver-containing stainless steel with various compositions under preset smelting parameters. The monitoring data includes historical melt surface tension distribution, historical temperature field distribution, historical silver concentration field distribution, and historical ingot final silver element distribution evaluation. For each different composition, a correspondence is established between the alloy composition and its standard melt viscosity coefficient and surface tension temperature coefficient at multiple preset temperatures, forming an alloy composition-physical property parameter mapping table. This mapping table is stored in a database table format, supporting rapid querying by composition and temperature. For each historical smelting process, based on the historical temperature field distribution and the alloy composition-physical property parameter mapping table, a dynamic spatiotemporal field of the melt viscosity coefficient during the historical smelting process is generated. This dynamic spatiotemporal field reflects the viscosity changes with time and space. Based on the historical melt surface tension distribution and historical silver concentration field distribution, the theoretical value of the convection velocity field driven by the surface tension gradient and concentration gradient within the melt is calculated. The calculation of the theoretical value of the convection velocity field is based on the coupling equation of Marangoni convection and solute convection.

[0029] In practical implementation, the final silver element distribution evaluation of historical ingots is quantified as a uniformity index of silver element distribution. This uniformity index can be represented by the standard deviation or coefficient of variation of the silver concentration distribution. Using the dynamic spatiotemporal field of the melt viscosity coefficient, the theoretical value of the convection velocity field, and parameters from the melt flow resistance model as model inputs, and the uniformity index of silver element distribution as the optimization objective, the model parameters are fitted. The fitting process uses the least squares method or a genetic algorithm to obtain a preset melt flow resistance model that predicts the convection mass transfer resistance value. It can be understood that when the melt flow resistance model is applied online, it directly calls the model function, combining the real-time monitored melt surface tension distribution value to quickly output the viscous convection resistance surface.

[0030] In one embodiment of the present invention, the fluctuation pattern of the molten pool composition is identified based on continuous monitoring records of silver concentration to predict the probability of silver phase aggregation and the location of segregation nuclei in different parts of the ingot. The continuous monitoring records of silver concentration are segmented according to a preset casting stage, and the variance of silver concentration fluctuation within each casting stage is calculated and its correlation coefficient with adjacent samples. Molten pool areas where the variance and correlation coefficient exceed the process stability threshold are marked as silver concentration abnormal fluctuation areas. The spatial evolution morphology and temporal change sequence of the silver concentration abnormal fluctuation areas are analyzed to identify silver concentration gradient abrupt change bands that conform to the micro-segregation characteristics in alloy solidification theory, and these silver concentration gradient abrupt change bands are determined as potential high-incidence areas of silver phase aggregation probability. Within the silver concentration abnormal fluctuation areas, high-frequency oscillation signals of silver concentration are further extracted, and the location of segregation nuclei is located using signal power spectral density identification technology. The location of segregation nuclei corresponds to the spatial core point with the strongest high-frequency oscillation power of silver concentration. An aggregation intensity level label is assigned to each identified high-incidence area of ​​silver phase aggregation probability and segregation nuclei location. The probability of silver phase aggregation and the location of segregation nuclei are mapped to uniform distribution control targets to generate alloy melting process parameter adjustment schemes with process risk labels. A spatial framework for uniform distribution control targets is established, and spatial overlay analysis is performed on high-incidence areas of silver phase aggregation probability labeled with aggregation intensity level tags and segregation nuclei locations. For uniform distribution control target units that fall into any high-incidence area of ​​silver phase aggregation probability or whose distance from any segregation nuclei location is less than the size of the critical solidification unit, the corresponding segregation risk type and risk level are recorded in their attributes. Based on the risk suppression strategy, one or more process parameter adjustment suggestions are generated for each uniform distribution control target unit with segregation risk type and risk level. The process parameter adjustment suggestions involve the local electromagnetic stirring intensity, cooling rate, or holding time of the target unit. The uniform distribution control target units that do not fall into the risk area, as well as all generated process parameter adjustment suggestions, are integrated into an alloy melting process parameter adjustment scheme with process risk labels. Each control unit in the scheme includes location coordinates, segregation risk labels, and corresponding process parameter suggestions.

[0031] In practical implementation, the identification of molten pool composition fluctuation patterns begins with the processing of continuous monitoring records of silver concentration, which are obtained from immersion spectral probes or laser-induced breakdown spectroscopy online monitoring systems. The continuous monitoring records of silver concentration are segmented according to preset casting stages, based on key time nodes defined in the smelting process specifications such as tapping, refining, and quenching. The variance of silver concentration fluctuation within each casting stage is calculated, along with the correlation coefficient between adjacent samples. In some embodiments, the variance and correlation coefficient calculations are performed on each independent analysis unit within the molten pool, divided into grids. The size of the analysis unit matches the spatial resolution of the molten pool monitoring module. Molten pool regions where the variance and correlation coefficient exceed the process stability threshold are marked as silver concentration abnormal fluctuation areas. The process stability threshold is obtained statistically from historical qualified melt data. The markings of silver concentration abnormal fluctuation areas are recorded in the system's storage as three-dimensional spatial blocks. The spatial evolution morphology and temporal variation sequence of the abnormal silver concentration fluctuation zone were analyzed. The spatial evolution morphology analysis focused on the shape changes, volume expansion or contraction trends of the abnormal silver concentration fluctuation zone in continuous time frames. The temporal variation sequence analysis focused on the temporal fluctuation characteristics of the average concentration within the abnormal silver concentration fluctuation zone. The abrupt change zone of silver concentration gradient that conforms to the micro-segregation characteristics in alloy solidification theory was identified. The identification criterion for the abrupt change zone of silver concentration gradient is a step change in concentration over a micrometer-scale spatial distance. The abrupt change zone of silver concentration gradient is identified as a potential high-probability area for silver phase aggregation.

[0032] In practical implementation, within the region of abnormal silver concentration fluctuations, high-frequency oscillation signals of silver concentration are further extracted. These high-frequency oscillation signals are obtained by subtracting the moving average of the original concentration time-series data. Signal power spectral density (SPD) identification technology is then used to locate the segregation nucleus generation positions. This technology calculates the power spectrum at different spatial locations by performing a Fourier transform on the high-frequency oscillation signal; the segregation nucleus generation positions correspond to the spatial core points where the high-frequency oscillation power of silver concentration is strongest. Each identified high-probability silver phase aggregation region and segregation nucleus generation position is assigned an aggregation intensity level label. This label is determined based on the steepness of the silver concentration gradient abrupt change band and the amplitude of the high-frequency oscillation power of silver concentration. The aggregation intensity level label can be represented using numerical levels or color coding. Optionally, the identification of abnormal silver concentration fluctuation regions can incorporate more complex time-series pattern matching algorithms to distinguish between random fluctuations and trend-based abnormal fluctuations.

[0033] A spatial framework for the uniform distribution control target is established, which is a three-dimensional mesh model defining the target silver concentration and its allowable fluctuation range. High-probability silver phase aggregation areas labeled with aggregation intensity levels are spatially superimposed with segregation nucleus formation locations. This spatial superposition analysis aligns the risk region with the mesh through coordinate system transformation, mapping the spatial coordinates of the risk area onto the spatial framework of the uniform distribution control target. For uniform distribution control target units falling within any high-probability silver phase aggregation area or within a distance of less than the critical solidification unit size from any segregation nucleus formation location, the corresponding segregation risk type and risk level are recorded in their attributes. The critical solidification unit size is calculated using a solidification model based on alloy composition and cooling rate. In some embodiments, the segregation risk type can be distinguished as "aggregation type" and "nucleation type," corresponding to the influence of high-probability silver phase aggregation areas and segregation nucleus formation locations, respectively. Based on risk mitigation strategies, one or more process parameter adjustment suggestions are generated for each uniform distribution control target unit with a segregation risk type and risk level. The risk mitigation strategies are stored in a database, recording recommended process intervention measures under different risk types and levels. The process parameter adjustment recommendations involve the local electromagnetic stirring intensity, cooling rate, or holding time of the target unit. For example, for a high-risk clustered risk, an adjustment recommendation of "increasing the local electromagnetic stirring intensity by 15%" might be generated. It can be understood that the generation of process parameter adjustment recommendations is a rule-based or optimization model-based decision-making process. The uniformly distributed control target units that do not fall into the risk zone, along with all generated process parameter adjustment recommendations, are integrated into an alloy melting process parameter adjustment scheme with process risk annotations. Each control unit in the scheme includes its location coordinates, segregation risk annotation, and corresponding process parameter recommendations.

[0034] See Figure 3In the identification of compositional fluctuation patterns throughout the entire process of smelting and casting silver-containing martensitic antibacterial stainless steel, process stability assessment and segregation risk prediction were achieved by quantifying the variance of silver concentration fluctuations and the time-series correlation coefficient at each stage. Specifically, the continuous monitoring records of silver concentration were segmented into five key casting stages: tapping, refining, quenching, early solidification, and mid-solidification. The variance of silver concentration fluctuations (characterizing concentration dispersion) and the time-series correlation coefficient between adjacent samples (characterizing concentration temporal continuity) were calculated for each stage, with a process stability threshold (concentration fluctuation variance = 0.100) used as the risk assessment benchmark. The quantitative results show that during the tapping stage, the concentration fluctuation variance (approximately 0.020) was far below the threshold, but the time-series correlation coefficient (approximately 0.92) was high, indicating that although the silver concentration fluctuations were relatively smooth at this stage, the temporal correlation was strong, and the compositional transfer trend was stable. Refining stage: The concentration fluctuation variance (approximately 0.050) remains below the threshold, and the time-series correlation coefficient (approximately 0.78) has decreased, reflecting the gradual emergence of component homogenization during refining and a weakening of concentration time-series continuity. Calming stage: The concentration fluctuation variance (approximately 0.120) exceeds the process stability threshold for the first time, and the time-series correlation coefficient (approximately 0.53) further decreases, indicating that while the molten pool is stabilizing and homogenizing, local concentration disturbances begin to accumulate, creating a risk for subsequent segregation. Initial solidification stage: The concentration fluctuation variance (approximately 0.180) reaches its peak value throughout the process, and the time-series correlation coefficient (approximately 0.42) drops to its lowest point, indicating intense convection at the solid-liquid interface and a significant silver repulsion effect in the initial solidification stage. The concentration time-series memory fracture represents a high-risk window for silver phase aggregation and segregation nucleus formation. Mid-solidation: The concentration fluctuation variance (approximately 0.080) falls below the threshold, while the time-series correlation coefficient (approximately 0.68) rebounds, indicating that as the solidification process progresses, melt fluidity weakens, concentration fluctuations become smoother, and the risk of segregation gradually converges. By comparing the coupling characteristics of fluctuation variance and time-series correlation coefficient at each stage, abnormal silver concentration fluctuation zones can be accurately identified, providing a quantitative basis for subsequent prediction of silver phase aggregation probability, location of segregation nuclei, and optimization of process parameters (electromagnetic stirring intensity, cooling rate, etc.).

[0035] In one embodiment of the present invention, the alloy melting process parameter adjustment scheme with process risk labeling can be used for solidification process simulation and iterative optimization. From the alloy melting process parameter adjustment schemes with process risk labeling, the control unit with the highest segregation risk level is selected as the first batch of process intervention units. Based on the solidification sequence in the thermodynamic state cloud diagram, the main solidification conditions for microstructure evolution are set. Using a dendrite growth simulator based on a phase field model, the solute redistribution and silver phase precipitation images generated when the first batch of process intervention units solidify under the main solidification conditions are calculated. For each virtual material element of the first batch of process intervention units, the phase field variables and solute concentration field are initialized under the main solidification conditions. Based on the main solidification conditions, the current thermodynamic parameters of the material element, and the diffusion coefficient of silver in the iron matrix, the phase field variables and solute concentration field are driven to evolve in the spatiotemporal domain. The solid-liquid interface advancement and solute repulsion process of each material element during the phase field evolution process are tracked, and the concentration distribution of silver in the solid and liquid phases is recorded. The simulation region is discretized into a microgrid. The final silver concentration of each microgrid cell at the solidification completion time is statistically analyzed and compared with the initial average concentration to obtain the concentration deviation value. The concentration deviation values ​​of all microgrid cells are integrated to generate a solute redistribution and silver precipitation image covering the entire simulation region. The boundary of the severely segregated micro-segregation region is determined based on a preset concentration deviation threshold. The solute redistribution and silver precipitation image is overlaid with a uniform distribution control target to evaluate the impact of micro-segregation on the uniformity of silver distribution in the final tool matrix. Based on the evaluation results, the expected silver distribution uniformity of significantly affected downstream control units is reduced. Based on the reduced uniformity index and segregation risk level, all control units to be intervened are reordered. From the reordered control units to be intervened, the optimal intervention unit is selected and added to the execution sequence. The steps from dendrite growth simulation to cell rearrangement are repeated until all control units to be intervened complete the process decision or reach the preset process control limit, ultimately generating an optimized performance control process scheme.

[0036] In practice, the iterative optimization process begins with loading and analyzing the parameter adjustment schemes for the alloy melting process with process risk labels. From these schemes, the system selects the control unit with the highest segregation risk level as the first batch of process intervention units according to a preset sorting rule. The segregation risk level can be sorted based on the numerical value of the risk level label; a higher value indicates a higher risk level. When multiple control units have the same risk level, a secondary sort can be performed based on their spatial location and the order of solidification in the thermodynamic state cloud map. Based on the solidification order in the thermodynamic state cloud map, the primary solidification conditions for microstructure evolution are set. These conditions include the initial conditions of the temperature gradient, local cooling rate, and solid-liquid interface propagation speed experienced by the control unit during solidification. A dendrite growth simulator based on a phase-field model is used to calculate the solute redistribution and silver phase precipitation images generated when the first batch of process intervention units solidify under the primary solidification conditions. For each virtual material element in the first batch of process intervention units, the phase field variable and solute concentration field are initialized under the main solidification conditions. The phase field variable is used to distinguish between the solid phase and the liquid phase, and the solute concentration field records the initial distribution of silver element in the material element.

[0037] In practice, based on the main solidification conditions, the thermodynamic parameters of the current material volume element, and the diffusion coefficient of silver in the iron matrix, the phase field variables and solute concentration field are driven to evolve in the spatiotemporal domain. The evolution process is achieved by solving the coupled phase field equations and diffusion equations. The solid-liquid interface propagation and solute repulsion processes of each material volume element during the phase field evolution are tracked, and the concentration distribution of silver in the solid and liquid phases is recorded. The position of the solid-liquid interface is determined by the isosurface of the phase field variables, and the solute repulsion process is calculated based on the difference in the distribution coefficient of silver in the solid and liquid phases. The simulation region is discretized into a microgrid, and the final concentration of silver in each microgrid element at the time of solidification completion is statistically analyzed and compared with the initial average concentration to obtain the concentration deviation value. The concentration deviation values ​​of all micro-grid units are integrated to generate solute redistribution and silver precipitation images covering the entire simulation area. The solute redistribution and silver precipitation images are visualized in the form of two-dimensional pseudo-color images or three-dimensional concentration cloud maps. The boundaries of the severely segregated micro-regions are determined based on the preset concentration deviation threshold. The severely segregated micro-regions refer to the set of grid units whose concentration deviation values ​​exceed the threshold.

[0038] The solute redistribution and silver phase precipitation images are overlaid with a uniform distribution control target to assess the impact of microsegregation on the uniformity of silver distribution in the final tool matrix. The overlay operation compares the simulated silver concentration distribution with the desired uniform distribution control target within the same spatial coordinate system. Based on the assessment results, a reduction calculation is performed on the desired silver distribution uniformity of significantly affected downstream control units. Downstream control units are defined as those located after the current first batch of process intervention units in the solidification sequence and exhibiting thermal or solute interactions with the current unit. The reduction calculation is based on the distance between the downstream control unit and the current simulation region, as well as the intensity of solute redistribution in the current simulation. One reduction calculation formula is as follows: in: Indicators representing uniformity after reduction This represents the original desired silver distribution uniformity index of the downstream control unit. It is a decay coefficient that is positively correlated with the severity of microsegregation in the current simulated region. This represents the normalized spatial distance between the downstream control unit and the centroid of the current simulation region. Based on the reduced uniformity index and segregation risk level, all control units to be intervened are reordered. The sorting logic can assign a higher weight to the reduced uniformity index. From the reordered control units to be intervened, the optimal intervention unit is selected and added to the execution sequence. The selection rule can be the unit with the highest comprehensive score or the greatest risk mitigation potential. The steps from dendrite growth simulation to unit rearrangement are repeated until all control units to be intervened complete the process decision or reach the preset process control limit, ultimately generating an optimized performance control process scheme.

[0039] In some embodiments, the iterative optimization process may generate a record of decision sequences processed by a control unit, as shown in Table 1.

[0040] Table 1: Example of iterative optimization sorting of control units to be intervened See Figure 4In the phase-field simulation of the solidification process of silver-containing martensitic antibacterial stainless steel, this figure visually presents the microscopic segregation behavior of silver phase precipitation and solute redistribution. The figure uses the X and Y coordinates of the microscopic mesh as spatial coordinates. The color scale on the right quantifies the relative deviation value of silver concentration, with colors ranging from green to red corresponding to deviation values ​​from low to high (approximately 0.10~0.25). The red core area represents the segregation zone where silver is significantly enriched, while the green outer area represents the relatively depleted silver concentration zone. The black circular boundary marks the boundary of the microscopic segregation region, serving as the boundary between the severely segregated area and the matrix region based on a preset concentration deviation threshold. This visually reflects the solute aggregation morphology formed by silver elements under the repulsive effect at the solid-liquid interface during solidification. The simulation results are derived from dendrite growth calculations based on a phase-field model: by initializing phase-field variables and solute concentration fields, the phase-field evolution equation and solute diffusion equation are solved in a coupled manner, and the repulsion and redistribution behavior of silver elements during the solid-liquid interface process is tracked. Finally, the deviation values ​​of silver concentration in each micro-grid unit from the initial average concentration are visualized, providing a direct quantitative basis for evaluating the impact of micro-segregation on the uniformity of silver element distribution in the tool matrix and optimizing smelting process parameters.

[0041] In one embodiment of the present invention, after small-scale stainless steel samples are fabricated according to a preliminary process plan, a dynamic fine-tuning step can be performed. Actual composition distribution data, microhardness data, and silver phase morphology image data of the samples are collected. The actual composition distribution data is compared with the predicted composition distribution of the corresponding uniform distribution control target to generate a local composition distribution correction field. For each sample detection area providing actual composition distribution data, the average silver concentration measurement value within a specific detection surface is extracted. From the corresponding region of the uniform distribution control target, the predicted average silver concentration value within the same specific detection surface of the sample detection area is read. The ratio of the measured average silver concentration value to the predicted average silver concentration value is calculated as the composition prediction correction coefficient for the sample detection area. Using spatial interpolation technology, the composition prediction correction coefficients of each sample detection area are extended to the simulated three-dimensional space of the entire tool material, forming a continuous local composition distribution correction field. The local composition distribution correction field is multiplied point-by-point with the original uniform distribution control target to obtain an updated distribution map of the silver-containing martensitic antibacterial stainless steel performance regulation target. By analyzing silver phase morphology image data, the size distribution, shape factor, and dispersion of the silver phase in the matrix are statistically analyzed to identify regions where the actual silver precipitation behavior deviates from theoretical predictions. Combining information from the local compositional distribution correction field and the regions where the actual silver precipitation behavior deviates, the expected silver distribution of remaining process control units that have not yet been executed or can be adjusted is reassessed. Based on the reassessment results, the process parameters or intervention priorities of the remaining process control units are adjusted to generate dynamically updated performance control process schemes.

[0042] In practice, the dynamic fine-tuning step is initiated after small-scale stainless steel samples are fabricated according to the preliminary process plan. These small-scale stainless steel samples are representative samples taken from the test ingots guided by the preliminary process plan. Actual composition distribution data, microhardness data, and silver phase morphology image data of the samples are collected. The actual composition distribution data is obtained through electron probe microscopy or micro-area X-ray fluorescence spectroscopy. The microhardness data is measured using a Vickers or Knoop hardness tester at gridded points on a specific cross-section of the sample. The silver phase morphology image data is acquired using a scanning electron microscope or optical metallurgical microscope under multiple fields of view. The actual composition distribution data is compared with the predicted composition distribution of the corresponding uniform distribution control target to generate a local composition distribution correction field. This comparison is performed in each detection area of ​​the sample. For each sample detection area providing actual composition distribution data, the average silver concentration measurement value within a specific detection surface is extracted. This specific detection surface is typically a cross-section perpendicular or parallel to the main axis of the ingot. From the corresponding region of the uniformly distributed control target, the predicted average concentration of silver in the sample detection area within the same specific detection surface is read. The predicted average concentration of silver is extracted from the three-dimensional distribution model output by the distribution prediction module through coordinate mapping.

[0043] In practice, the ratio of the measured average silver concentration to the predicted average silver concentration is calculated and used as the composition prediction correction coefficient for the sample detection area. This correction coefficient reflects the accuracy deviation of the model prediction in that local region. Using spatial interpolation techniques, the composition prediction correction coefficients for each sample detection area are extended to the entire simulated three-dimensional space of the tool material, forming a continuous local composition distribution correction field. Spatial interpolation techniques can employ inverse distance weighting, kriging, or radial basis function methods. The local composition distribution correction field is then multiplied point-by-point with the original uniform distribution control target to obtain the updated distribution map of the silver-containing martensitic antibacterial stainless steel performance regulation target. The update operation is performed at each node of the three-dimensional mesh. It can be understood that the composition prediction correction coefficient... The calculation formula is: in: This represents the average measured concentration of silver. This represents the predicted average concentration of silver. In some embodiments, when the number of sample detection areas is limited, spatial interpolation introduces uncertainty, which can be identified by setting a confidence interval for the interpolation results. Optionally, for microhardness data, an empirical relationship between it and local silver concentration can be established to indirectly verify or supplement the component distribution correction field.

[0044] Analyzing silver phase morphology image data, the size distribution, shape factor, and dispersion of the silver phase in the matrix are statistically analyzed. Size distribution is obtained by measuring the equivalent circle diameter of all visible silver phase particles in the image. The shape factor is calculated using the relationship between particle perimeter and area. Dispersion is obtained by calculating the number of silver phase particles per unit area or by statistically analyzing nearest neighbor distances. Regions where actual silver phase precipitation behavior deviates from theoretical predictions are identified. This deviation is identified by comparing the actual measured silver phase size, shape, and dispersion with theoretical values ​​predicted based on solidification kinetics simulations using threshold comparisons. Combining information from local compositional distribution correction fields and regions of deviation in actual silver phase precipitation behavior, the expected silver distribution of unexecuted or adjustable remaining process control units is reassessed. This reassessment comprehensively considers both compositional distribution corrections and deviations in silver phase precipitation morphology to predict the impact on the final performance control target. In some embodiments, remaining process control units refer to the set of control commands in the alloy melting process parameter adjustment scheme that have not yet been applied to actual melting, or process parameters that can still be adjusted in subsequent batches. Based on the reassessment results, the process parameters or intervention priorities of the remaining process control units are adjusted. Adjustments to process parameters include modifying the power curve of the electromagnetic stirring or the local cooling rate. The intervention priorities are adjusted based on the degree of influence of each unit on the final distribution uniformity target after the reassessment. A dynamically updated performance control process scheme is generated, which will serve as the process input for subsequent formal smelting production or the next round of testing.

[0045] See Figure 5The cloud map visually represents the concentration gradient and segregation trend of silver in the two-dimensional plane of the molten pool. Specifically, the silver concentration distribution across the entire molten pool exhibits significant spatial heterogeneity: low concentration areas are concentrated in the upper left region (X≈20–45 mm, Y≈50–75 mm) and lower right region (X≈75–100 mm, Y≈0–20 mm), with a concentration range of approximately 1.12–1.44 wt.%, corresponding to the blue-cyan areas in the cloud map. This reflects the high melt convection resistance or solute repulsion effect driven by the temperature gradient in these areas, which inhibits the local enrichment of silver. High-concentration zone: Mainly distributed in the lower left region (X≈20–45mm, Y≈0–25mm) and upper right region (X≈75–100mm, Y≈50–80mm) of the molten pool, with a concentration range of approximately 2.08–2.40 wt.%, corresponding to the orange-red area in the contour map. This is a potential high-incidence area for silver segregation and is highly coupled with the heat output path of the high-energy hot zone in the thermodynamic state contour map and the low-resistance channel of the viscous convection resistance surface. Transition zone: A large-area transition zone with a concentration of 1.60–1.92 wt.% exists throughout the region, forming a concentration buffer interface between the low-concentration and high-concentration zones, reflecting the dynamic balance of silver convection mass transfer and diffusion within the molten pool. The cloud map can be directly mapped to the spatial framework of the uniformly distributed control target, providing a basis for marking segregation risk for the process optimization module: high concentration areas are marked as areas with a high probability of silver phase aggregation, and their core locations can be further located as candidate points for segregation nucleus generation. Subsequently, by adjusting process parameters such as local electromagnetic stirring intensity and cooling rate, the excessive enrichment of silver in this area will be suppressed, the uniformity of silver distribution in the final ingot will be improved, and the antibacterial performance stability of the kitchen knife matrix will be ensured.

[0046] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives, characterized in that, The system includes: The molten pool monitoring module continuously records the instantaneous temperature, surface tension distribution, and silver concentration of each region of the molten pool during the smelting process of silver-containing martensitic antibacterial stainless steel. The thermal analysis module performs spatial and temporal correlation analysis on the instantaneous temperature values ​​of each region of the molten pool to generate a thermodynamic state cloud map of the molten pool. The flow analysis module calculates the viscous convection resistance surface of the molten pool by fusing the surface tension distribution value of the melt with a preset melt flow resistance model. The viscous convection resistance surface is used to quantify the obstruction effect of different molten pool regions on the convective mass transfer of silver elements. The distribution prediction module determines the uniform distribution control target of silver elements during the solidification process of the ingot based on the coupling relationship between the thermodynamic state cloud map and the viscous convection resistance surface, and performs a simulation of the silver phase precipitation process based on solidification kinetics on the uniform distribution control target to predict the initial distribution morphology of silver elements in the matrix. The process optimization module identifies the fluctuation pattern of the molten pool composition based on the continuous monitoring records of the silver concentration, predicts the probability of silver phase aggregation and the location of segregation nuclei in different parts of the ingot, maps the probability of silver phase aggregation and the location of segregation nuclei to the uniform distribution control target, and generates an alloy smelting process parameter adjustment scheme with process risk labeling.

2. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 1, characterized in that, The step of performing spatial and temporal correlation analysis on the instantaneous temperature values ​​of each region of the molten pool to generate a thermodynamic state cloud map of the molten pool includes: The thermodynamic state cloud map includes the core location of the high-energy hot zone, the location of the steep change line of the temperature gradient, and the expected segregation path of silver in the melt. The instantaneous temperature values ​​from all synchronous molten pool monitoring devices are time-synchronized to form a snapshot of the spatial temperature distribution at the same moment; Heat transfer path tracing is performed on a series of consecutive temperature spatial distribution snapshots to extract high-temperature regions that are stable in the time dimension and identify the core location of the high-energy thermal zone. Around the core of the high-energy thermal zone, dense bands of temperature isopleths and split points of temperature diffusion direction are detected, and the spatial coordinates of the dense bands and split points are recorded to form the position of the steep change line of the temperature gradient. Along the main heat output path of the high-energy hot zone, the average temperature decay rate is calculated at fixed step intervals, and the calculated average temperature decay rates are connected in sequence along the path to form the expected segregation path of the silver element in the melt. The core location of the high-energy hot zone, the location of the steep change line of the temperature gradient, and the expected segregation path of the silver element in the melt are integrated and encoded into the thermodynamic state cloud map.

3. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 2, characterized in that, The viscous convection resistance surface of the molten pool is generated by fusing the surface tension distribution value of the melt with a preset melt flow resistance model, including: Gradient calculations are performed on the surface tension distribution values ​​of the melt to obtain the force direction and force field strength values ​​of each micro-element region of the molten pool. By combining a pre-set database of alloying elements and a database of physical properties of furnace lining, corresponding rheological parameters are matched for each micro-element region of the molten pool. Based on the force direction, force field strength value, and rheological property parameters, the effective viscosity coefficient and convection inhibition factor of the melt corresponding to each micro-element region of the molten pool are obtained by looking up the table. The effective viscosity coefficient of the melt and the convection inhibition factor are used to parameterize the reference fluid dynamics model, and the theoretical convection mass resistance value of each micro-element region of the molten pool is calculated. By fitting the theoretical convection mass resistance values ​​of all micro-regions within the entire molten pool to a spatial surface, a viscous convection resistance surface characterizing the degree to which the melt flow hinders the diffusion of silver is constructed.

4. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 3, characterized in that, Based on the continuous monitoring records of the silver concentration, the compositional fluctuation pattern of the molten pool is identified to predict the probability of silver phase aggregation and the location of segregation nuclei in different parts of the ingot, including: The continuous monitoring records of silver concentration are divided into segments according to the preset pouring stages, and the variance of silver concentration fluctuation within each pouring stage is calculated and the correlation coefficient between the two samples is calculated. The molten pool regions where the variance and correlation coefficient exceed the process stability threshold are marked as silver concentration abnormal fluctuation regions; By analyzing the spatial evolution morphology and temporal variation sequence of the abnormal silver concentration fluctuation zone, abrupt changes in the silver concentration gradient that conform to the microsegregation characteristics in alloy solidification theory are identified, and the abrupt changes in the silver concentration gradient are determined to be potential high-probability areas for silver phase aggregation. Within the abnormal fluctuation region of silver concentration, the high-frequency oscillation signal of silver concentration is further extracted, and the location of the segregation nucleus generation is located using signal power spectral density identification technology. The location of the segregation nucleus generation corresponds to the spatial core point where the high-frequency oscillation power of silver concentration is strongest. Each identified high-probability silver phase aggregation region and the location of segregation nucleus generation is assigned an aggregation intensity level label.

5. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 4, characterized in that, Mapping the silver phase aggregation probability and segregation nucleus generation location to the uniform distribution control target generates an alloy smelting process parameter adjustment scheme with process risk annotations, including: A spatial framework for the uniformly distributed control target is established, and the high-probability silver phase aggregation area labeled with aggregation intensity level is spatially superimposed with the segregation nucleus generation location for analysis. For uniformly distributed control target units that fall into any of the high probability zones of silver phase aggregation or whose distance from any of the segregation nuclei generation locations is less than the size of the critical solidification unit, the corresponding segregation risk type and risk level are recorded in their attributes. Based on the risk suppression strategy, one or more process parameter adjustment suggestions are generated for each uniformly distributed control target unit with the aforementioned segregation risk type and risk level. The process parameter adjustment suggestions involve the local electromagnetic stirring intensity, cooling rate, or holding time of the target unit. The uniformly distributed control target units that do not fall into the risk zone and all the generated process parameter adjustment suggestions are integrated into the alloy melting process parameter adjustment scheme with process risk labeling. Each control unit in the scheme includes position coordinates, segregation risk labeling and corresponding process parameter suggestions.

6. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 5, characterized in that, Also includes: The iterative optimization module is used to simulate and iteratively optimize the solidification process of the alloy melting process parameter adjustment scheme with process risk labeling. From the alloy melting process parameter adjustment schemes marked with process risks, the control unit with the highest segregation risk level is selected as the first batch of process intervention units. Based on the solidification sequence in the thermodynamic state cloud diagram, the main solidification conditions for tissue evolution are set; A dendrite growth simulator based on a phase field model was used to calculate the solute redistribution and silver phase precipitation images generated when the first batch of process intervention units solidified under the main solidification conditions. The solute redistribution and silver precipitation images are superimposed on the uniform distribution control target to evaluate the degree of influence of microsegregation on the uniformity of silver element distribution in the final tool matrix. Based on the assessment results, the expected silver distribution uniformity of the significantly affected downstream control units was reduced, and all control units to be intervened were reordered based on the reduced uniformity index and segregation risk level. From the reordered control units to be intervened, the optimal intervention unit is selected and added to the execution sequence. The steps from dendrite growth simulation to unit rearrangement are repeated until all control units to be intervened complete the process decision or reach the preset process control limit, and finally the optimized performance control process scheme is generated.

7. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 6, characterized in that, The dendrite growth simulator based on the phase field model is used to calculate the solute redistribution and silver phase precipitation images generated when the first batch of process intervention units solidifies under the main solidification conditions, including: For each virtual material element of the first batch of process intervention units, the phase field variables and solute concentration field are initialized under the main solidification conditions; Based on the main solidification conditions, the thermodynamic parameters of the current material volume element, and the diffusion coefficient of silver in the iron matrix, the phase field variables and solute concentration field are driven to evolve and be calculated in the spatiotemporal domain. The solid-liquid interface propagation and solute repulsion processes of each material volume element during phase field evolution were tracked, and the concentration distribution of silver in the solid and liquid phases was recorded. The simulation region is discretized into a micro-grid, and the final concentration of silver in each micro-grid unit at the moment of solidification is calculated. The concentration deviation value is obtained by comparing it with the initial average concentration. The concentration deviation values ​​of all micro-grid units are integrated to generate a solute redistribution and silver phase precipitation image covering the entire simulation area, and the boundary of the severely segregated micro-segregation area is determined based on a preset concentration deviation threshold.

8. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 7, characterized in that, Also includes: The dynamic fine-tuning module is used for the dynamic fine-tuning steps of the process for small-scale sample test results. After small-scale stainless steel samples were prepared according to the preliminary process plan, the actual composition distribution data, microhardness data and silver phase morphology image data of the samples were collected. The actual component distribution data is compared with the predicted component distribution of the uniform distribution control target in the corresponding region to generate a local component distribution correction field. Analyze the silver phase morphology image data, statistically analyze the size distribution, shape factor and dispersion of the silver phase in the matrix, and identify the areas where the actual silver phase precipitation behavior deviates from the theoretical prediction; By combining the information of the local component distribution correction field and the actual silver precipitation behavior deviation region, the expected silver distribution of the remaining process control units that have not yet been executed or can be adjusted is re-evaluated; Based on the results of the reassessment, the process parameters or intervention priorities of the remaining process control units are adjusted to generate a dynamically updated performance regulation process scheme.

9. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 8, characterized in that, The actual component distribution data is compared with the predicted component distribution of the uniform distribution control target in the corresponding region to generate a local component distribution correction field, including: For each sample detection area that provides the actual component distribution data, the average concentration of silver element within a specific detection surface is extracted. From the corresponding region of the uniformly distributed control target, read the predicted value of the average silver concentration in the sample detection area within the same specific detection surface; The ratio of the measured average concentration of silver to the predicted average concentration of silver is calculated and used as the component prediction correction coefficient for the sample detection area. Using spatial interpolation technology, the component prediction correction coefficient of each sample detection area is extended to the simulated three-dimensional space range of the entire tool material, forming a continuous local component distribution correction field; The local component distribution correction field is multiplied point by point with the original uniform distribution control target to obtain the updated distribution map of the performance regulation target of silver-containing martensitic antibacterial stainless steel.

10. The performance regulation system for silver-containing martensitic antibacterial stainless steel for kitchen knives according to claim 9, characterized in that, The steps for constructing the preset melt flow resistance model are as follows: The historical smelting process data of various silver-containing stainless steels with different compositions under preset smelting parameters were collected. The monitoring data included historical melt surface tension distribution, historical temperature field distribution, historical silver concentration field distribution, and historical ingot final silver element distribution evaluation. For each different composition, a correspondence between the alloy composition and its standard melt viscosity coefficient and surface tension temperature coefficient at multiple preset temperatures is established, forming an alloy composition-physical property parameter mapping table; For each historical smelting process, a dynamic spatiotemporal field of melt viscosity coefficient is generated based on the historical temperature field distribution and the alloy composition-physical property parameter mapping table. Based on the historical melt surface tension distribution and the historical silver concentration field distribution, the theoretical value of the convection velocity field driven by the surface tension gradient and concentration gradient inside the melt is calculated. The final silver element distribution evaluation of the historical ingots is quantified as an index of the uniformity of silver element distribution; Using the dynamic spatiotemporal field of the melt viscosity coefficient, the theoretical value of the convection velocity field, and the parameters in the melt flow resistance model as model inputs, and the uniformity index of silver element distribution as the optimization objective, the model parameters are fitted to obtain a preset melt flow resistance model that predicts the convection mass transfer resistance value.

Images