Laser-hardened surface strengthening method and system for mold steel

Abstract

This invention discloses a method and system for surface strengthening treatment of mold steel based on laser hardening, belonging to the field of mold steel strengthening technology. The method includes: constructing a multi-dimensional state-space model characterizing the evolution behavior of the surface microstructure of the mold steel; determining the evolution trajectory of the target microstructure in the multi-dimensional state-space model and calculating the stability margin distribution of the target microstructure evolution trajectory relative to the critical transition surface of the microstructure evolution; performing joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of the laser action during the laser strengthening process; and performing execution control management of laser strengthening based on the joint planning and optimization results. This invention solves the technical problem of insufficient laser hardening control precision in existing technologies, leading to poor uniformity of the hardened layer on the mold steel surface and low wear resistance and thermal fatigue resistance. It achieves precise control of laser hardening, improving the uniformity, wear resistance, and thermal fatigue resistance of the hardened layer on the mold steel surface.

Description

Technical Field

[0001] This invention relates to the field of mold steel strengthening technology, specifically to a method and system for surface strengthening treatment of mold steel based on laser hardening. Background Technology

[0002] In modern manufacturing, mold steel, as a core material for critical forming components, directly determines the service life of the mold and the quality of the formed products due to its surface wear resistance and thermal fatigue resistance. Laser hardening technology, with its advantages of rapid heating, high cooling efficiency, and minimal workpiece deformation, has become one of the mainstream methods for surface strengthening of mold steel. However, existing laser hardening methods largely rely on empirically preset laser parameters, failing to fully consider the impact of initial microstructure differences, surface heterogeneity, and non-uniform thermal diffusion on microstructure evolution. This makes it difficult to accurately control the phase transformation process during laser treatment. Furthermore, traditional techniques lack systematic division and dynamic monitoring of stable and phase-transformation sensitive regions, easily leading to problems such as localized thermal drift, uneven energy absorption, and sluggish response in certain areas. This results in uneven hardened layer thickness and large fluctuations in hardness distribution, ultimately affecting the improvement of the mold steel's wear resistance and thermal fatigue resistance, failing to meet the stringent requirements of high-precision and long-life molds in high-end manufacturing.

[0003] The existing technology has insufficient precision in controlling laser hardening, resulting in poor uniformity of the hardened layer on the surface of mold steel and low wear resistance and thermal fatigue resistance. Summary of the Invention

[0004] This application provides a method and system for surface strengthening treatment of mold steel based on laser hardening, which is used to address the technical problems in the prior art where insufficient control precision of laser hardening leads to poor uniformity of the hardened layer on the surface of mold steel and low wear resistance and thermal fatigue resistance.

[0005] In view of the above problems, this application provides a method and system for surface strengthening treatment of mold steel based on laser hardening.

[0006] The first aspect of this application provides a method for surface strengthening treatment of mold steel based on laser hardening, the method comprising:

[0007] Before laser strengthening, a multidimensional state-space model characterizing the surface microstructure evolution behavior of the mold steel is constructed based on its initial microstructure, surface aggregate structure characteristics, and material thermal diffusion inhomogeneity. This multidimensional state-space model includes a stable phase region, a phase transition sensitive region, and corresponding critical transition surfaces for microstructure evolution. Based on the strengthening performance requirements corresponding to the target service conditions, the target microstructure evolution trajectory is determined in the multidimensional state-space model, and the stability margin distribution of the target microstructure evolution trajectory relative to the critical transition surfaces is calculated. During laser strengthening, the evolution pose and stability margin change trend of the current microstructure evolution state in the multidimensional state space are used as key control variables, and the stability margin distribution is used as the microstructure evolution risk weight field. Joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of the laser action are performed. The execution control management of laser strengthening is then implemented based on the joint planning and optimization results.

[0008] A second aspect of this application provides a laser-hardened surface strengthening treatment system for mold steel, the system comprising:

[0009] The multidimensional state-space model construction module is used to construct a multidimensional state-space model characterizing the surface microstructure evolution behavior of the mold steel based on its initial microstructure state, surface aggregate structure characteristics, and material thermal diffusion non-uniformity before laser strengthening treatment. This multidimensional state-space model includes a stable microstructure phase region, a phase transition sensitive region, and the corresponding critical transition surface for microstructure evolution. The stability margin distribution calculation module is used to determine the target microstructure evolution trajectory in the multidimensional state-space model according to the strengthening performance requirements corresponding to the target service conditions, and to calculate the stability margin distribution of the target microstructure evolution trajectory relative to the critical transition surface. The joint planning and optimization module is used during laser strengthening treatment to perform joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of the laser action, using the current microstructure evolution pose and stability margin change trend in the multidimensional state space as key control variables, and the stability margin distribution as the microstructure evolution risk weight field. The control and management module is used to perform execution control management of laser strengthening based on the joint planning and optimization results.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] Before laser hardening, a multidimensional state-space model characterizing the microstructure evolution behavior of the mold steel surface is constructed. Based on the strengthening performance requirements corresponding to the target service conditions, the target microstructure evolution trajectory is determined in the multidimensional state-space model, and the stability margin distribution of the target microstructure evolution trajectory relative to the critical transition surface of the microstructure evolution is calculated. During laser hardening, the evolution pose and stability margin change trend of the current microstructure evolution state in the multidimensional state space are used as key control variables, and the stability margin distribution is used as the microstructure evolution risk weight field. Joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of the laser action are performed. Based on the joint planning and optimization results, the execution control management of laser hardening is implemented. This achieves precise control of laser hardening, improving the uniformity, wear resistance, and thermal fatigue resistance of the hardened layer on the mold steel surface. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic diagram of the process for surface strengthening treatment of mold steel based on laser hardening provided in the embodiments of this application;

[0014] Figure 2 A schematic diagram of a laser-hardened mold steel surface strengthening treatment system provided in this application embodiment.

[0015] Figure labeling: Multidimensional state space model construction module 10, stability margin distribution calculation module 20, joint planning optimization module 30, control management module 40. Detailed Implementation

[0016] This application provides a method and system for surface strengthening treatment of mold steel based on laser hardening, which addresses the technical problems in the prior art where insufficient control precision of laser hardening leads to poor uniformity of the hardened layer on the surface of mold steel and low wear resistance and thermal fatigue resistance.

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0018] Example 1, as Figure 1As shown, this application provides a method for surface strengthening treatment of mold steel based on laser hardening, the method comprising:

[0019] Step S100: Before performing laser strengthening treatment, based on the initial microstructure state, surface aggregate structure characteristics and material thermal diffusion non-uniformity of the mold steel, a multidimensional state space model characterizing the surface microstructure evolution behavior of the mold steel is constructed. The multidimensional state space model includes a microstructure phase stability region, a phase transformation sensitive region and the corresponding microstructure evolution critical transition surface.

[0020] Specifically, before performing laser strengthening, initial microstructure data of the mold steel is collected to clarify the initial phase composition ratio, dislocation density, grain morphology factor, and carbide distribution density, which serve as the structural basis for the microstructure evolution space. Then, the surface aggregate structure characteristics of the mold steel are extracted, encompassing grain orientation distribution, surface structure interval boundaries, and concentrated areas of surface micro-defects. Surface orientation partitions are constructed through regionalization and merging. Subsequently, the thermal diffusion non-uniformity of the material is spatially mapped, and a thermal diffusion response matrix is ​​generated by combining the gradient change of the thermal diffusion coefficient with location, the perturbation bandwidth of the surface thermal conductivity, and the thermal trapping characteristics. Based on the above... Based on the structural substrate, surface orientation partitioning, and thermal diffusion response matrix, a multidimensional microstructure evolution state vector is established, including phase sensitivity factor, thermally induced nucleation potential factor, thermal flux response coefficient, and microstructure evolution gain coefficient. Finally, based on this vector, a stable microstructure phase region is constructed by stable clustering of the low-sensitivity interval of the phase sensitivity factor. The phase transition sensitive region is determined by the combination condition of the coupling abrupt change of the thermally induced nucleation potential factor and the thermal flux response coefficient. A continuous transition boundary is formed by the microstructure evolution gain coefficient crossing the critical response threshold, thereby constructing a complete multidimensional state space model to achieve accurate characterization of the microstructure evolution behavior of mold steel surface.

[0021] Step S200: Based on the enhanced performance requirements corresponding to the target service condition, determine the target tissue evolution trajectory in the multidimensional state space model, and calculate the stability margin distribution of the target tissue evolution trajectory relative to the critical transition surface of tissue evolution.

[0022] Specifically, the core strengthening performance requirements of mold steel corresponding to the target service conditions are first clarified, such as wear resistance indicators under specific loads and thermal fatigue resistance standards under high-temperature environments. These requirements are then transformed into quantifiable microstructure performance parameters, such as the target martensite layer hardness range, hardened layer thickness uniformity threshold, and grain refinement requirements. Subsequently, based on the constructed multidimensional state-space model, within its included microstructure phase stability region (constructed by stable clustering based on the low-sensitivity interval of the phase sensitivity factor), phase transformation sensitive region (determined by the coupling abrupt change of the thermally induced nucleation potential factor and the thermal flux response coefficient), and the critical transition surface of microstructure evolution, within the spatial framework formed by the microstructure evolution gain coefficient crossing the critical response threshold, the steel is screened to select those that accurately match the aforementioned microstructure performance parameters. The evolution path must avoid the unstable range of the phase transition sensitive zone and fit within a reasonable range of the stable phase state of the microstructure, ultimately determining the target microstructure evolution trajectory. Finally, by quantitatively calculating key parameters such as the spatial distance between each node on the target microstructure evolution trajectory and the critical transition surface of microstructure evolution, the deviation of the phase state sensitivity factor, and the difference in the thermal flux response coefficient, the stability margin value corresponding to each node is obtained. Then, combined with the global distribution characteristics of the trajectory and the regional characteristics of the mold steel surface, a stability margin distribution covering the entire processing area is formed. This distribution intuitively reflects the safety degree of the target trajectory in moving away from the critical phase transition state during microstructure evolution, providing a precise risk quantification basis for the joint planning and optimization of the temporal energy distribution, spatial action trajectory, and scanning rhythm of subsequent laser action.

[0023] Step S300: During the laser enhancement process, the evolutionary pose and stability margin change trend of the current tissue evolution state in the multidimensional state space are used as key control variables, and the stability margin distribution is used as the tissue evolution risk weight field. The joint planning and optimization of the temporal energy distribution, spatial trajectory and scanning rhythm of the laser action are performed.

[0024] Specifically, in the laser strengthening treatment implementation stage, the evolution pose of the current microstructure evolution state of the mold steel surface in the multi-dimensional state space, namely the state vector parameters such as the associated phase sensitivity factor and thermal flux response coefficient, and the stability margin change trend, are taken as the core key control variables. Simultaneously, the obtained stability margin distribution is directly used as the microstructure evolution risk weight field. Based on this, the joint planning and optimization of laser action parameters is initiated. First, the mold steel surface is discretized according to a predetermined spatial scale, and the stability margin value of each discrete unit is encoded to form a risk weight matrix. Then, directional information reflecting the spatial change trend of the stability margin is extracted from this matrix. On the one hand, this information is used to construct a risk guidance field to constrain the scanning path planning; on the other hand, it is used to identify stability margin changes and construct key monitoring units with phase transition tendencies. The stability margin change patterns of adjacent regions of the key monitoring units are extracted from the risk weight matrix and converted into a local risk adjustment parameter set. Subsequently, based on the risk weight matrix, risk guidance field, and local risk... A structured set of organizational evolution risk information is constructed using a risk adjustment parameter set. Temporal energy distribution, spatial action trajectory, and scanning rhythm are set as three-dimensional joint decision variables. A composite performance index is constructed with the cumulative amount of organizational evolution risk weighted by the risk weight matrix, the trajectory bias requirement determined by the risk guidance field, the local energy input limit constrained by the local risk adjustment parameter set, and the enhanced organizational deviation corresponding to the target organizational evolution trajectory as the core. This index is used as the optimization objective to perform a phased solution. First, a preset scanning path framework that conforms to the risk guidance field constraint is established through a global path generator. Within the framework, the temporal energy distribution and scanning rhythm are coupled and solved. Then, based on the solution results, the turning angle, coverage width, and overlap area of ​​the local segments of the path are dynamically reconstructed to reduce the local risk accumulation to a preset threshold. After that, a global consistency correction is performed on the dynamically reconstructed path framework to uniformly calibrate the reference frequency of the scanning rhythm and the reference flux of the energy input. Finally, the temporal energy distribution, spatial action trajectory, and scanning rhythm are jointly summarized to complete the joint planning optimization.

[0025] Step S400: Perform laser enhancement execution control management based on the joint planning optimization results.

[0026] Specifically, firstly, based on the joint planning and optimization results of the obtained laser action time energy distribution, spatial action trajectory, and scanning rhythm, a high-power laser is driven to perform laser strengthening scanning on the surface of the mold steel according to preset parameters, achieving localized precise heating to promote the formation of a high-hardness martensite layer. During the actual processing, the dynamic working conditions such as local thermal drift trend, surface energy absorption changes, and regional response hysteresis on the mold steel surface are identified in real time, and relevant data are collected to construct a monitoring dataset. Based on this monitoring dataset, the determined stability margin distribution is dynamically scaled and adjusted, and local strengthening treatment is carried out for abnormal areas. The joint planning and optimization results are updated synchronously to adapt to real-time working condition changes, ensuring optimized heating and cooling processes and uniform formation of the hardened layer. At the same time, in the entire execution control and management process, the joint planning and optimization results, the initial microstructure of the mold steel, surface aggregate structure characteristics, material thermal diffusion non-uniformity, and strengthening treatment detection results are integrated to construct a mapped storage data group, and encrypted storage management is performed to provide data support for subsequent strengthening treatments of similar mold steels, ultimately ensuring the precise improvement of the wear resistance and thermal fatigue resistance of the mold steel.

[0027] In one possible implementation, step S300 further includes:

[0028] Step S310: Discretize the surface of the mold steel according to a predetermined spatial scale, encode the stability margin value of the discrete unit, and form a risk weight matrix.

[0029] Step S320: Based on the risk weight matrix, extract directional information reflecting the trend of change in the stability margin space, and construct a risk guidance field based on the directional information. The risk guidance field is used to constrain the scanning path planning.

[0030] Step S330: Use the directional information to identify changes in stability margin and construct a key monitoring unit for phase transition tendency.

[0031] Step S340: For the key monitoring unit, extract the stability margin change pattern of adjacent regions from the risk weight matrix and convert it into a local risk adjustment parameter set.

[0032] Step S350: Based on the risk weight matrix, risk guidance field, and local risk adjustment parameter set, construct a structured tissue evolution risk information set, and perform joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of laser action according to the tissue evolution risk information set.

[0033] Specifically, firstly, based on the dimensions of the mold steel surface, the required precision of the strengthening process, and the planning needs of the subsequent scanning path, a predetermined spatial scale is set, such as dividing it into uniform grid units of 5mm×5mm or 10mm×10mm. A spatial discretization algorithm is then used to divide the entire mold steel surface into several independent discrete units according to this scale, ensuring that the regional characteristics of each unit can be quantified and analyzed individually. Next, the calculated stability margin distribution data is retrieved, and the stability margin value corresponding to each discrete unit, reflecting the safety of the unit's microstructure evolution away from the critical transition surface, is standardized and encoded through data matching. The stability margin value is then mapped to a risk weight coefficient in the 0-1 range according to preset rules. For example, a higher stability margin corresponds to a lower risk weight coefficient, indicating a smaller risk of microstructure evolution; conversely, a lower stability margin corresponds to a higher risk weight coefficient, indicating a greater risk of approaching the critical phase transition state. Finally, the risk weight coefficients of all discrete units are arranged sequentially according to their spatial position on the mold steel surface, constructing a two-dimensional risk weight matrix. Each element in the matrix corresponds one-to-one with a discrete unit, intuitively presenting the microstructure evolution risk distribution of each region on the mold steel surface, providing structured data support for subsequent risk guidance field construction and key monitoring unit identification.

[0034] The spatial gradient algorithm is used to perform a global analysis of the risk weight matrix. By calculating the difference in risk weight coefficients and direction vectors between adjacent discrete units, directional information reflecting the spatial change trend of the stability margin is extracted. This includes the orientation, rate of change, and gradient magnitude of the risk weight increase / decrease. For example, if the risk weight of a unit increases towards the adjacent unit to the right and the gradient magnitude is greater than a preset threshold, the directional information indicates an increase in risk to the right at a rate of 0.3 per unit. Subsequently, a risk guidance field is constructed based on this directional information. First, field constraint rules are defined, such as the path must extend along the direction of decreasing risk weight gradient and avoid abrupt changes where the gradient magnitude exceeds the threshold. Then, the directional information is mapped to a vector field distribution within the field. Each discrete unit corresponds to a guidance vector, with the vector direction pointing to the optimal orientation of risk reduction. The vector magnitude is positively correlated with the rate of risk change. Finally, a continuous path constraint boundary is formed through this vector field, enabling subsequent scanning path planning to adaptively adjust along the direction of the guidance vector, prioritizing the construction of paths in low-risk, stable-change areas, and avoiding high-risk phase transition sensitive areas, thus providing precise directional constraints for the laser scanning path.

[0035] A preset stability margin change threshold is established, including a change rate threshold and a critical difference threshold. For example, a stability margin change exceeding 0.3 per unit distance is the rate threshold, and a stability margin value less than 0.2 from the critical transition surface of tissue evolution is the critical difference threshold. Based on the extracted directional information, a feature recognition algorithm is used to traverse the entire discrete unit domain, screening out units whose stability margin change rate exceeds the threshold or whose stability margin value is close to the critical transition surface. Subsequently, through regional connectivity analysis, adjacent high-risk screening units are merged into continuous regions. At the same time, combined with prior information such as the surface orientation partitioning and thermal diffusion response matrix of the mold steel surface, false change regions caused by the material's inherent characteristics are eliminated. Finally, the merged continuous high-risk regions are marked as key monitoring units for phase transition tendency, and each unit is assigned a unique identifier and corresponding monitoring parameters, such as real-time sampling frequency and state feedback threshold, to ensure that these regions prone to phase transition anomalies can be monitored and dynamically controlled during subsequent laser strengthening processes.

[0036] The spatial coordinates of each key monitoring unit in the risk weight matrix are located. A predetermined neighborhood, such as a 3×3 or 5×5 discrete unit matrix, is defined centered on each key monitoring unit. The stability margin values ​​and risk weight coefficients of all units within the neighborhood are extracted by traversing the matrix index. Subsequently, pattern recognition algorithms, such as the K-nearest neighbor algorithm and cluster analysis algorithms, are used to analyze the changing characteristics of the stability margin within the neighborhood, identifying patterns such as gradual trends, abrupt change nodes, and uniform distribution. Examples include a gradual decrease in stability margin from the center to the edge, and a sudden drop in stability margin in a certain direction. Next, these change patterns are matched with preset parameter mapping rules, and characteristic quantities such as gradual change rate, abrupt change amplitude, and uniformity deviation are converted into corresponding local risk adjustment parameters, including quantifiable control parameters such as local energy input upper / lower limits, scan rate adjustment range, path coverage overlap rate threshold, and temperature monitoring response sensitivity. Finally, the various local risk adjustment parameters corresponding to each key monitoring unit are integrated according to a preset format to form a structured local risk adjustment parameter group, which provides a precise parameter constraint basis for subsequent local optimization of laser action time energy distribution, scanning rhythm, and path dynamic reconstruction.

[0037] By integrating structured data, the risk weight matrix, risk guidance field, and local risk adjustment parameter set are correlated and mapped to clarify their spatial correspondence and parameter linkage rules. This constructs a structured organizational evolution risk information set containing a risk quantification layer, a constraint guidance layer, and a local control layer, comprehensively covering the global risk distribution, path constraint direction, and key area local control requirements. Then, the temporal energy distribution, spatial trajectory, and scanning rhythm of laser action are set as three-dimensional joint decision variables. Based on the organizational evolution risk information set, a composite performance index is constructed. This index includes the cumulative organizational evolution risk calculated weighted by the risk weight matrix, the trajectory bias requirement determined by the risk guidance field, the local energy input limit constrained by the local risk adjustment parameter set, and the enhanced organizational deviation calculated based on the target organizational evolution trajectory. Subsequently, the composite performance index is used as the optimization objective, and a phased solution of the three-dimensional joint decision variables is performed, first utilizing the global path... The path generator establishes a preset scanning path framework that conforms to the directional constraints of the risk guidance field. Within this framework, the temporal energy distribution and scanning rhythm are coupled and solved through local optimization. Based on the coupled solution results, the local segments of the preset scanning path framework are dynamically reconstructed. According to the local risk adjustment parameter set, the turning angle, coverage width, and segment overlap area between path segments are fine-tuned in real time to reduce the cumulative risk of local segments of the path to a preset threshold range. Then, global consistency correction is performed on the dynamically reconstructed preset scanning path framework. The path drift, rhythm difference, and energy input deviation introduced by the local fine-tuning are used to construct a global offset. This global offset is used to uniformly calibrate the scanning rhythm reference frequency and energy input reference flux globally. Finally, after completing the global consistency correction, the temporal energy distribution, spatial action trajectory, and scanning rhythm are jointly summarized to form a joint planning optimization result that meets the requirements of controllable tissue evolution risk and enhanced performance.

[0038] In one possible implementation, step S350 further includes:

[0039] Step S351: Use the time energy distribution, spatial action trajectory, and scanning rhythm as three-dimensional joint decision variables.

[0040] Step S352: Construct a composite performance index for joint planning optimization based on the organizational evolution risk information set. The composite performance index includes the cumulative amount of organizational evolution risk calculated according to the risk weight matrix, the trajectory bias requirement determined based on the risk guidance field, the local energy input limit constrained by the local risk adjustment parameter group, and the enhanced organizational deviation calculated based on the target organizational evolution trajectory.

[0041] Step S353: Using the composite performance index as the optimization objective, perform a phased solution of the three-dimensional joint decision variables to complete the joint planning optimization.

[0042] Specifically, the three key controllable parameters in laser enhancement processing are defined as three-dimensional joint decision variables. Temporal energy distribution corresponds to energy allocation parameters in the time dimension, such as the energy input flux and duration of energy action per unit time. Spatial trajectory corresponds to motion parameters in the spatial dimension, such as the coordinates of path nodes, path segment connection methods, and coverage area of ​​laser scanning. Scanning rhythm corresponds to execution parameters in the rhythm dimension, such as laser scanning rate, path switching frequency, and interval between adjacent paths. These three parameters together constitute an interconnected and synergistically regulated three-dimensional decision variable system. This provides a clear optimization target for subsequent joint planning and optimization based on the tissue evolution risk information set, ensuring that the key parameters of laser action can achieve controllable tissue evolution risk and meet enhancement performance standards through systematic optimization.

[0043] First, relying on a structured set of organizational evolution risk information, a composite performance index is constructed by systematically integrating four core constraint dimensions. For the cumulative amount of organizational evolution risk, a weighted summation method is used, based on the risk weight coefficients of each discrete unit in the risk weight matrix, combined with the processing priority weights of the corresponding regions to calculate the overall comprehensive risk value. For trajectory bias requirements, by analyzing the vector distribution rules of the risk guidance field, the deviation angle and distance between the preset scanning path and the optimal travel direction of the guidance field are calculated, transforming them into quantifiable trajectory constraint parameters. Regarding local energy input limitations, the energy control standards in the local risk adjustment parameter group are directly mapped to the upper and lower limits of energy input for each key monitoring unit and adjacent regions, forming precise local energy constraints. For strengthening organizational deviation, the core parameters of the target organizational evolution trajectory and the corresponding indicators of the actual organizational evolution state are extracted. By calculating the absolute difference and relative deviation rate between the two, the degree of fit between organizational performance and target requirements is quantified. Finally, the four dimensions of indicators are weighted and integrated according to preset weight ratios to form a unified composite performance index, providing a comprehensive and quantifiable target basis for the optimization solution of three-dimensional joint decision variables.

[0044] A global path generator is invoked to generate a preset scanning path framework covering the surface treatment area of ​​the mold steel, based on the directional constraint rules of the risk guidance field, ensuring that the overall path meets the guidance requirements for risk reduction. Then, within this preset path framework, a local optimization algorithm is used to couple the temporal energy distribution and scanning rhythm, establishing a coupled solution result that coordinates the two, achieving local adaptation between energy input and scanning rhythm. Subsequently, based on the coupled solution result and combined with local risk adjustment parameter sets, local segments of the preset scanning path framework are dynamically reconstructed. By fine-tuning the turning angles, coverage widths, and overlapping areas between path segments in real time, the cumulative risk of local segments is controlled within a preset threshold range. Next, global consistency correction is performed on the dynamically reconstructed path framework. By calculating the path drift, rhythm differences, and energy input deviations caused by local fine-tuning, a global offset is constructed. This offset is used to uniformly calibrate the reference frequency of the scanning rhythm and the reference flux of energy input across the entire domain, eliminating the impact of local adjustments on global consistency. Finally, after completing the global calibration, the optimized temporal energy distribution, spatial action trajectory, and scanning rhythm are jointly summarized to form a joint planning optimization result that satisfies the optimal composite performance index.

[0045] In one possible implementation, step S353 further includes:

[0046] Step S3531: Use the global path generator to establish a preset scan path framework that conforms to the directional constraints of the risk guidance field.

[0047] Step S3532: Within the preset scanning path framework, use local optimization to perform coupled solution of time energy distribution and scanning rhythm, and establish coupled solution results.

[0048] Step S3533: Based on the coupling solution results, dynamically reconstruct the local segments of the preset scanning path framework. The dynamic reconstruction includes real-time fine-tuning of the turning angle, coverage width and segment overlap area between path segments according to the local risk adjustment parameter group, so that the cumulative risk of the local segments of the path decreases to a preset threshold range.

[0049] Step S3534: Perform global consistency correction on the dynamically reconstructed preset scan path framework, construct a global offset based on the path drift, rhythm difference and energy input deviation introduced by local fine-tuning, and use the global offset to uniformly calibrate the scan rhythm reference frequency and energy input reference flux in the global range.

[0050] Step S3535: After completing the global consistency correction, the temporal energy distribution, spatial action trajectory, and scanning rhythm are jointly summarized to form the joint planning optimization result.

[0051] Specifically, the existing classic A-path search algorithm is adopted as the core algorithm of the global path generator. The discrete unit grid after discretization of the mold steel surface is used as the environment map for path search. The directional constraint of the risk guidance field is transformed into the heuristic function weight factor of the A algorithm. The vector direction cost term of the risk guidance field is introduced into the heuristic function, so that the movement path consistent with the optimal direction of the risk guidance field is selected first during the path search process. At the same time, the passage cost of high-risk units in the risk weight matrix is ​​set to the maximum value, forming the forbidden region of path search. In the algorithm initialization phase, the coordinates of the starting and ending points of the mold steel surface strengthening treatment are set, and a search graph is constructed using discrete units as nodes. During the search process, the cost function f(n) = g(n) + h(n) of each node is calculated, where g(n) is the actual travel cost from the starting point to the current node, and h(n) is the heuristic cost including the risk guidance direction weight. The node with the smallest cost function value is expanded first. At the same time, combined with the vector constraint of the risk guidance field, the node expansion direction is restricted to the effective direction of risk reduction, avoiding extension to high-risk areas. Finally, through step-by-step iterative search, a continuous path covering the entire processing area, conforming to the directional constraint of the risk guidance field, and avoiding high-risk taboo areas is generated, which serves as the preset scanning path framework.

[0052] Within the established pre-defined scanning path framework, and based on the risk weight matrix and local risk adjustment parameter set in the organizational evolution risk information set, the particle swarm optimization algorithm is used as the core algorithm for local optimization. The temporal energy distribution, including energy input flux per unit time and energy duration, and the scanning rhythm, including scanning rate and path switching frequency, are used as coupled optimization variables to construct a collaborative constraint model. The energy input needs to be adapted to the scanning rate to avoid local overheating or energy deficiency due to energy accumulation, and the scanning rhythm needs to be dynamically adjusted to match the risk level of different regions. In the algorithm initialization phase, reasonable ranges for energy input and scan rate are set, and the optimization boundary is defined by combining the upper and lower energy thresholds in the local risk adjustment parameter set. During the iteration process, the cumulative amount of tissue evolution risk and local energy input limit in the composite performance index are used as adaptation targets. By calculating the fitness value of each particle, that is, the risk control effect and energy fitness corresponding to the coupling parameter combination, the particle position and velocity are dynamically updated, gradually converging to the optimal coupling parameter combination. At the same time, a constraint processing mechanism is introduced to penalize and correct coupling solutions that exceed the local risk adjustment parameter limit, ensuring that the solution results meet the local control requirements of the key monitoring area. Finally, through multiple rounds of iteration convergence, a coupling solution that takes into account both accurate energy allocation and scan rhythm adaptation is established, providing parameter support for subsequent local dynamic reconstruction of the path.

[0053] For local segments exceeding the preset threshold range, a gradient descent algorithm is introduced to iteratively optimize the turning angles between path segments, constrained by the temporal energy distribution and scanning rhythm parameters in the coupled solution results. By calculating the gradient relationship between the turning angle adjustment and the cumulative risk, the angle is gradually corrected to avoid areas of abrupt risk gradient changes. Simultaneously, an adaptive dynamic programming algorithm is used to optimize the path coverage width, using the energy control standard in the local risk adjustment parameter group as the reward function to dynamically adjust the coverage width to adapt to different risk levels in different areas. For overlapping segments, a K-means clustering algorithm is used to perform cluster analysis on the risk distribution characteristics of adjacent path segments to determine the optimal overlap range and density. Throughout the process, a Bayesian filtering algorithm is used to update the calculated results of the cumulative risk of local segments in real time, dynamically comparing them with the preset threshold range, and iteratively adjusting the turning angle, coverage width, and overlapping parameters until the cumulative risk meets the threshold requirements, thus completing the dynamic reconstruction of local segments of the preset scanning path framework.

[0054] An iterative nearest-point algorithm is used to register all path nodes in the dynamically reconstructed preset scanning path framework. The spatial coordinates of the actual path nodes are compared with those of the theoretical path nodes point by point, and the deviation values ​​in the x / y / z axes are calculated. Noise is reduced by sliding window filtering to accurately quantify path drift. At the same time, a dynamic time warping algorithm is used to align the actual scanning rhythm of each path segment with the reference rhythm sequence. Rhythm difference features are extracted through variance analysis. The Kalman filter algorithm is combined with real-time data from the laser power sensor and theoretical energy input values ​​to calculate the absolute deviation and relative deviation rate of energy input. Then, a weighted fusion algorithm is used to integrate path drift, rhythm difference, and energy input deviation according to preset weights to construct a global offset matrix containing three dimensions: space, time, and energy. Based on this matrix, a PID control algorithm is used to build a global calibration model. The calibration correction amount of the reference frequency of the scanning rhythm and the reference flux of the energy input is calculated according to the offset distribution characteristics. Finally, the reference frequency of the scanning rhythm and the reference flux of the energy input are synchronously adjusted in the entire domain according to the correction amount to ensure that the parameters of each path segment meet the preset accuracy after calibration, and completely eliminate the influence of local fine-tuning on the synergy of global laser action parameters.

[0055] After completing the global consistency correction, the optimized parameters are systematically integrated and verified. The time-energy distribution parameters after global calibration are extracted, including energy input flux per unit time, energy duration, and spatial trajectory data after dynamic reconstruction and calibration, including final path node coordinates, path segment connection methods, coverage area range, and local segment adjustment parameters. The scan rhythm parameters after unified calibration are also included, including scan rate, path switching frequency, and reference frequency, ensuring that all three types of parameters meet the optimal requirements of composite performance indicators and preset accuracy standards. Subsequently, a structured data integration method is used to map and match the three types of parameters according to the three-dimensional correlation logic of time-space-rhythm, clarifying the energy input standards and rhythm execution rules corresponding to different path segments, and forming a parameter linkage reference table. Finally, the integrity and synergy of the integrated parameter set are verified, redundant data is eliminated, parameter conflicts are corrected, and finally, a standardized joint planning optimization result that can be directly used for laser strengthening execution control is formed, providing comprehensive, coordinated, and accurate parameter support for the subsequent precise implementation of mold steel surface strengthening treatment.

[0056] In one possible implementation, step S100 further includes:

[0057] Step S110: The initial microstructure includes the initial phase composition ratio, dislocation density, grain morphology factor and carbide distribution density, and the initial microstructure is used as the structural basis of the microstructure evolution space.

[0058] Step S120: The surface aggregate structure features include grain orientation distribution, surface structure interval boundaries and surface micro-defect concentration areas. The surface aggregate structure is adjusted by regional merging to construct surface orientation partitions.

[0059] Step S130: Spatial mapping of the thermal diffusion non-uniformity of the material is performed, and a thermal diffusion response matrix is ​​formed based on the gradient change of the thermal diffusion coefficient with position, the perturbation bandwidth of the surface thermal conductivity, and the thermal trapping characteristics.

[0060] Step S140: Construct a multidimensional state-space model characterizing the surface microstructure evolution behavior of mold steel based on the structural substrate, surface orientation partitioning, and thermal diffusion response matrix.

[0061] Specifically, before constructing a multidimensional state-space model of the microstructure evolution of mold steel surface, metallographic analysis, electron microscopy, and X-ray diffraction are used to comprehensively and quantitatively characterize the microstructure features of the mold steel in its original state. This allows for the precise acquisition of the initial phase composition ratios, such as the proportions of ferrite, pearlite, and cementite, the dislocation density (the number of dislocations per unit volume), the internal stress state of the material, grain morphology factors (including the aspect ratio and shape factor of grains), the grain geometric morphology, and the carbide distribution density (the number and uniformity of carbide particles per unit area or volume). These parameters are then standardized to eliminate detection errors and dimensional differences, forming a dataset that comprehensively reflects the essential properties of the initial microstructure of the mold steel. This dataset serves as the structural basis for the microstructure evolution space, providing solid microstructural data support for the subsequent integration of surface aggregate structure features and thermal diffusion characteristics to construct a complete multidimensional state-space model. This ensures that the model accurately reflects the initial state of the mold steel and represents its microstructure evolution behavior.

[0062] In constructing the multidimensional state-space model, the aggregate structure characteristics of the mold steel surface are first accurately detected and data collected using electron backscatter diffraction (EBSD) and scanning electron microscopy. The focus is on acquiring three core data types: grain orientation distribution (i.e., crystallographic orientation information of different grains), surface structure boundaries (such as the transition boundary between the surface and the substrate, the delineation of different microstructure regions), and concentrated areas of surface micro-defects (such as the location and extent of microcracks, pores, and inclusions). Subsequently, the collected discretized surface aggregate structure data is preprocessed to remove detection noise and anomalies. Using constant data points, a region growing algorithm combined with clustering analysis logic is employed to regionalize and merge dispersed surface structural features according to the principles of grain orientation consistency, structural interval continuity, and micro-defect distribution correlation. Regions with similar orientation features, located in the same structural interval, and with similar micro-defect distribution patterns are integrated into unified units. Ultimately, a surface orientation partition with clear boundaries and well-defined partition characteristics is constructed. This partition can systematically reflect the spatial distribution pattern of the mold steel surface structure, providing accurate surface structural support for the subsequent integration of the initial microstructure and thermal diffusion characteristics to construct a complete multi-dimensional state space model.

[0063] In the multidimensional state-space model construction stage, considering the non-uniformity of thermal diffusion in mold steel, infrared thermal imaging and laser scintillation methods were used to accurately collect and quantify thermal diffusion-related parameters across the entire material. The focus was on acquiring data on the gradient variation of the thermal diffusivity with spatial location, the perturbation bandwidth of surface thermal conductivity, and the material's inherent heat trapping characteristics. These discrete thermal diffusivity parameters were precisely matched and meshed according to the spatial coordinates of the mold steel surface using a spatial mapping algorithm, establishing a one-to-one correspondence between parameters and spatial locations. Based on this, key indicators such as the gradient variation of the thermal diffusivity, the perturbation amplitude of thermal conductivity, and heat trapping efficiency were integrated through matrix encoding to form a thermal diffusion response matrix that comprehensively quantifies the thermal diffusion response patterns in different regions of the mold steel. This provides accurate thermophysical data support for the subsequent construction of a complete multidimensional state-space model integrating the structural substrate and surface orientation zones, ensuring that the model can realistically characterize the impact of thermal diffusion non-uniformity on microstructure evolution.

[0064] After constructing the structural substrate, surface orientation partitioning, and thermal diffusion response matrix, a multidimensional microstructure evolution state vector was established using a multidimensional data fusion algorithm. This vector included phase sensitivity factors, thermally induced nucleation potential factors, thermal flux response coefficients, and microstructure evolution gain coefficients. This comprehensively integrated the essential properties of the initial microstructure, the spatial distribution of the surface structure, and the non-uniformity of thermal diffusion into a unified state characterization system. Subsequently, based on the stable clustering results of the phase sensitivity factors in the low-sensitivity range, the microstructure phase stability region was delineated. The phase transition sensitive region was defined by combining the coupling abrupt combination conditions of the thermally induced nucleation potential factor and the thermal flux response coefficient. The critical transition surface of microstructure evolution was determined according to the continuous transition boundary formed by the microstructure evolution gain coefficient crossing the critical response threshold. Finally, by spatially associating and logically integrating the microstructure phase stability region, the phase transition sensitive region, and the critical transition surface of microstructure evolution, a multidimensional state space model capable of comprehensively characterizing the microstructure evolution behavior of the mold steel surface was constructed. This provides accurate model support for the subsequent determination of the target microstructure evolution trajectory and the joint planning and optimization of laser parameters.

[0065] In one possible implementation, step S140 further includes:

[0066] Step S141: Establish a multidimensional tissue evolution state vector based on the structural substrate, surface orientation partition, and thermal diffusion response matrix. The multidimensional tissue evolution state vector includes a phase sensitivity factor, a thermally induced nucleation potential factor, a thermal flow response coefficient, and a micro-region tissue evolution gain coefficient.

[0067] Step S142: Determine the stable phase region, phase transition sensitive region, and critical transition surface of tissue evolution based on the multidimensional tissue evolution state vector to construct a multidimensional state space model.

[0068] Specifically, principal component analysis (PCA) is used to perform dimensionality reduction on multi-source data, including the initial phase composition ratio, dislocation density, grain morphology factor, and carbide distribution density of the structural substrate; the grain orientation distribution, surface structural interval boundaries, and surface micro-defect concentration areas covered by the surface orientation partition; and the gradient change of thermal diffusivity coefficient with position, surface thermal conductivity perturbation bandwidth, and heat trapping characteristics involved in the thermal diffusivity response matrix, to extract core correlation features. A mapping relationship between features and target factors is established using a multiple linear regression algorithm. The phase sensitivity factor is calculated using the initial phase composition ratio and grain orientation consistency as core parameters. The thermally induced nucleation potential factor is derived by combining dislocation density and thermally induced nucleation conditions. The thermal flux response coefficient is quantified based on the gradient change of thermal diffusivity coefficient and heat transfer efficiency. The micro-region microstructure evolution gain coefficient is determined by integrating carbide distribution density, the influence of micro-defect concentration areas, and the microstructure evolution gain law. The four factors are standardized and normalized to eliminate dimensional differences and data redundancy, ultimately forming a multi-dimensional microstructure evolution state vector that can comprehensively characterize the core features of the surface microstructure evolution of mold steel.

[0069] The DBSCAN density clustering algorithm was used to perform cluster analysis on the phase sensitivity factors in the multidimensional microstructure evolution state vector, and the dataset with concentrated density in the low sensitivity interval was screened out. Based on the cluster boundary characteristics and data distribution stability, the microstructure phase stability region was delineated. A coupled correlation model between the thermally induced nucleation potential factor and the thermal flux response coefficient was established using the multivariate logistic regression algorithm. A threshold for the determination of two-factor co-mutation was set, and the regions that meet the mutation conditions were screened out using threshold logic and identified as phase transformation sensitive regions. The support vector machine (SVM) algorithm was used to train and fit the microstructure evolution gain coefficient, and its critical response threshold was calculated. The discrete points that cross the threshold were fitted into a continuous and smooth microstructure evolution critical transition surface using the boundary interpolation algorithm. Finally, the spatial coordinate registration algorithm was used to establish a three-dimensional spatial correlation between the microstructure phase stability region, the phase transformation sensitive region, and the microstructure evolution critical transition surface. Combined with logical rules, a multidimensional state space model that can completely characterize the microstructure evolution behavior of the mold steel surface was formed.

[0070] In one possible implementation, step S140 further includes:

[0071] The stable region of tissue phase is constructed based on the stable clustering of phase sensitivity factors in the low sensitivity range. The phase change sensitive region is determined based on the combination of thermally induced nucleation potential factor and thermal flux response coefficient coupling abrupt change. The critical transition surface of tissue evolution forms a continuous transition boundary according to the micro-region tissue evolution gain coefficient crossing the critical response threshold.

[0072] Specifically, a data preprocessing algorithm is used to screen the phase sensitivity factors in the multidimensional tissue evolution state vector, removing outliers and noise data. Low-sensitivity intervals are then defined using a threshold determination. The DBSCAN density clustering algorithm is then used to stably cluster the data within these intervals. Based on the density distribution characteristics and boundary integrity of the clusters, the spatial range and boundary features of the tissue phase stability region are determined. The specific method for determining the phase transition sensitive region is as follows: a coupled correlation model between the thermally induced nucleation potential factor and the heat flux response coefficient is established using a multiple linear regression algorithm, and the dual-factor model is calibrated using experimental data. The threshold for determining the sub-synergistic mutation is determined by using a logical discrimination algorithm to detect the data point by point across the entire domain. Regions that meet the conditions of thermally induced nucleation potential factor exceeding the threshold and thermal flux response coefficient mutation are designated as phase transition sensitive regions. The specific method for forming the critical transition surface of tissue evolution is as follows: the support vector machine (SVM) algorithm is used to train and fit the historical data of the micro-region tissue evolution gain coefficient to determine its critical response threshold. The discrete points that cross the threshold in the entire domain are fitted using a spatial interpolation algorithm to form a continuous, smooth transition boundary with clear physical meaning, namely the critical transition surface of tissue evolution.

[0073] In one possible implementation, step S400 further includes:

[0074] Step S410: During the actual processing, identify local thermal drift trends, changes in surface energy absorption, and regional response hysteresis, and construct a monitoring dataset.

[0075] Step S420: Use the monitoring dataset to perform dynamic scaling and local reinforcement processing on the stability margin distribution to update the joint planning optimization results.

[0076] Specifically, in the actual execution of laser strengthening processing of mold steel, an infrared thermal imaging sensor is used to collect temperature field data of the processing area in real time. A time-series data analysis algorithm is used to extract the slope of temperature change and the cumulative deviation from the preset temperature threshold, accurately identifying the direction, rate, and range of local thermal drift. A laser power feedback detection device is used to simultaneously collect incident laser power and laser power reflected from the material surface. Combined with an energy absorption calculation model, the surface energy absorption efficiency is calculated. Through fluctuation feature extraction and abrupt change threshold determination, the dynamic change law and abnormal fluctuation nodes of surface energy absorption are captured. A high-speed industrial camera is used to... Images of tissue morphology evolution in each region were captured at a preset frame rate. Image registration, feature point tracking, and time-series comparison algorithms were used to calculate the time difference between different regions reaching the preset tissue evolution stage, clarifying the distribution range and degree of regional response lag. Finally, the quantitative parameters of local thermal drift trends, dynamic data of surface energy absorption changes, and characteristic indicators of regional response lag were standardized and integrated according to spatial coordinates and time series to construct a monitoring dataset that covers the three dimensions of heat, energy, and response, is time-continuous, and structurally unified. This provides real-time data support for subsequent dynamic scaling of stability margin distribution and updating of joint planning optimization results.

[0077] A weighted least squares algorithm is employed, using the slope of local thermal drift, fluctuations in surface energy absorption efficiency, and duration of regional response hysteresis in the monitored dataset as dynamic weighting factors to dynamically scale the initial stability margin distribution in a matrix manner. By adjusting the margin weight coefficients of each discrete unit, an updated stability margin distribution matching the real-time processing state is formed. For local abnormal regions where the scaled margin is lower than a preset threshold, a PID control algorithm is used to dynamically adjust the output amplitude of the local laser power density, incorporating local risk adjustment parameters from the tissue evolution risk information set. Furthermore, local segments of the scanning path are optimized through Bezier curve interpolation. The dwell time is calculated, and the triggering logic and path planning for the secondary scan are supplemented based on the time-series scheduling algorithm to complete targeted local enhancement processing. The dynamically scaled stability margin distribution is used as the updated risk weight field and integrated into the original joint planning optimization framework. The particle swarm optimization algorithm is used to perform a second iteration to solve the three-dimensional joint decision variables of the laser action time energy distribution, spatial action trajectory and scanning rhythm. The weight coefficients of the tissue evolution risk accumulation, trajectory bias requirement and local energy input limit in the composite performance index are simultaneously calibrated. Finally, the updated joint planning optimization result is output to realize the closed-loop optimization adjustment of the processing process.

[0078] In one possible implementation, step S400 further includes:

[0079] The joint planning optimization results, initial microstructure state, surface aggregate structure characteristics, material thermal diffusion non-uniformity, and strengthening treatment detection results are constructed into a mapped storage data group, and encrypted storage management is performed.

[0080] Specifically, a comprehensive data normalization process is first performed on the joint planning optimization results, initial microstructure, surface aggregate structure characteristics, material thermal diffusion inhomogeneity, and strengthening treatment detection results. The joint planning optimization results include the temporal energy distribution, spatial trajectory, and scanning rhythm parameters of the laser action; the initial microstructure includes the initial phase composition ratio, dislocation density, grain morphology factor, and carbide distribution density; the surface aggregate structure characteristics include data related to grain orientation distribution, surface structure interval boundaries, and concentrated areas of surface micro-defects; the material thermal diffusion inhomogeneity involves the gradient change of the thermal diffusivity coefficient with location, the perturbation bandwidth of surface thermal conductivity, and thermal trapping characteristic parameters; and the strengthening treatment detection results include indicators such as hardness distribution of the hardened layer, martensitic phase transformation rate, and layer thickness uniformity. These data are standardized in format and dimensionless. Subsequently, according to the mapping logic of core parameters, associated features, and detection results, a key-value pair structured mapping data set is constructed to store the data, clarifying the relationships and index paths between each data item. The AES-256 symmetric encryption algorithm is used to encrypt the entire mapped storage data group, generating encrypted data packets. Simultaneously, a unique decryption key is generated by combining this with the device's unique identifier. Finally, the encrypted data packets and decryption keys are stored separately. The encrypted data packets are stored in a secure database with a strict access control mechanism, while the decryption keys are separately stored through a hardware encryption module or a trusted key management system. A tiered authorization encryption storage management process is implemented to ensure the integrity, confidentiality, and traceability of the entire data group.

[0081] Example 2 is based on the same inventive concept as the laser-hardened mold steel surface strengthening treatment method in the previous examples, such as... Figure 2 As shown, this application provides a laser-hardened surface strengthening treatment system for mold steel. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0082] The multidimensional state space model construction module 10 is used to construct a multidimensional state space model characterizing the surface microstructure evolution behavior of the mold steel based on the initial microstructure state, surface aggregate structure characteristics and material thermal diffusion non-uniformity of the mold steel before performing laser strengthening treatment. The multidimensional state space model includes a microstructure phase stability region, a phase transformation sensitive region and the corresponding microstructure evolution critical transition surface.

[0083] The stability margin distribution calculation module 20 is used to determine the target tissue evolution trajectory in the multidimensional state space model according to the enhanced performance requirements corresponding to the target service conditions, and to calculate the stability margin distribution of the target tissue evolution trajectory relative to the tissue evolution critical transition surface.

[0084] The joint planning and optimization module 30 is used to perform joint planning and optimization of the temporal energy distribution, spatial trajectory and scanning rhythm of laser action during the laser enhancement process, using the evolutionary pose and stability margin change trend of the current tissue evolution state in the multidimensional state space as key control variables and the stability margin distribution as the tissue evolution risk weight field.

[0085] The control and management module 40 is used to perform laser enhancement execution control and management based on the joint planning optimization results.

[0086] Furthermore, the system is also used to implement the following functions:

[0087] The surface of the mold steel is discretized according to a predetermined spatial scale, and the stability margin value of the discrete unit is encoded to form a risk weight matrix. Based on the risk weight matrix, directional information reflecting the spatial change trend of the stability margin is extracted, and a risk guidance field is constructed according to the directional information. The risk guidance field is used to constrain the scanning path planning. The directional information is used to identify stability margin changes and construct key monitoring units with phase transition tendencies. For the key monitoring units, the stability margin change patterns of adjacent regions are extracted from the risk weight matrix and converted into a set of local risk adjustment parameters. Based on the risk weight matrix, the risk guidance field, and the set of local risk adjustment parameters, a structured set of tissue evolution risk information is constructed. Based on the set of tissue evolution risk information, the joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of laser action are performed.

[0088] Furthermore, the system is also used to implement the following functions:

[0089] The temporal energy distribution, spatial action trajectory, and scanning rhythm are used as three-dimensional joint decision variables. Based on the organizational evolution risk information set, a composite performance index for joint planning optimization is constructed. The composite performance index includes the cumulative amount of organizational evolution risk calculated according to the risk weight matrix, the trajectory bias requirement determined based on the risk guidance field, the local energy input limit constrained by the local risk adjustment parameter set, and the enhanced organizational deviation calculated based on the target organizational evolution trajectory. The composite performance index is used as the optimization objective, and the three-dimensional joint decision variables are solved in stages to complete the joint planning optimization.

[0090] Furthermore, the system is also used to implement the following functions:

[0091] A preset scanning path framework conforming to the directional constraints of the risk guidance field is established using a global path generator. Within the preset scanning path framework, local optimization is used to couple the temporal energy distribution and scanning rhythm, establishing the coupled solution results. Based on the coupled solution results, local segments of the preset scanning path framework are dynamically reconstructed. The dynamic reconstruction includes real-time fine-tuning of the turning angles, coverage widths, and overlapping areas between path segments according to a set of local risk adjustment parameters, so that the cumulative risk of local segments decreases to a preset threshold range. The dynamically reconstructed preset scanning path framework is then subjected to global consistency correction. The path drift, rhythm differences, and energy input deviations introduced by the local fine-tuning are used to construct a global offset. The global offset is then used to uniformly calibrate the reference frequency of the scanning rhythm and the reference flux of the energy input globally. After completing the global consistency correction, the temporal energy distribution, spatial action trajectory, and scanning rhythm are jointly summarized to form a joint planning optimization result.

[0092] Furthermore, the system is also used to implement the following functions:

[0093] The initial microstructure includes the initial phase composition ratio, dislocation density, grain morphology factor, and carbide distribution density, which serves as the structural basis for the microstructure evolution space. The surface aggregate structure features include grain orientation distribution, surface structure interval boundaries, and concentrated areas of surface microdefects. The surface aggregate structure is adjusted by regional merging to construct surface orientation partitions. The thermal diffusion non-uniformity of the material is spatially mapped, and a thermal diffusion response matrix is ​​formed based on the gradient change of the thermal diffusion coefficient with position, the perturbation bandwidth of surface thermal conductivity, and thermal trapping characteristics. Based on the structural basis, surface orientation partitions, and thermal diffusion response matrix, a multidimensional state-space model characterizing the surface microstructure evolution behavior of the mold steel is constructed.

[0094] Furthermore, the system is also used to implement the following functions:

[0095] Based on the aforementioned structural substrate, surface orientation partitions, and thermal diffusion response matrix, a multidimensional tissue evolution state vector is established. This multidimensional tissue evolution state vector includes a phase sensitivity factor, a thermally induced nucleation potential factor, a thermal flux response coefficient, and a micro-region tissue evolution gain coefficient. Based on this multidimensional tissue evolution state vector, the tissue phase stability region, the phase transition sensitive region, and the tissue evolution critical transition surface are determined to construct a multidimensional state space model.

[0096] Furthermore, the system is also used to implement the following functions:

[0097] The stable region of tissue phase is constructed based on the stable clustering of phase sensitivity factors in the low sensitivity range. The phase change sensitive region is determined based on the combination of thermally induced nucleation potential factor and thermal flux response coefficient coupling abrupt change. The critical transition surface of tissue evolution forms a continuous transition boundary according to the micro-region tissue evolution gain coefficient crossing the critical response threshold.

[0098] Furthermore, the system is also used to implement the following functions:

[0099] During the actual processing, local thermal drift trends, changes in surface energy absorption, and regional response hysteresis are identified, and a monitoring dataset is constructed. The monitoring dataset is then used for dynamic scaling and local enhancement processing of the stability margin distribution to update the joint planning optimization results.

[0100] Furthermore, the system is also used to implement the following functions:

[0101] The joint planning optimization results, initial microstructure state, surface aggregate structure characteristics, material thermal diffusion non-uniformity, and strengthening treatment detection results are constructed into a mapped storage data group, and encrypted storage management is performed.

[0102] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0103] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0104] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for surface strengthening treatment of mold steel based on laser hardening, characterized in that, The method includes: Before performing laser strengthening treatment, a multidimensional state space model characterizing the surface microstructure evolution behavior of the mold steel is constructed based on the initial microstructure state, surface aggregate structure characteristics and material thermal diffusion non-uniformity of the mold steel. The multidimensional state space model includes a microstructure phase stability region, a phase transformation sensitive region and the corresponding microstructure evolution critical transition surface. Based on the enhanced performance requirements corresponding to the target service conditions, the target tissue evolution trajectory is determined in the multidimensional state space model, and the stability margin distribution of the target tissue evolution trajectory relative to the critical transition surface of tissue evolution is calculated. During laser enhancement, the evolutionary pose and stability margin change trend of the current tissue evolution state in the multidimensional state space are used as key control variables, and the stability margin distribution is used as the tissue evolution risk weight field. The joint planning and optimization of the temporal energy distribution, spatial trajectory and scanning rhythm of laser action are performed. The execution control management of laser enhancement is carried out based on the joint planning and optimization results; The joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of laser action includes: The surface of the mold steel is discretized according to a predetermined spatial scale, and the stability margin value of the discrete unit is encoded to form a risk weight matrix. Based on the risk weight matrix, directional information reflecting the trend of stability margin space change is extracted, and a risk guidance field is constructed according to the directional information. The risk guidance field is used to constrain scan path planning. The directional information is used to identify changes in stability margin, and a key monitoring unit for phase transition tendency is constructed. For the key monitoring unit, the stability margin change pattern of adjacent regions is extracted from the risk weight matrix and converted into a local risk adjustment parameter set; Based on the risk weight matrix, risk guidance field, and local risk adjustment parameter set, a structured tissue evolution risk information set is constructed. Based on the tissue evolution risk information set, joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of laser action are performed. Based on the initial microstructure, surface aggregate structure characteristics, and non-uniform thermal diffusion of mold steel, a multi-dimensional state-space model characterizing the surface microstructure evolution behavior of mold steel is constructed, including: The initial microstructure includes the initial phase composition ratio, dislocation density, grain morphology factor and carbide distribution density, and the initial microstructure is used as the structural basis of the microstructure evolution space. The surface aggregate structure features include grain orientation distribution, surface structure interval boundaries, and surface micro-defect concentration areas. The surface aggregate structure is adjusted by regional merging to construct surface orientation partitions. The thermal diffusion non-uniformity of the material is spatially mapped, and a thermal diffusion response matrix is ​​formed based on the gradient change of the thermal diffusion coefficient with position, the perturbation bandwidth of the surface thermal conductivity, and the thermal trapping characteristics. Based on the aforementioned structural substrate, surface orientation partitioning, and thermal diffusion response matrix, a multidimensional state-space model characterizing the surface microstructure evolution behavior of mold steel is constructed. The execution control management of laser enhancement is carried out based on the joint planning and optimization results, including: During the actual processing, local thermal drift trends, changes in surface energy absorption, and regional response hysteresis are identified, and a monitoring dataset is constructed. The monitoring dataset is used to perform dynamic scaling and local reinforcement processing on the stability margin distribution in order to update the joint programming optimization results.

2. The method for surface strengthening treatment of mold steel based on laser hardening as described in claim 1, characterized in that, The joint planning and optimization of the temporal energy distribution, spatial trajectory, and scanning rhythm of laser treatment based on the aforementioned tissue evolution risk information set includes: The temporal energy distribution, spatial action trajectory, and scanning rhythm are used as three-dimensional joint decision variables; Based on the organizational evolution risk information set, a composite performance index for joint planning optimization is constructed. The composite performance index includes the cumulative amount of organizational evolution risk calculated by weighting according to the risk weight matrix, the trajectory bias requirement determined based on the risk guidance field, the local energy input limit constrained by the local risk adjustment parameter set, and the enhanced organizational deviation calculated based on the target organizational evolution trajectory. Using the composite performance index as the optimization objective, a phased solution of the three-dimensional joint decision variables is performed to complete the joint planning optimization.

3. The method for surface strengthening treatment of mold steel based on laser hardening as described in claim 2, characterized in that, Using the composite performance index as the optimization objective, a phased solution of the three-dimensional joint decision variables is performed to complete the joint programming optimization, including: A preset scanning path framework that conforms to the directional constraints of the risk guidance field is established using a global path generator; Within the preset scanning path framework, local optimization is used to couple the solution of time energy distribution and scanning rhythm, and the coupled solution result is established. Based on the coupling solution results, the local segments of the preset scanning path framework are dynamically reconstructed. The dynamic reconstruction includes real-time fine-tuning of the turning angle, coverage width and segment overlap area between path segments according to the local risk adjustment parameter group, so that the cumulative risk of the local segments of the path is reduced to a preset threshold range. The preset scan path framework after dynamic reconstruction is subjected to global consistency correction. The path drift, rhythm difference and energy input deviation introduced by local fine-tuning are used to construct a global offset. The global offset is used to uniformly calibrate the scan rhythm reference frequency and energy input reference flux in the global range. After completing the global consistency correction, the temporal energy distribution, spatial action trajectory, and scanning rhythm are jointly summarized to form the joint planning optimization result.

4. The method for surface strengthening treatment of mold steel based on laser hardening as described in claim 1, characterized in that, Based on the aforementioned structural substrate, surface orientation partitioning, and thermal diffusion response matrix, a multidimensional state-space model characterizing the surface microstructure evolution behavior of mold steel is constructed, including: A multidimensional tissue evolution state vector is established based on the aforementioned structural substrate, surface orientation partitioning, and thermal diffusion response matrix. The multidimensional tissue evolution state vector includes a phase sensitivity factor, a thermally induced nucleation potential factor, a thermal flow response coefficient, and a micro-region tissue evolution gain coefficient. Based on the multidimensional tissue evolution state vector, determine the tissue phase stability region, phase transition sensitive region, and critical transition surface of tissue evolution to construct a multidimensional state space model.

5. The method for surface strengthening treatment of mold steel based on laser hardening as described in claim 4, characterized in that, The stable region of tissue phase is constructed based on the stable clustering of phase sensitivity factors in the low sensitivity range. The phase change sensitive region is determined based on the combination of thermally induced nucleation potential factor and thermal flux response coefficient coupling abrupt change. The critical transition surface of tissue evolution forms a continuous transition boundary according to the micro-region tissue evolution gain coefficient crossing the critical response threshold.

6. The method for surface strengthening treatment of mold steel based on laser hardening as described in claim 1, characterized in that, The joint planning optimization results, initial microstructure state, surface aggregate structure characteristics, material thermal diffusion non-uniformity, and strengthening treatment detection results are constructed into a mapped storage data group, and encrypted storage management is performed.

7. A surface strengthening treatment system for mold steel based on laser hardening, characterized in that, The system is used to implement the laser-hardened surface strengthening treatment method for mold steel according to any one of claims 1-6, the system comprising: The multidimensional state space model construction module is used to construct a multidimensional state space model characterizing the surface microstructure evolution behavior of the mold steel based on the initial microstructure state, surface aggregate structure characteristics and material thermal diffusion non-uniformity of the mold steel before performing laser strengthening treatment. The multidimensional state space model includes a microstructure phase stability region, a phase transformation sensitive region and the corresponding microstructure evolution critical transition surface. The stability margin distribution calculation module is used to determine the target tissue evolution trajectory in the multidimensional state space model according to the enhanced performance requirements corresponding to the target service conditions, and to calculate the stability margin distribution of the target tissue evolution trajectory relative to the tissue evolution critical transition surface. The joint planning and optimization module is used in the laser enhancement process to perform joint planning and optimization of the temporal energy distribution, spatial trajectory and scanning rhythm of laser action, taking the evolutionary pose and stability margin change trend of the current tissue evolution state in the multidimensional state space as the key control quantity, and the stability margin distribution as the tissue evolution risk weight field. The control and management module is used to perform laser enhancement execution control and management based on the joint planning optimization results.

Images