Hydraulic support surface treatment method and system based on laser cladding

Abstract

The invention discloses a hydraulic support surface treatment method and system based on laser cladding, and relates to the technical field related to data processing.The method comprises the steps that a structure data set and a function data set are constructed, and the cladding response precision of a hydraulic support is configured; performing partition clustering, and constructing N surface partitions; establishing a performance index set and N fitness functions; creating a first optimization space and a second optimization space; configuring a temporary memory bank; establishing an initial selected probability, and performing probability search; an attention adjustment factor is constructed, and optimization of the initial selected probability is executed; and performing partition laser cladding control by optimizing the parameter set. The technical problems that according to existing hydraulic support surface treatment, corresponding cladding treatment is difficult to conduct according to accurate subareas of the support surface, so that an optimal parameter combination cannot be found, and then the quality and reliability of hydraulic support surface treatment are poor are solved. The technical effect of improving the surface treatment quality, precision and reliability of the hydraulic support is achieved.

Description

Technical Field

[0001] The present application relates to data processing technology-related fields, and specifically to a surface treatment method and system for a hydraulic support based on laser cladding. Background Art

[0002] With the continuous advancement of industrial technology, hydraulic supports are the main supporting equipment in the comprehensive mining process of coal mines, and their performance directly affects the safe production of coal mines. However, due to the harsh environment of mines, especially in humid and corrosive media environments, the surface of hydraulic supports is prone to rust and scratches, which in turn affects its sealing and lifting force, and even causes the entire set of hydraulic supports to fail, posing serious hidden dangers to the safe production of coal mines. Strengthening the surface of hydraulic supports to improve their wear resistance and corrosion resistance is of great significance to ensuring the safe production of coal mines. Traditional hydraulic support surface treatment mainly uses electroplating chromium technology, but the bonding strength of the electroplating layer is relatively poor and it is easy to crack. The plating layer is prone to fall off when impacted by coal particles, etc., and the quality stability of the hydraulic support cannot be guaranteed.

[0003] Therefore, in the current hydraulic support surface treatment related technologies, it is difficult to perform corresponding cladding treatment according to the precise partitioning of the support surface, making it impossible to find the optimal parameter combination, which in turn leads to technical problems such as poor quality and reliability of the hydraulic support surface treatment. Summary of the invention

[0004] The present application provides a surface treatment method and system for a hydraulic support based on laser cladding, and adopts technical means such as establishing an optimization space, configuring a memory library, and probabilistic search to solve the technical problem that it is difficult to perform corresponding cladding treatment according to the precise partitioning of the support surface in the existing surface treatment of hydraulic supports, making it impossible to find the optimal parameter combination, which in turn leads to poor quality and reliability of the surface treatment of the hydraulic support, thereby achieving the technical effect of improving the quality, accuracy and reliability of the surface treatment of the hydraulic support.

[0005] The present application provides a surface treatment method for a hydraulic support based on laser cladding, the method comprising: acquiring data on surface features of the hydraulic support, constructing a structural data set, a functional data set, and configuring the cladding response accuracy of the hydraulic support; using the cladding response accuracy as the minimum clustering constraint, using the structural data set and the functional data set to normalize the position of the hydraulic support, performing partition clustering, and constructing N surface partitions, wherein the N surface partitions are all regions composed of continuous regions; establishing a performance index set, the performance index set including a coating thickness uniformity index, a coating and substrate bonding strength index, a surface roughness index, and a heat-affected zone index, respectively configuring the performance index set index weights through the N surface partitions, and constructing N fitness functions corresponding to the N surface partitions; creating a first optimization space and a second optimization space, the first optimization space being a data space constructed by collecting laser cladding processing data in a factory, the The second optimization space is a data space constructed based on the laser cladding processing data of big data; a temporary memory library is configured, the temporary memory library contains H groups of memory harmonies, and each group of memory harmonies represents a group of laser cladding parameters, including laser power, scanning speed, focusing distance, powder spraying rate, and spot parameters; the initial selection probabilities of the temporary memory library, the first optimization space, and the second optimization space are established, and a probability search is performed based on the initial selection probability, and the probability search results are evaluated based on the corresponding N fitness functions, and the temporary memory library is iterated; after iterating to a preset period, the temporary memory library is evaluated, and an attention adjustment factor is constructed according to the evaluation result, and the initial selection probability is optimized through the attention adjustment factor, and continuous iteration is completed according to the optimization result; when the iteration result satisfies the end constraint, an optimization parameter set is constructed according to the final temporary memory library, and the partitioned laser cladding control corresponding to the N surface partitions is executed through the optimization parameter set.

[0006] In a possible implementation, after iterating to a preset period, the temporary memory bank is evaluated, an attention adjustment factor is constructed according to the evaluation result, the initial selection probability is optimized through the attention adjustment factor, continuous iteration is completed according to the optimization result, and the following processing is performed: perform iteration number analysis, establish a stage analysis of the current iteration, and construct an iteration stage constraint; obtain the memory harmony in the current temporary memory bank, perform a difference analysis on the memory harmony, and generate an initial attention adjustment factor according to the difference analysis result; optimize the initial attention adjustment factor through the iteration stage constraint, construct the attention adjustment factor, and perform initial selection probability optimization based on the attention adjustment factor.

[0007] In a possible implementation, the memory harmony in the current temporary memory bank is obtained, and a difference analysis is performed on the memory harmony, and the following processing is performed: a fitness analysis is performed based on the memory harmony to construct a preferred solution set; an additional selection probability is configured for the preferred solution set, and when a subsequent iterative probability search is performed, if the probability selection result is the selection result of the temporary memory bank, the memory harmony in the temporary memory bank is selected according to the additional selection probability; the bandwidth is fine-tuned according to the memory harmony selection result, and the temporary memory bank is updated based on the bandwidth fine-tuning result to complete this round of iteration.

[0008] In a possible implementation, the cladding response accuracy is used as the minimum clustering constraint, and after the position of the hydraulic support is normalized and identified using the structural data set and the functional data set, partition clustering is performed to construct N surface partitions, and the following processing is also performed: a normalized ratio of structure and function is configured, and the position normalization identification of the hydraulic support is completed based on the normalized ratio; initial partitioning is performed based on the structural data set and the functional data set, respectively, to establish k structural partitions and b functional partitions; cluster centers are distributed in the corresponding k structural partitions and b functional partitions, and position similarity analysis of structure and function is performed through position normalization identification to perform initial clustering; it is determined whether there is a clustering result in the initial clustering that does not meet the minimum clustering constraint. If not, cluster search continues to be performed, and N surface partitions are constructed according to the clustering search results.

[0009] In a possible implementation, the determination of whether there are clustering results in the initial clustering that do not satisfy the minimum clustering constraint further performs the following processing: if there are clustering results in the initial clustering that do not satisfy the minimum clustering constraint, an upward covering clustering instruction is generated; clustering merging of the corresponding clustering result and the neighborhood is performed through the upward covering clustering instruction, wherein the upward covering clustering instruction is a constraint instruction for constraining the current clustering result to cluster with high surface quality; and N surface partitions are constructed according to the cluster merging result.

[0010] In a possible implementation, the creation of the first optimization space and the second optimization space also performs the following processing: after the first optimization space is created, big data control parameters of the same model equipment are collected based on big data to establish a big data control parameter set; the big data control parameter set and the first optimization space are analyzed for overlap, and a non-overlapping control parameter set in the big data control parameter set is obtained, and the non-overlapping control parameter set is used as the second optimization space.

[0011] In a possible implementation, partitioned laser cladding control corresponding to N surface partitions is performed by optimizing the parameter set, and the following processing is also performed: a control test is performed by optimizing the parameter set to establish a test result; a fitness consistency evaluation of the control parameters in the optimized parameter set is performed based on the test result; N fitness functions are updated according to the consistency evaluation result, and fine-tuning and optimizing the optimized parameter set is performed based on the N fitness functions; and partitioned laser cladding control of the N surface partitions is performed according to the fine-tuning optimization result.

[0012] The present application also provides a surface treatment system for a hydraulic support based on laser cladding, comprising: A data set construction module, the data set construction module is used to acquire data on surface features of the hydraulic support, construct a structural data set, a functional data set, and configure the cladding response accuracy of the hydraulic support; a surface partition construction module, the surface partition construction module is used to use the cladding response accuracy as the minimum clustering constraint, use the structural data set and the functional data set to normalize the position of the hydraulic support, perform partition clustering, and construct N surface partitions, wherein the N surface partitions are all regions composed of continuous regions; a performance indicator set establishment module, the performance indicator set establishment module is used to establish a performance indicator set, the performance indicator set includes a coating thickness uniformity index, a coating and substrate bonding strength index, a surface roughness index, and a heat-affected zone index, and the performance indicator set weights are respectively configured through the N surface partitions to construct N fitness functions corresponding to the N surface partitions; an optimization space creation module, the optimization space creation module is used to create a first optimization space and a second optimization space, the first optimization space is a data space constructed by collecting laser cladding processing data in the factory, and the second optimization space is a data space constructed by laser cladding processing data based on big data. a data space constructed; a temporary memory library configuration module, the temporary memory library configuration module is used to configure the temporary memory library, the temporary memory library contains H groups of memory harmonics, and each group of memory harmonics represents a group of laser cladding parameters, including laser power, scanning speed, focusing distance, powder spraying rate, and spot parameters; an initial selection probability establishment module, the initial selection probability establishment module is used to establish the initial selection probability of the temporary memory library, the first optimization space, and the second optimization space, and perform probability search based on the initial selection probability, evaluate the probability search results based on the corresponding N fitness functions, and iterate the temporary memory library; an attention adjustment factor construction module, the attention adjustment factor construction module is used to evaluate the temporary memory library after iterating to a preset period, construct an attention adjustment factor according to the evaluation result, perform optimization of the initial selection probability through the attention adjustment factor, and complete continuous iteration according to the optimization result; a partitioned laser cladding control module, the partitioned laser cladding control module is used to construct an optimization parameter set according to the final temporary memory library after the iteration result meets the end constraint, and perform partitioned laser cladding control corresponding to N surface partitions through the optimization parameter set.

[0013] Through the surface treatment method and system of hydraulic support based on laser cladding proposed in this application, it is planned to construct a structural data set, a functional data set, and configure the cladding response accuracy of the hydraulic support; perform partition clustering, construct N surface partitions; establish a performance indicator set, N fitness functions; create a first optimization space and a second optimization space; configure a temporary memory library; establish an initial selection probability and perform a probability search; construct an attention adjustment factor and perform optimization of the initial selection probability; and perform partition laser cladding control corresponding to N surface partitions by optimizing the parameter set. The technical problem that it is difficult to perform corresponding cladding treatment according to the precise partition of the support surface in the existing surface treatment of hydraulic supports is solved, which makes it impossible to find the optimal parameter combination, and thus leads to poor quality and reliability of the surface treatment of hydraulic supports, and achieves the technical effect of improving the quality, accuracy and reliability of the surface treatment of hydraulic supports. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic flow chart of a surface treatment method for a hydraulic support based on laser cladding provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a hydraulic support surface treatment system based on laser cladding provided in an embodiment of the present application.

[0016] Explanation of the accompanying drawings: data set construction module 10, surface partition construction module 20, performance indicator set establishment module 30, optimization space creation module 40, temporary memory library configuration module 50, initial selection probability establishment module 60, attention adjustment factor construction module 70, partition laser cladding control module 80. DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The present application embodiment provides a surface treatment method for a hydraulic support based on laser cladding, such as Figure 1 As shown, the method includes: Laser cladding is an advanced surface treatment technology. Through the action of a high-energy laser beam, a coating with special properties is formed on the surface of the base material, thereby significantly improving the wear resistance, corrosion resistance, heat resistance, oxidation resistance and electrical properties of the base material. Specifically, laser cladding technology places the selected coating material on the surface of the coated base in different filler forms, and then melts the coating material and a thin layer on the base surface at the same time through laser irradiation, and quickly solidifies to form a surface coating with extremely low dilution and metallurgical bonding with the base material. The high energy density and high power of the laser beam are used to achieve rapid melting and cooling of the material surface, thereby obtaining a coating with excellent performance.

[0021] Step S100, acquiring data of surface features of the hydraulic support, constructing a structural data set and a functional data set, and configuring the cladding response accuracy of the hydraulic support. The acquisition of surface feature data of hydraulic supports refers to the use of visual inspection (such as high-resolution cameras, etc.) and physical measurement (such as hardness testers, roughness testers, etc.) to obtain surface feature data such as surface morphology, material, hardness, roughness, etc. of hydraulic supports, and the construction of structural data sets and functional data sets based on the acquired surface feature data. Specifically, the structural data set mainly describes the physical structural characteristics of the hydraulic supports, such as size, shape, weight, etc., which helps to analyze the structural stability and bearing capacity of the hydraulic supports, and provide a basis for design optimization and fault diagnosis; the functional data set describes the performance of the hydraulic supports during operation, such as lifting force, sealing, wear resistance, etc., which helps to evaluate whether the performance of the hydraulic supports meets the requirements and provide direction for product improvement and upgrading; the cladding response accuracy refers to the control accuracy of parameters such as the bonding quality between the coating and the substrate, the coating thickness, and the uniformity of the composition when the laser cladding technology forms a coating on the surface of the hydraulic support. For example, the cladding response accuracy of laser cladding is configured according to laser parameters (laser power, scanning speed, spot size, etc.), powder parameters (powder composition, particle size, fluidity, etc.) and substrate characteristics (material, hardness, surface state, etc. of the substrate material), so as to achieve high-quality and high-precision control of the surface coating of the hydraulic support.

[0022] Step S200, taking the cladding response accuracy as the minimum clustering constraint, after normalizing and identifying the position of the hydraulic support with the structural data set and the functional data set, perform partition clustering to construct N surface partitions, wherein the N surface partitions are all regions composed of continuous regions. Taking the cladding response accuracy as the minimum clustering constraint means that in the partition clustering process, the cladding accuracy of any two regions classified as the same surface must meet this set minimum standard, ensuring that in the subsequent partition clustering process, the cladding response accuracy of each region is consistent, thereby ensuring the quality and performance of the coating; performing position normalization on the structural data set and the functional data set of the hydraulic support means mapping each position point or region on the hydraulic support to a unified coordinate system. After the normalization process, each position point or region is identified, which may include its coordinates, structural characteristics (such as shape, size) and functional characteristics (such as lifting force, sealing) and other information; by performing partition clustering on the surface of the hydraulic support, for example, K-means clustering or hierarchical clustering can be used to divide the surface of the hydraulic support into N regions (N is a natural number), each region is composed of a series of continuous position points or regions, and these regions have a certain degree of similarity in cladding response accuracy, structural characteristics and functional characteristics, and meet the minimum constraints of cladding response accuracy. In addition, since each region is continuous, it is easier to achieve uniformity and consistency of the coating during the laser cladding process, so as to achieve high quality and high precision control of the coating in the subsequent laser cladding process.

[0023] In a possible implementation, step S200 further includes step S210, configuring a normalized ratio of structure and function, and completing a normalized position identification of the hydraulic support based on the normalized ratio. According to the structure and function of the hydraulic support, the parameters or indicators that need to be normalized are determined, such as structural dimensions, load-bearing capacity, adjustment range, etc. According to the actual values ​​and value ranges of these parameters or indicators, a normalized ratio is configured for them through a normalization algorithm, such as mapping the actual value to a value between 0 and 1, or mapping to a preset value interval. The position normalization identification refers to converting the actual position information of the hydraulic support into a normalized value, so as to facilitate comparison and analysis on a unified scale, which is helpful to achieve accurate control and optimized management of the hydraulic support. It also includes step S220, performing initial partitioning based on the structural data set and the functional data set, respectively, to establish k structural partitions and b functional partitions. The k value represents the number of structural partitions, which is determined according to the characteristics of the dataset and the analysis objectives. For example, if there are several obvious different structural types in the dataset, then the k value is set to the number of these types. The b value represents the number of functional partitions. Similar to the determination of the k value, the b value is determined according to the characteristics of the dataset and the analysis objectives. Specifically, using the selected partitioning method, the data in the structural dataset and the functional dataset are divided into k structural partitions and b functional partitions, respectively.

[0024] Step S200 also includes step S230, respectively distributing cluster centers in corresponding k structural partitions and b functional partitions, performing structural and functional positional similarity analysis through position normalization identification, and performing initial clustering. Select the mean of the data in the k structural partitions and b functional partitions (such as the centroid in the K-Means algorithm) as the cluster center of the corresponding partition, and use the position normalization identification to perform structural and functional positional similarity analysis, such as calculating the similarity (such as distance) between different objects (or cluster centers), analyzing the similarity between the object and the cluster center of the partition where it is located, and the similarity with the cluster centers of other partitions. According to the results of the similarity analysis, each object is preliminarily assigned to the partition where the cluster center most similar to it is located, and the initial clustering is completed. It also includes step S240, judging whether there is a clustering result that does not meet the minimum clustering constraint in the initial clustering. If not, continue to perform cluster search, and construct N surface partitions according to the cluster search results. Check whether each initial cluster (or partition) satisfies the minimum cluster constraint. For example, if the minimum cluster size is set to m, any cluster with less than m samples should be considered as not satisfying the constraint. If the initial clustering results all satisfy the minimum cluster constraint, that is, the samples of all initial clusters are greater than m, the cluster search can be continued. When the cluster search is completed and the cluster results that meet the constraints are found, N surface partitions can be constructed based on the cluster search results, representing the surface areas of the hydraulic support with similar characteristic attributes.

[0025] In a possible implementation, step S240 further includes step S241, if there are clustering results in the initial clustering that do not meet the minimum clustering constraint, then an upward covering clustering instruction is generated. The upward covering clustering instruction is a specific clustering adjustment strategy, and its core purpose is to ensure that the clustering results that do not currently meet the minimum clustering constraint can be merged into higher quality clusters. If there are clustering results in the initial clustering that do not meet the minimum clustering constraint, that is, there are some clusters whose sample quantity, density or other indicators are lower than the preset minimum clustering constraint, then an upward covering clustering instruction is generated to adjust the clusters that do not meet the constraint so that they meet the minimum clustering constraint, for example, merging the clusters that do not meet the constraint with adjacent, similar clusters, thereby increasing the size of the cluster or increasing the density of the cluster; for clusters that do not meet the constraint, their boundaries can be expanded outward to include more samples or increase the density of the cluster. The method also includes step S242, in which the corresponding clustering result and the clustering of the neighborhood are merged by the upward covering clustering instruction, wherein the upward covering clustering instruction is a constraint instruction for constraining the current clustering result to cluster with high surface quality. First, the clusters that do not meet the minimum clustering constraint in the initial clustering result are identified. For each cluster that does not meet the constraint, other clusters in its neighborhood are searched, and the possibility of merging the current cluster with each cluster in the neighborhood is evaluated. The possibility of merging mainly depends on the surface quality of the neighborhood cluster. The surface quality may involve multiple factors, such as the smoothness, continuity, area or volume of the cluster. If one or more neighborhood clusters with higher surface quality are found, a merge operation is performed to merge the current cluster with these high-quality clusters into a new cluster, including merging samples of the two clusters into a new cluster, i.e., a high surface quality cluster. The method also includes step S243, in which N surface partitions are constructed according to the cluster merging result. The clustering results are converted into surface partitions with clear boundaries and attributes. Specifically, based on the results of cluster merging, the boundaries of each partition are determined, and a set of attributes or features are assigned to each partition, and finally the surface partitions of N hydraulic supports are obtained.

[0026] Step S300, establish a performance index set, the performance index set includes a coating thickness uniformity index, a coating and substrate bonding strength index, a surface roughness index, and a heat-affected zone index. The performance index set index weights are configured through N surface partitions, and N fitness functions corresponding to the N surface partitions are constructed. Specifically, the coating thickness uniformity index measures the uniformity of the coating on the surface of the object. The uniformity of the coating thickness is crucial to the performance and service life of the product. A certain threshold or range can be set to ensure that the uniformity of the coating meets the expected standard; the coating and substrate bonding strength index is the adhesion strength of the coating and the substrate, which is an important parameter for evaluating the quality of the coating. High bonding strength means that the connection between the coating and the substrate is firm and can resist peeling and impact; the surface roughness index measures the unevenness of the small spacing and tiny peaks and valleys on the surface of the hydraulic support. Smaller surface roughness usually means a smoother surface, which has a positive effect on the adhesion and service life of the coating; the heat-affected zone index is an area that undergoes temperature changes but does not reach the melting point, and its performance changes may have an impact on the overall coating quality. For N surface partitions obtained by partition clustering, it is necessary to configure the indicator weight of the performance indicator set for each partition. The weight configuration is based on factors such as the importance of each partition in the hydraulic support, the frequency of use, and the stress conditions. For example, for areas with greater stress, a higher weight can be given to the coating-substrate bonding strength indicator; for areas with higher appearance requirements, a higher weight can be given to the surface roughness indicator; N fitness functions corresponding to the N surface partitions are constructed to evaluate the performance of the performance indicators of the N surface partitions under different configurations (i.e., different indicator weight combinations). Specifically, a target value or expectation is first set for each performance indicator. range, and then calculate the performance index score of each partition according to the index weight and measured value of each partition. Finally, the scores of all partitions are weighted and summed to obtain the fitness value of the entire coating system. For example, assuming that the target value of the coating thickness uniformity index is 90% (that is, the coating thickness deviation is within ±10%), and the measured value of a surface partition is 95%, if the coating thickness uniformity index weight of the surface partition is 0.3, then the coating thickness uniformity index score of the surface partition is 0.3×95%=28.5%. Similarly, the scores of other performance indicators and surface partitions can be calculated, and finally the fitness value of the entire coating system can be obtained.

[0027] Step S400, creating a first optimization space and a second optimization space, wherein the first optimization space is a data space constructed by collecting laser cladding processing data in a factory, and the second optimization space is a data space constructed based on laser cladding processing data of big data. The first optimization space is a data space constructed by collecting laser cladding processing data in the factory. The data in this space mainly comes from the laser cladding processing in the actual production process of the factory, which directly reflects the actual situation and performance of the laser cladding processing in the factory. Specifically, various data in the laser cladding processing process, such as temperature, speed, power, coating thickness, etc., are obtained through sensors and monitoring equipment installed in the factory or directly from production equipment. The collected original data are cleaned, sorted, converted and standardized, and the pre-processed data are integrated into a unified data space to form the first optimization space; the second optimization space is a data space constructed based on big data laser cladding processing data. The data source of this space is more extensive, which may include laser cladding processing data of multiple factories, multiple equipment, multiple materials and process conditions. Specifically, a large amount of laser cladding processing data is collected through web crawlers, data exchange, public data sets, etc., and pre-processed and integrated. The integrated data is deeply analyzed and mined using data mining technology to discover the laws and patterns in the data, thereby forming the second optimization space.

[0028] In a possible implementation, step S400 further includes step S410, after the first optimization space is created, big data control parameters of the same type of equipment are collected based on big data to establish a big data control parameter set. The same type of equipment may operate in different working environments, and its control parameters may be different. The control parameter data are collected by using web crawlers (used to capture data from online databases or monitoring systems), sensor technology (used to collect equipment operating status data in real time), etc., including the optimal control parameter settings of the equipment under different working conditions, parameter adjustment ranges, and the relationship between parameters and equipment performance, etc., to reflect the performance of the equipment under different working conditions, and finally establish a big data control parameter set containing multiple control parameters and their value ranges, that is, a big data control parameter set. It also includes step S420, performing overlap analysis on the big data control parameter set and the first optimization space, obtaining a non-overlapping control parameter set in the big data control parameter set, and using the non-overlapping control parameter set as the second optimization space. Compare and analyze the big data control parameter set with the first optimization space to find out the overlapping and non-overlapping parts between the two. Through the overlap analysis, understand the degree of match between the actual operating conditions of the equipment reflected in the big data and the initially set optimization space. Specifically, the non-overlapping control parameter set refers to the parameter set in the big data control parameter set but has no overlap with the first optimization space, which represents the optimal operating state of the equipment under certain specific conditions. Use the non-overlapping control parameter set as the second optimization space, that is, include these parameters that have not been fully considered before into the new search range for more in-depth optimization and search.

[0029] Step S500, configure a temporary memory library, the temporary memory library contains H groups of memory harmonics, and each group of memory harmonics represents a group of laser cladding parameters, including laser power, scanning speed, focusing distance, powder spraying rate, and spot parameters. The temporary memory library is a database for storing and retrieving temporary data. It is used to store parameter information related to laser cladding processing, and contains H groups of memory harmonics. Each group of memory harmonics represents a specific set of laser cladding parameters. Specifically, laser power is the energy output by the laser per unit time, which directly affects the quality and efficiency of the cladding layer. Laser cladding generally uses kW-class lasers, and the power range may be from a few kilowatts to tens of thousands of watts; the scanning speed indicates the speed at which the laser beam moves on the surface of the workpiece, which determines the efficiency of the cladding process and the microstructure of the cladding layer. Different scanning speeds are set according to different equipment and materials; the focusing distance The distance is the distance from the laser beam outlet to the workpiece surface, which affects the energy density of the laser beam and the uniformity of the cladding layer. The focusing distance needs to be adjusted according to the specific equipment and materials. The powder injection rate refers to the speed of adding powder to the molten pool, which determines the composition and performance of the cladding layer. The powder injection rate needs to match the laser power and scanning speed to ensure sufficient melting of the powder and the uniformity of the cladding layer. The spot parameters include the shape (such as circle, rectangle, etc.) and size of the spot. The size of the spot affects the energy density of the laser beam and the microstructure of the cladding layer. The selection of the spot parameters needs to be adjusted according to the specific processing object and process requirements. Configuring a temporary memory library specifically refers to defining a data structure to store each set of laser cladding parameters, and then creating a temporary memory library to store these data structures, adding each set of laser cladding parameters as a set of memory harmony to the temporary memory library, which can be accessed and retrieved through a unique identifier (such as index number, name, etc.). For example, a set of laser cladding parameters (such as laser power of 5000W, scanning speed of 10m / min, focusing distance of 10mm, powder spraying rate of 20g / min, spot shape of circular and diameter of 5mm) is added to the temporary memory library as a set of memory harmony, and this set of laser cladding parameters is retrieved from the temporary memory library through a unique identifier.

[0030] Step S600, establish the initial selection probability of the temporary memory bank, the first optimization space, and the second optimization space, and perform a probability search based on the initial selection probability, evaluate the probability search results based on the corresponding N fitness functions, and iterate the temporary memory bank. The initial selection probability is the initial possibility of each parameter or parameter combination being selected during the search process, and can be initialized based on strategies such as uniform distribution and Gaussian distribution. For example, under uniform distribution, the probability of each parameter or parameter combination being selected is equal. According to the established initial selection probability, randomly select parameters or parameter combinations from the first optimization space and the second optimization space. For each parameter or parameter combination obtained by the search, use N fitness functions for evaluation. By calculating the fitness value of each parameter or parameter combination, its pros and cons in the current search space can be determined. According to the evaluation results of the search results, update the temporary memory bank, for example, by replacing poor candidate solutions, adding new high-quality solutions, or performing some form of memory update (such as forgetting strategy).

[0031] Step S700, after the iteration reaches a preset period, the temporary memory bank is evaluated, an attention adjustment factor is constructed according to the evaluation result, the initial selection probability is optimized through the attention adjustment factor, and continuous iteration is completed according to the optimization result. When the iteration reaches a preset period (such as every 10 iterations, every 50 iterations, etc.), the candidate solutions in the temporary memory bank are evaluated as a whole, for example, the average fitness, optimal fitness, fitness distribution and other indicators of all candidate solutions are calculated to understand the overall quality and distribution of the candidate solutions in the current search process, and the attention adjustment factor is constructed according to the evaluation result of the memory bank. The attention adjustment factor is used to guide the adjustment direction of the initial selection probability in the next round of iteration. For example, if a certain type of candidate solution performs well in the memory bank evaluation (such as having a higher average fitness or optimal fitness), then the candidate solution of this type can be improved in the next round of iteration. On the contrary, if a certain type of candidate solution performs poorly, its initial selection probability can be reduced; according to the constructed attention adjustment factor, the initial selection probability is optimized by updating the distribution function of the initial selection probability, adjusting the parameters of the probability distribution, etc. For example, if a uniform distribution is used as the initial selection probability, the upper and lower limits or the center position of the uniform distribution can be adjusted according to the attention adjustment factor; if a Gaussian distribution is used, parameters such as the mean and standard deviation can be adjusted; continuous iteration is completed according to the optimization results, that is, continuous iteration is performed using the optimized initial selection probability to continuously update the temporary memory bank.

[0032] In a possible implementation, step S700 further includes step S710, performing iteration number analysis, establishing a stage analysis of the current iteration, and constructing an iteration stage constraint. The iteration number analysis includes counting the time required for each iteration, the trend of the number of iterations, the change in the quality of the solution during the iteration process, etc., understanding the convergence speed and stability of the iteration, and judging whether the iterative algorithm meets the preset number of iterations constraint, or whether the iteration strategy needs to be adjusted to improve efficiency. The current iteration stage analysis is to evaluate and analyze the performance of the iterative algorithm at the current stage (such as within a preset number of iterations), including calculating the average number of iterations in the current stage, the quality change rate of the solution, the coverage of the search space, etc., and it can be judged whether the iteration strategy needs to be adjusted, such as increasing or decreasing the number of iterations, changing the search direction, etc., to optimize the search process. The iteration stage constraint is a restriction condition set for different stages of the iterative algorithm, which can be set based on factors such as the number of iterations, the quality of the solution, and the change of the search space. For example, a maximum number of iterations constraint can be set to avoid an infinite loop of the algorithm; a solution quality threshold constraint can be set to ensure that the searched solution meets certain requirements; and a search space change constraint can also be set to limit the search range and avoid unnecessary searches.

[0033] Step S700 also includes step S720, obtaining the memory harmony in the current temporary memory bank, performing difference analysis on the memory harmony, and generating an initial attention adjustment factor according to the difference analysis result. Difference analysis is to evaluate the similarity or difference between different memory harmonies, for example, using Euclidean distance, Manhattan distance, cosine similarity, etc. to calculate the difference between different memory harmonies, and generating an initial attention adjustment factor based on the difference analysis result, which is used to guide the selection and adjustment of memory harmonies in subsequent iterations. Specifically, if the difference analysis result shows that there are large differences between some memory harmonies, the initial attention adjustment factor will tend to increase the probability of these memory harmonies with large differences being selected in subsequent iterations to promote extensive exploration of the search space; on the contrary, if the difference analysis result shows that some memory harmonies are relatively similar, the initial attention adjustment factor will reduce the probability of these memory harmonies being selected in subsequent iterations to avoid falling into a local optimal solution. It also includes step S730, optimizing the initial attention adjustment factor through the iteration stage constraint, constructing the attention adjustment factor, and performing initial selection probability optimization based on the attention adjustment factor. The initial attention adjustment factor is optimized using the iterative stage constraints to ensure that the adjustment factor not only considers the differences in the memory harmonies, but also the stage requirements and constraints of the iterative process. For example, if the iterative stage constraints require rapid exploration of the search space within a certain stage, the initial attention adjustment factor may be optimized to be more inclined to select memory harmonies with greater differences. The optimized initial attention adjustment factor will be used to construct the final attention adjustment factor. Based on the constructed attention adjustment factor, the initial selection probability can be optimized. For example, the weights of each memory harmony can be adjusted according to the attention adjustment factor, or a certain probability allocation strategy can be used to redistribute the initial selection probability, so that those memory harmonies that are more likely to produce high-quality solutions have a higher probability of being selected in subsequent iterations.

[0034] In a possible implementation, step S720 further includes step S721, performing fitness analysis based on memory harmony to construct a preferred solution set. A fitness function is defined for each memory harmony (candidate solution), and the fitness value of each memory harmony is calculated. The higher the fitness value, the better the quality of the memory harmony, and the more likely it is to be a high-quality solution to the problem. According to the fitness value of the memory harmony, a portion of high-quality candidate solutions are selected to construct a preferred solution set. It also includes step S722, configuring an additional selection probability for the preferred solution set. When performing a subsequent iterative probability search, if the probability selection result is the selection result of the temporary memory bank, the memory harmony in the temporary memory bank is selected using the additional selection probability. An additional selection probability is configured for each solution in the preferred solution set, which is usually higher than the memory harmony in the non-preferred solution set, ensuring that it has a greater chance of being selected in the subsequent search process. The value of the additional selection probability can be adjusted according to the actual situation, for example, according to the fitness value of the solution, the diversity of the solution set, etc. In the subsequent probability search process, if the solution selected according to the probability distribution happens to be in the temporary memory bank (that is, the solution generated in the previous iteration process), it is necessary to select a specific memory harmony from the temporary memory bank. When it is necessary to select a memory harmony from the temporary memory bank, the algorithm will give priority to the solution in the preferred solution set. Specifically, the solution is selected from the preferred solution set according to the additional selection probability, rather than simply selecting in order or randomly in the temporary memory bank, thereby improving the search efficiency and the quality of the solution. It also includes step S723, performing bandwidth fine-tuning according to the memory harmony selection result, updating the temporary memory bank based on the bandwidth fine-tuning result, and completing this round of iteration. Bandwidth fine-tuning refers to fine-tuning one or more parameters of the selected memory harmony. The adjustment amplitude is determined by the pitch fine-tuning bandwidth, which is a preset maximum adjustment amplitude parameter. After bandwidth fine-tuning is completed, a new set of harmonies (candidate solutions) will be generated and compared with the existing harmonies in the temporary memory bank. If the new harmony is better than one or some harmonies in the temporary memory bank in terms of fitness or other evaluation indicators, it will be added to the temporary memory bank and replace the poorly performing harmony, thereby updating the temporary memory bank and completing this round of iteration.

[0035] Step S800, when the iteration result satisfies the end constraint, an optimization parameter set is constructed according to the final temporary memory library, and the partition laser cladding control corresponding to the N surface partitions is executed through the optimization parameter set. The iteration process continues until a certain end constraint is met, such as reaching a preset maximum number of iterations, the searched solution has no obvious improvement in multiple consecutive iterations, the quality of the solution (such as fitness value) reaches a preset threshold, etc. Specifically, when the iteration meets the end constraint, the optimal candidate solution is selected from the final temporary memory library as the optimization parameter set. This optimization parameter set usually contains one or more groups of laser cladding parameters, such as laser power, scanning speed, focusing distance, powder spraying rate, etc. If there are multiple high-quality solutions (such as solutions with similar fitness values) in the temporary memory library, one or more groups of solutions can be selected as the optimization parameter set as needed. For each partition, the laser cladding equipment is set according to the corresponding parameters in the optimization parameter set to adapt to the differences in materials, defect types or locations of different partitions. On each partition, cladding repair is performed according to the set laser cladding parameters to ensure the quality and stability of the cladding layer. During the laser cladding process, it is necessary to monitor the quality, temperature distribution and other parameters of the cladding layer in real time, and adjust them as needed to ensure that the cladding process is carried out in the optimal state, thereby improving the service life of the hydraulic support and the effect of laser cladding repair.

[0036] In a possible implementation, step S800 further includes step S810, performing a control test by optimizing the parameter set and establishing a test result. Performing a control test using the optimized parameter set specifically refers to configuring the laser cladding equipment using the optimized parameter set, performing the test, and collecting and organizing the test data after performing the control test, and establishing the test result, including the test data, the test result record, the performance index, etc. It also includes step S820, performing a fitness consistency evaluation of the control parameters in the optimized parameter set based on the test result. Evaluate the performance of each control parameter in the optimized parameter set in actual application, and check whether the control parameter meets the expected target or standard. For example, the consistency can be evaluated by comparing the difference between the test result and the expected target or standard. If the difference between the test result and the expected target or standard is small, it indicates that the fitness of the control parameter is high and the consistency is good. It also includes step S830, updating N fitness functions according to the consistency evaluation results, and fine-tuning the optimization parameter set based on the N fitness functions. According to the results of the consistency evaluation, the N fitness functions are updated. For example, their calculation method, weight distribution, threshold setting, etc. can be adjusted to better reflect the actual effect of the parameter combination. The updated N fitness functions are used to fine-tune the optimization parameter set. Fine-tuning optimization is an iterative optimization method that continuously tries different parameter combinations and calculates their fitness values ​​to find the optimal or near-optimal parameter configuration. Specifically, in the fine-tuning optimization process, the parameter set is continuously iterated and optimized, and control tests are performed based on the new parameter set. The test results are collected and analyzed, and the performance of the new parameter set is evaluated. If the test results do not meet the requirements, the fitness function or optimization algorithm is adjusted according to the test results, and the iterative optimization is continued until a set of optimized parameter sets with high fitness and good consistency is obtained. It also includes step S840, which performs partitioned laser cladding control of N surface partitions according to the fine-tuning optimization results. Based on the optimized parameter set obtained by fine-tuning optimization, specific laser cladding parameters are set for each partition, and laser cladding equipment is used to perform laser cladding operations on each partition according to the set parameters, including laser beam scanning, powder feeding, molten pool cooling, etc. During the cladding process, sensors and control systems are used to monitor the cladding process in real time, and parameters are fine-tuned according to actual conditions. This can ensure the consistency and stability of the cladding quality.

[0037] In the above, refer to Figure 1 The surface treatment method of the hydraulic support based on laser cladding according to the embodiment of the present invention is described in detail. Figure 2 A surface treatment system for a hydraulic support based on laser cladding according to an embodiment of the present invention is described.

[0038] The surface treatment system of the hydraulic support based on laser cladding according to the embodiment of the present invention is used to solve the technical problem that it is difficult to perform the corresponding cladding treatment according to the precise partitioning of the support surface in the existing surface treatment of the hydraulic support, which makes it impossible to find the optimal parameter combination, thereby resulting in poor quality and reliability of the surface treatment of the hydraulic support, and achieves the technical effect of improving the quality, accuracy and reliability of the surface treatment of the hydraulic support. The surface treatment system of the hydraulic support based on laser cladding includes: a data set construction module 10, a surface partition construction module 20, a performance indicator set establishment module 30, an optimization space creation module 40, a temporary memory library configuration module 50, an initial selection probability establishment module 60, an attention adjustment factor construction module 70, and a partition laser cladding control module 80.

[0039] A data set construction module 10, the data set construction module 10 is used to acquire data of surface features of the hydraulic support, construct a structural data set, a functional data set, and configure the cladding response accuracy of the hydraulic support; a surface partition construction module 20, the surface partition construction module 20 is used to use the cladding response accuracy as the minimum clustering constraint, use the structural data set and the functional data set to normalize the position of the hydraulic support, perform partition clustering, and construct N surface partitions, wherein the N surface partitions are all regions composed of continuous regions; a performance indicator set establishment module 30, the performance indicator set establishment module 30 is used to establish a performance indicator set, the performance indicator set includes a coating thickness uniformity index, a coating and substrate bonding strength index, a surface roughness index, and a heat affected zone index, and the performance indicator set index weights are respectively configured through the N surface partitions to construct N fitness functions corresponding to the N surface partitions; an optimization space creation module 40, the optimization space creation module 40 is used to create a first optimization space and a second optimization space, the first optimization space is a data space constructed by collecting laser cladding processing data in the factory, and the second optimization space is a data space constructed by laser cladding processing data based on big data. A data space is created; a temporary memory library configuration module 50, the temporary memory library configuration module 50 is used to configure a temporary memory library, the temporary memory library contains H groups of memory harmonics, and each group of memory harmonics represents a group of laser cladding parameters, including laser power, scanning speed, focusing distance, powder spraying rate, and spot parameters; an initial selection probability establishment module 60, the initial selection probability establishment module 60 is used to establish the initial selection probability of the temporary memory library, the first optimization space, and the second optimization space, and perform a probability search based on the initial selection probability, and evaluate the probability search results based on the corresponding N fitness functions , iterate the temporary memory library; pay attention to the adjustment factor construction module 70, the attention adjustment factor construction module 70 is used to evaluate the temporary memory library after iterating to a preset period, construct the attention adjustment factor according to the evaluation result, perform the optimization of the initial selection probability through the attention adjustment factor, and complete the continuous iteration according to the optimization result; the partitioned laser cladding control module 80, the partitioned laser cladding control module 80 is used to construct the optimization parameter set according to the final temporary memory library when the iteration result meets the end constraint, and perform the partitioned laser cladding control corresponding to the N surface partitions through the optimization parameter set.

[0040] The specific configuration of the attention adjustment factor construction module 70 will be described in detail below. The attention adjustment factor construction module 70 further includes: performing iteration number analysis, establishing the current iteration phase analysis, and constructing the iteration phase constraint; obtaining the memory harmony in the current temporary memory bank, performing difference analysis on the memory harmony, and generating the initial attention adjustment factor according to the difference analysis result; optimizing the initial attention adjustment factor through the iteration phase constraint, constructing the attention adjustment factor, and performing the initial selection probability optimization based on the attention adjustment factor.

[0041] The following will continue to describe in detail the specific configuration of the attention adjustment factor construction module 70. The attention adjustment factor construction module 70 may further include: performing fitness analysis based on memory harmony to construct a preferred solution set; configuring an additional selection probability for the preferred solution set, and when performing subsequent iterative probability search, if the probability selection result is the selection result of the temporary memory bank, then selecting the memory harmony in the temporary memory bank through the additional selection probability; performing bandwidth fine-tuning according to the memory harmony selection result, and updating the temporary memory bank based on the bandwidth fine-tuning result to complete this round of iteration.

[0042] The specific configuration of the surface partition construction module 20 will be described in detail below. The surface partition construction module 20 may further include: configuring a normalized ratio of structure and function, and completing a normalized position identification of the hydraulic support based on the normalized ratio; performing initial partitioning based on the structure data set and the function data set, respectively, to establish k structure partitions and b function partitions; distributing cluster centers in the corresponding k structure partitions and b function partitions, respectively, performing position similarity analysis of the structure and function through position normalized identification, and performing initial clustering; determining whether there is a clustering result in the initial clustering that does not meet the minimum clustering constraint, and if not, continuing to perform cluster search, and constructing N surface partitions according to the clustering search results.

[0043] The specific configuration of the surface partition construction module 20 will be described in detail below. The surface partition construction module 20 further includes: if there are clustering results that do not meet the minimum clustering constraint in the initial clustering, an upward covering clustering instruction is generated; the clustering merging of the corresponding clustering result and the neighborhood is performed through the upward covering clustering instruction, wherein the upward covering clustering instruction is a constraint instruction for constraining the current clustering result to cluster with high surface quality; and N surface partitions are constructed according to the cluster merging result.

[0044] The specific configuration of the optimization space creation module 40 will be described in detail below. The optimization space creation module 40 further includes: after the first optimization space is created, big data control parameter collection of the same type of equipment is performed based on big data to establish a big data control parameter set; overlap analysis is performed on the big data control parameter set and the first optimization space to obtain a non-overlapping control parameter set in the big data control parameter set, and the non-overlapping control parameter set is used as the second optimization space.

[0045] The specific configuration of the partitioned laser cladding control module 80 will be described in detail below. The partitioned laser cladding control module 80 may further include: performing a control test by optimizing a parameter set and establishing a test result; performing a fitness consistency evaluation of the control parameters in the optimized parameter set based on the test result; updating N fitness functions according to the consistency evaluation result, and performing fine-tuning and optimization of the optimized parameter set based on the N fitness functions; and performing partitioned laser cladding control of N surface partitions according to the fine-tuning and optimization result.

[0046] The hydraulic support surface treatment system based on laser cladding provided in the embodiment of the present invention can execute the hydraulic support surface treatment method based on laser cladding provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0047] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0048] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A surface treatment method for a hydraulic support based on laser cladding, characterized in that: The method comprises: Acquire surface feature data of hydraulic supports, build structural data sets and functional data sets, and configure the cladding response accuracy of hydraulic supports; Taking the cladding response accuracy as the minimum clustering constraint, after normalizing and marking the position of the hydraulic support with the structural data set and the functional data set, performing partition clustering to construct N surface partitions, wherein the N surface partitions are all regions composed of continuous regions; Establishing a performance index set, the performance index set includes a coating thickness uniformity index, a coating and substrate bonding strength index, a surface roughness index, and a heat-affected zone index, respectively configuring the performance index set index weights through N surface partitions, and constructing N fitness functions corresponding to the N surface partitions; Creating a first optimization space and a second optimization space, wherein the first optimization space is a data space constructed by collecting laser cladding processing data in a factory, and the second optimization space is a data space constructed based on the laser cladding processing data of big data; A temporary memory bank is configured, wherein the temporary memory bank contains H groups of memory harmonics, and each group of memory harmonics represents a group of laser cladding parameters, including laser power, scanning speed, focusing distance, powder spraying rate, and spot parameters; Establishing the initial selection probabilities of the temporary memory bank, the first optimization space, and the second optimization space, and performing a probability search based on the initial selection probabilities, evaluating the probability search results based on the corresponding N fitness functions, and iterating the temporary memory bank; After the iteration reaches a preset period, the temporary memory bank is evaluated, an attention adjustment factor is constructed according to the evaluation result, the initial selection probability is optimized by the attention adjustment factor, and continuous iteration is completed according to the optimization result; When the iteration result satisfies the end constraint, an optimization parameter set is constructed according to the final temporary memory library, and the partition laser cladding control corresponding to the N surface partitions is executed through the optimization parameter set.

2. The surface treatment method of hydraulic support based on laser cladding according to claim 1, characterized in that: After the iteration reaches a preset period, the temporary memory bank is evaluated, an attention adjustment factor is constructed according to the evaluation result, the initial selection probability is optimized by the attention adjustment factor, and continuous iteration is completed according to the optimization result, further comprising: Perform iteration number analysis, establish the current iteration phase analysis, and construct iteration phase constraints; Obtaining the memory harmony in the current temporary memory bank, performing difference analysis on the memory harmony, and generating an initial attention adjustment factor according to the difference analysis result; The initial attention adjustment factor is optimized through the iterative phase constraint, the attention adjustment factor is constructed, and the initial selection probability optimization is performed based on the attention adjustment factor.

3. The surface treatment method of hydraulic support based on laser cladding according to claim 2 is characterized in that: The obtaining of the memory harmony in the current temporary memory bank and performing difference analysis on the memory harmony also includes: Perform fitness analysis based on memory harmony and construct an optimal solution set; Additional selection probabilities are configured for the preferred solution set, and when a subsequent iterative probability search is performed, if the probability selection result is the selection result of the temporary memory bank, the memory harmony selection in the temporary memory bank is performed by using the additional selection probability; The bandwidth is fine-tuned according to the memory harmony selection result, and the temporary memory library is updated based on the bandwidth fine-tuning result to complete this round of iteration.

4. The surface treatment method of hydraulic support based on laser cladding according to claim 1, characterized in that: The method uses the cladding response accuracy as the minimum clustering constraint, uses the structural data set and the functional data set to normalize and identify the position of the hydraulic support, performs partition clustering, and constructs N surface partitions, and further includes: A normalized ratio of configuration structure and function is configured, and a normalized position identification of the hydraulic support is completed based on the normalized ratio; Performing initial partitioning based on the structural data set and the functional data set, respectively, to establish k structural partitions and b functional partitions; The cluster centers are distributed in k structural partitions and b functional partitions respectively, and the position similarity analysis of structure and function is performed through position normalization identification to perform initial clustering; It is determined whether there is a clustering result in the initial clustering that does not meet the minimum clustering constraint. If not, the clustering search is continued, and N surface partitions are constructed according to the clustering search results.

5. The surface treatment method of hydraulic support based on laser cladding according to claim 4, characterized in that: The determining whether there is a clustering result in the initial clustering that does not satisfy the minimum clustering constraint further includes: If there are clustering results in the initial clustering that do not meet the minimum clustering constraint, an upward covering clustering instruction is generated; The clustering merging of the corresponding clustering result and the neighborhood is performed by the upward covering clustering instruction, wherein the upward covering clustering instruction is a constraint instruction for constraining the current clustering result to cluster with high surface quality; Construct N surface partitions based on the clustering merging results.

6. The surface treatment method of hydraulic support based on laser cladding according to claim 1, characterized in that: The creating of the first optimization space and the second optimization space also includes: After the first optimization space is created, the big data control parameters of the same type of equipment are collected based on the big data to establish a big data control parameter set; An overlap analysis is performed on the large data control parameter set and the first optimization space to obtain a non-overlapping control parameter set in the large data control parameter set, and the non-overlapping control parameter set is used as the second optimization space.

7. The surface treatment method of hydraulic support based on laser cladding according to claim 1, characterized in that: The method further comprises: Perform control tests by optimizing parameter sets and establish test results; Performing a control parameter fitness consistency evaluation in an optimization parameter set based on the test results; Update N fitness functions according to the consistency evaluation results, and perform fine-tuning and optimization of the optimization parameter set based on the N fitness functions; The partitioned laser cladding control of N surface partitions is carried out according to the fine-tuning optimization results.

8. The surface treatment system of hydraulic support based on laser cladding is characterized by: The system is used to implement the surface treatment method of the hydraulic support based on laser cladding according to any one of claims 1 to 7, and the system comprises: A data set construction module, wherein the data set construction module is used to acquire data of surface features of the hydraulic support, construct a structural data set, a functional data set, and configure the cladding response accuracy of the hydraulic support; A surface partition construction module, wherein the surface partition construction module is used to use the cladding response accuracy as the minimum clustering constraint, perform partition clustering after normalizing the position of the hydraulic support with the structural data set and the functional data set, and construct N surface partitions, wherein the N surface partitions are all regions composed of continuous regions; A performance index set establishment module, the performance index set establishment module is used to establish a performance index set, the performance index set includes a coating thickness uniformity index, a coating and substrate bonding strength index, a surface roughness index, and a heat-affected zone index, and the performance index set index weights are respectively configured through N surface partitions to construct N fitness functions corresponding to the N surface partitions; An optimization space creation module, wherein the optimization space creation module is used to create a first optimization space and a second optimization space, wherein the first optimization space is a data space constructed by collecting laser cladding processing data in a factory, and the second optimization space is a data space constructed based on laser cladding processing data of big data; A temporary memory library configuration module, the temporary memory library configuration module is used to configure a temporary memory library, the temporary memory library contains H groups of memory harmonics, and each group of memory harmonics represents a group of laser cladding parameters, including laser power, scanning speed, focusing distance, powder spraying rate, and spot parameters; An initial selection probability establishment module, the initial selection probability establishment module is used to establish the initial selection probabilities of the temporary memory bank, the first optimization space, and the second optimization space, and perform a probability search based on the initial selection probability, evaluate the probability search results based on the corresponding N fitness functions, and iterate the temporary memory bank; An attention adjustment factor construction module, wherein the attention adjustment factor construction module is used to evaluate the temporary memory bank after iterating to a preset period, construct an attention adjustment factor according to the evaluation result, optimize the initial selection probability through the attention adjustment factor, and complete continuous iteration according to the optimization result; The partitioned laser cladding control module is used to construct an optimization parameter set according to a final temporary memory library after the iteration result satisfies the end constraint, and execute the partitioned laser cladding control corresponding to N surface partitions through the optimization parameter set.

Images