Vacuum brazing method for a needle array heat spreader

Abstract

The present application relates to the technical field of vacuum brazing, and discloses a vacuum brazing method for a needle array radiator, which comprises the following steps: fusing benchmark process parameters of a historical process knowledge base and multi-dimensional characteristic parameters of the needle array radiator to obtain brazing process characteristic parameters; performing process strategy matching on the needle array radiator to confirm a vacuum brazing strategy for the needle array radiator; applying the vacuum brazing strategy to drive a vacuum brazing furnace to perform phased process operations of preheating and degassing, filler melting and spreading, diffusion connection and controllable cooling; performing situation analysis on real-time process data flow of the phased process operations to obtain deviation data; performing dynamic parameter reconstruction on a target brazing stage of the needle array radiator to obtain temperature regulation parameters and process logic parameters; and performing cyclic brazing control on the needle array radiator to obtain a post-weld component of the needle array radiator; and the present application can improve the efficiency of the vacuum brazing method for the needle array radiator.

Description

Technical Field

[0001] This invention relates to the field of vacuum brazing technology, and in particular to a vacuum brazing method for a pin array heat sink. Background Technology

[0002] In the field of vacuum brazing of pin array heat sinks, existing processes often rely on fixed empirical parameters or single-dimensional operating condition information when determining brazing process parameters. They fail to effectively integrate benchmark data from historical process knowledge bases with the multi-dimensional characteristic parameters of the pin array heat sink, resulting in low compatibility between the determined process parameters and actual brazing conditions. This problem directly leads to process defects such as incomplete preheating and degassing, poor solder melting and spreading, and insufficient diffusion bonding. This not only reduces the brazed joint qualification rate but also significantly increases the defect rework rate, affecting the production quality and efficiency of pin array heat sinks.

[0003] Meanwhile, existing vacuum brazing processes lack a systematic capability for analyzing real-time data streams of staged process operations. This makes it impossible to accurately capture deviations in operating conditions and corresponding brazing risk categories, and also hinders the dynamic reconstruction of temperature control parameters and process logic parameters for the target brazing stage based on deviation data. The lag in process adjustments results in poor post-weld quality consistency for different batches and with different characteristics of pin array heat sinks, failing to meet the production requirements of high-precision heat dissipation devices and further restricting the large-scale application of vacuum brazing technology for pin array heat sinks. Therefore, improving the process adaptability, process controllability, and post-weld quality stability of vacuum brazing for pin array heat sinks has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a vacuum brazing method for a pin array heat sink to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a vacuum brazing method for a pin array heat sink, comprising:

[0006] S1. By integrating the baseline process parameters from the historical process knowledge base and the multi-dimensional characteristic parameters of the pin array heat sink, the brazing process characteristic parameters of the pin array heat sink are obtained.

[0007] S2. Based on the brazing process characteristic parameters, perform process strategy matching on the pin array heat sink to confirm the vacuum brazing strategy of the pin array heat sink.

[0008] S3. Apply the vacuum brazing strategy to drive the vacuum brazing furnace to perform the staged process operations of preheating and degassing, brazing filler metal melting and spreading, diffusion bonding and controllable cooling.

[0009] S4. Perform situational analysis on the real-time process data stream of the staged process operation to obtain the deviation data of the real-time process data stream;

[0010] S5. Based on the deviation data, the target brazing stage of the pin array heat sink is dynamically reconstructed to obtain the temperature control parameters and process logic parameters of the target brazing stage.

[0011] S6. Based on the temperature control parameters and the process logic parameters, perform cyclic brazing control on the pin array heat sink to obtain the welded component of the pin array heat sink.

[0012] In a preferred embodiment, the baseline process parameters from the fusion historical process knowledge base and the multidimensional characteristic parameters of the pin array heat sink are used to obtain the brazing process characteristic parameters of the pin array heat sink, including:

[0013] Retrieve baseline process parameters from the historical process knowledge base that match the material combination and structural type of the pin array heat sink;

[0014] The multidimensional feature parameters of the pin array heat sink are deconstructed to obtain the brazing condition state vector of the pin array heat sink.

[0015] Map the brazing condition state vector to the adjustable parameter node in the baseline process parameters;

[0016] Based on the brazing joint quality requirements of the pin array heat sink, process values ​​are assigned to the adjustable parameter nodes to obtain the brazing process characteristic parameters of the pin array heat sink.

[0017] In a preferred embodiment, the step of performing process strategy matching on the pin array heat sink based on the brazing process characteristic parameters to confirm the vacuum brazing strategy of the pin array heat sink includes:

[0018] Based on the brazing temperature requirement, vacuum requirement and heat input constraint in the brazing process characteristic parameters, candidate strategies for the pin array heat sink are selected.

[0019] A comprehensive impact assessment is performed on the candidate strategies to obtain the brazing condition adaptability of the candidate strategies.

[0020] Based on the brazing condition adaptability, the candidate strategies are optimized to obtain the target strategy for the pin array heat sink.

[0021] The brazing process characteristic parameters are filled into the target strategy to obtain the vacuum brazing strategy for the pin array heat sink.

[0022] In a preferred embodiment, the step of performing a comprehensive impact assessment on the candidate strategies to obtain the brazing condition suitability of the candidate strategies includes:

[0023] The historical brazing data of the candidate strategies are traced back, and the brazing joint pass rate and defect rework rate in the historical brazing data are statistically analyzed to obtain the historical performance baseline index of the candidate strategies.

[0024] The candidate strategies are subjected to real-time constraint quantization to obtain real-time context constraint indices for the candidate strategies.

[0025] Based on the preset brazing quality impact weights, the brazing condition adaptability of the candidate strategies is calculated by integrating the historical performance baseline index and the real-time situation constraint index.

[0026] In a preferred embodiment, the formula for calculating the brazing condition adaptability is as follows:

[0027] ;

[0028] In the formula, Indicates the first Brazing condition adaptability of each candidate strategy Indicates the first The first candidate strategy A historical performance baseline indicator, Indicates the first The first candidate strategy A real-time contextual constraint indicator Indicates the first Local weights of historical performance baseline indicators, Indicates the first Local weights of real-time context constraint indicators This represents the total number of historical performance baseline indicators. This represents the total number of real-time context constraint metrics. This represents the preset historical performance fusion coefficient. This represents the preset real-time scenario fusion coefficient. This indicates a summation operation.

[0029] In a preferred embodiment, the application of the vacuum brazing strategy to drive the vacuum brazing furnace to perform staged process operations including preheating and degassing, solder melting and spreading, diffusion bonding, and controlled cooling includes:

[0030] The vacuum brazing strategy is analyzed to extract the target control parameters and stage switching conditions of the pin array heat sink;

[0031] The target control parameters and the stage switching conditions are encoded into the brazing adjustment command for the pin array heat sink.

[0032] The brazing adjustment command is sent to the vacuum brazing furnace to drive the vacuum brazing furnace to sequentially perform the staged process operations of preheating and degassing the pin array heat sink, melting and spreading the brazing filler metal, diffusion connection and controlled cooling.

[0033] In a preferred embodiment, the step of performing situational analysis on the real-time process data stream of the staged process operation to obtain the deviation data of the real-time process data stream includes:

[0034] Spatiotemporal correlation reconstruction is performed on the real-time process data stream of the aforementioned staged process operations to obtain a dynamic process status diagram of the real-time process data stream;

[0035] Analyze the evolution deviation trend of the dynamic process situation diagram, and determine the abnormal mode of the evolution deviation trend to obtain the brazing risk category of the evolution deviation trend;

[0036] The deviation trend of the evolution is quantified to obtain the working condition deviation degree of the evolution deviation trend;

[0037] The deviation of the operating condition and the brazing risk category are integrated into the deviation data of the real-time process data stream.

[0038] In a preferred embodiment, the step of analyzing the evolutionary deviation trend of the dynamic process status diagram and determining the abnormal mode of the evolutionary deviation trend to obtain the brazing risk category of the evolutionary deviation trend includes:

[0039] The dynamic process situation diagram is deconstructed using multi-dimensional features to obtain the key trend feature vector of the dynamic process situation diagram.

[0040] Anomaly similarity analysis is performed on the key trend feature vectors to obtain the evolutionary deviation trend of the key trend feature vectors;

[0041] Based on the abnormal patterns in the historical process knowledge base, the matching degree of the evolutionary deviation trend is evaluated to determine the brazing risk category of the evolutionary deviation trend.

[0042] In a preferred embodiment, the step of dynamically reconstructing the target brazing stage of the pin array heat sink based on the deviation data to obtain the temperature control parameters and process logic parameters of the target brazing stage includes:

[0043] In the historical process knowledge base, the temperature adjustment mapping relationship and logical adaptation mapping relationship that match the deviation data are called;

[0044] Based on the temperature adjustment mapping relationship, the temperature parameters of the target brazing stage of the pin array heat sink are adapted to obtain the temperature control parameters of the target brazing stage.

[0045] Based on the logical adaptation mapping relationship, the process decision is modified for the target brazing stage of the pin array heat sink to obtain the process logic parameters of the target brazing stage.

[0046] In a preferred embodiment, the step of performing cyclic brazing control on the pin array heat sink based on the temperature control parameters and the process logic parameters to obtain the post-brazing component of the pin array heat sink includes:

[0047] The temperature control parameters and the process logic parameters are encoded to obtain the brazing optimization instructions for the pin array heat sink.

[0048] The brazing optimization command is sent to the vacuum brazing furnace to control the vacuum brazing furnace to perform the optimized brazing operation;

[0049] During the optimized brazing operation, the brazing quality of the pin array heat sink is continuously monitored. When the brazing quality meets the preset brazing product conditions, the post-welded component of the pin array heat sink is obtained.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. This invention integrates the baseline process parameters from a historical process knowledge base with the multi-dimensional characteristic parameters of the pin array heat sink, and combines them with the brazed joint quality requirements to complete the process assignment, accurately obtaining the brazing process characteristic parameters suitable for the current working conditions; then, based on these parameters, candidate strategies are screened and the brazing working condition suitability is calculated, the target strategy is selected and the parameters are filled to form a vacuum brazing strategy, effectively improving the matching degree between the process and the actual working conditions, reducing process deviations in preheating and degassing, brazing filler metal melting and spreading, improving the brazed joint qualification rate, and improving the overall efficiency of vacuum brazing.

[0052] 2. Furthermore, this invention performs situational analysis on the real-time data stream of staged process operations to accurately obtain deviation data; then, based on the historical knowledge base and mapping relationships, it dynamically reconstructs the parameters of the target brazing stage to obtain suitable temperature control and process logic parameters. Through cyclic brazing control and continuous quality monitoring, the controllability of the process is significantly improved, ensuring stable and consistent quality of the welded components. This meets the production requirements of high-precision pin array heat sinks and provides reliable technical support for their large-scale manufacturing. Attached Figure Description

[0053] Figure 1 A schematic flowchart of a vacuum brazing method for a pin array heat sink according to an embodiment of the present invention;

[0054] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0055] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0056] This application provides a vacuum brazing method for a pin array heat sink. The execution subject of this vacuum brazing method for a pin array heat sink includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the vacuum brazing method for a pin array heat sink can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0057] Reference Figure 1 The diagram shown is a schematic flowchart of a vacuum brazing method for a pin array heat sink according to an embodiment of the present invention. In this embodiment, the vacuum brazing method for the pin array heat sink includes:

[0058] S1. By integrating the baseline process parameters from the historical process knowledge base and the multi-dimensional characteristic parameters of the pin array heat sink, the brazing process characteristic parameters of the pin array heat sink are obtained.

[0059] In this embodiment of the invention, the reference process parameters from the fusion historical process knowledge base and the multi-dimensional feature parameters of the pin array heat sink are used to obtain the brazing process feature parameters of the pin array heat sink, including:

[0060] Retrieve baseline process parameters from the historical process knowledge base that match the material combination and structural type of the pin array heat sink;

[0061] The multidimensional feature parameters of the pin array heat sink are deconstructed to obtain the brazing condition state vector of the pin array heat sink.

[0062] Map the brazing condition state vector to the adjustable parameter node in the baseline process parameters;

[0063] Based on the brazing joint quality requirements of the pin array heat sink, process values ​​are assigned to the adjustable parameter nodes to obtain the brazing process characteristic parameters of the pin array heat sink.

[0064] First, clarify the material combination and structural type of the current pin array heat sink to be processed. The material combination refers to the specific combination of the base material and the solder material used in the heat sink. The structural type refers to the structural characteristics of the heat sink, such as the pin arrangement density, the ratio of pin height to base material thickness, and the cross-sectional shape of the pins. Then, retrieve all stored vacuum brazing process records of pin array heat sinks from the historical process knowledge base. Compare the material combination and structural type of the pin array heat sink recorded in each process record with the material combination and structural type of the current pin array heat sink to be processed. Select process records that are completely consistent with the current product to be processed. Extract the corresponding process parameters from these selected process records. These extracted process parameters are the benchmark process parameters that match the material combination and structural type of the pin array heat sink.

[0065] First, we need to clarify the specific contents of the multidimensional characteristic parameters of the pin array heat sink. These parameters typically cover material properties, structural geometric properties, and process requirement properties. Material properties include the thermal conductivity, coefficient of thermal expansion, and melting point of the substrate and solder. Structural geometric properties include the number of pins, pin diameter, and substrate planar dimensions. Process requirement properties include the expected heat dissipation efficiency after welding and the temperature range to withstand. Then, according to the fixed classification rule of "material properties → structural geometric properties → process requirement properties", these multidimensional characteristic parameters are broken down and classified. The specific parameters in each category are arranged in a preset order to form a parameter set containing all the classified parameters in a fixed order. This parameter set is the brazing condition state vector of the pin array heat sink.

[0066] First, from the acquired baseline process parameters, identify the parameters that can be adjusted according to the actual working conditions as adjustable parameter nodes. These nodes typically include preheating temperature, holding time, vacuum control value, cooling rate, etc. Then, analyze the correlation between each parameter in the brazing condition state vector and each adjustable parameter node. For example, the thermal conductivity parameter in the material properties is directly related to the adjustable preheating temperature, and the pin density parameter in the structural geometry is directly related to the adjustable holding time. Based on these correlations, map each parameter in the brazing condition state vector to the corresponding adjustable parameter node in the baseline process parameters, so that each adjustable parameter node can correspond to at least one related parameter in the brazing condition state vector, thus completing the mapping from the brazing condition state vector to the adjustable parameter node.

[0067] First, clarify the quality requirements for the brazed joints of the pin array heat sink. These requirements specifically include the tensile strength standard, airtightness level, and the criteria for judging whether the joint is free of cracks and pores. Then, for each adjustable parameter node that has been mapped, determine the specific value of the adjustable parameter node by combining its corresponding brazing condition state vector parameters and the brazed joint quality requirements. For example, if the brazed joint quality requirement is high tensile strength and the corresponding material has high thermal conductivity, then set the value of the preheating temperature adjustable point to a specific temperature that can fully preheat the material and avoid overheating damage. Complete the value setting of all adjustable parameter nodes in the same way, and integrate all the adjustable parameter nodes with set values ​​to form a complete parameter set. This parameter set is the brazing process characteristic parameter of the pin array heat sink.

[0068] The beneficial effect is that it can accurately obtain brazing process characteristic parameters that are fully adapted to the current operating conditions of the pin array heat sink through systematic steps. These parameters rely on effective experience data in the historical process knowledge base and fully combine the multi-dimensional characteristics and quality requirements of the product itself, ensuring the rationality and pertinence of the parameters. This provides a reliable basis for the formulation of subsequent vacuum brazing strategies, helps to reduce defects caused by improper parameters in subsequent process links, and ensures the stability of the brazing process and the quality of the final welded components.

[0069] S2. Based on the brazing process characteristic parameters, perform process strategy matching on the pin array heat sink to confirm the vacuum brazing strategy of the pin array heat sink.

[0070] In this embodiment of the invention, the step of performing process strategy matching on the pin array heat sink based on the brazing process characteristic parameters to confirm the vacuum brazing strategy of the pin array heat sink includes:

[0071] Based on the brazing temperature requirement, vacuum requirement and heat input constraint in the brazing process characteristic parameters, candidate strategies for the pin array heat sink are selected.

[0072] A comprehensive impact assessment is performed on the candidate strategies to obtain the brazing condition adaptability of the candidate strategies.

[0073] Based on the brazing condition adaptability, the candidate strategies are optimized to obtain the target strategy for the pin array heat sink.

[0074] The brazing process characteristic parameters are filled into the target strategy to obtain the vacuum brazing strategy for the pin array heat sink.

[0075] The historical brazing data of the candidate strategies are traced back, and the brazing joint pass rate and defect rework rate in the historical brazing data are statistically analyzed to obtain the historical performance baseline index of the candidate strategies.

[0076] The candidate strategies are subjected to real-time constraint quantization to obtain real-time context constraint indices for the candidate strategies.

[0077] Based on the preset brazing quality impact weights, the brazing condition adaptability of the candidate strategies is calculated by integrating the historical performance baseline index and the real-time situation constraint index.

[0078] The formula for calculating the brazing condition adaptability is as follows:

[0079] ;

[0080] In the formula, Indicates the first Brazing condition adaptability of each candidate strategy Indicates the first The first candidate strategy A historical performance baseline indicator, Indicates the first The first candidate strategy A real-time contextual constraint indicator Indicates the first Local weights of historical performance baseline indicators, Indicates the first Local weights of real-time context constraint indicators This represents the total number of historical performance baseline indicators. This represents the total number of real-time context constraint metrics. This represents the preset historical performance fusion coefficient. This represents the preset real-time scenario fusion coefficient. This indicates a summation operation.

[0081] First, clarify the specific values ​​of the brazing temperature requirement, vacuum requirement, and heat input constraint included in the brazing process characteristic parameters of the current pin array heat sink. Then, retrieve all stored vacuum brazing process strategies from the preset process strategy library, and check the brazing temperature range, vacuum range, and heat input range corresponding to each process strategy one by one. Select the process strategies whose temperature range includes the brazing temperature requirement, whose vacuum range includes the vacuum requirement, and whose heat input range includes the heat input constraint. These selected process strategies are the candidate strategies for the pin array heat sink.

[0082] First, retrieve the historical brazing data generated from the past applications of each candidate strategy. This data includes the total number of welds, the number of qualified brazed joints, and the number of defective joints requiring rework each time the candidate strategy is used for brazing the pin array heat sink. Then, perform statistical analysis on this data. Divide the number of qualified brazed joints by the total number of welds to obtain the brazed joint qualification rate, and divide the number of defective joints requiring rework by the total number of welds to obtain the defect rework rate. Integrate the obtained brazed joint qualification rate and defect rework rate to form the set of values, which is the historical performance baseline index of the candidate strategy.

[0083] First, determine the real-time constraints faced when applying the candidate strategy. These constraints include the maximum heating temperature that the vacuum brazing furnace can currently stably reach, the minimum vacuum level, and the stable output range of the heat input. Then, convert these real-time constraints into specific values. For example, record the maximum stable heating temperature of the furnace as a specific value in degrees Celsius, the minimum stable vacuum level as a specific value in Pascals, and the stable output range of the heat input as a specific value in watts. These converted specific values ​​are the real-time situational constraint indicators of the candidate strategy.

[0084] First, obtain the preset brazing quality impact weights. These weights include the local weights of historical performance baseline indicators and the local weights of real-time scenario constraint indicators, as well as preset global weight coefficients corresponding to the historical and real-time dimensions, namely the historical performance fusion coefficient and the real-time scenario fusion coefficient. The local weights of historical performance baseline indicators are used to distinguish the degree of impact of brazed joint pass rate and defect rework rate on quality, while the local weights of real-time scenario constraint indicators are used to distinguish the degree of impact of indicators such as furnace temperature, vacuum degree, and heat input on quality. During the calculation, each historical performance baseline indicator is first multiplied by its corresponding local weight, and all the multiplication results are summed to obtain a weighted sum of historical performance. This sum is then multiplied by the historical performance fusion coefficient to obtain the contribution value of the historical dimension to the fit. Next, each real-time scenario constraint indicator is multiplied by its corresponding local weight, and all the multiplication results are summed to obtain a weighted sum of real-time scenarios. This sum is then multiplied by the real-time scenario fusion coefficient to obtain the contribution value of the real-time dimension to the fit. Finally, the contribution values ​​of the historical dimension and the real-time dimension are added together to obtain the brazing condition fit of the candidate strategy.

[0085] First, compile a list of the brazing condition adaptability values ​​corresponding to all candidate strategies. Then, sort the candidate strategies in descending order of their values. Select the candidate strategy that is first in the sorted list, i.e., the one with the highest brazing condition adaptability value. This selected candidate strategy is the target strategy for the pin array heat sink.

[0086] First, identify the empty spaces in the target strategy that need to be filled with specific parameters. These empty spaces include the positions of parameters related to brazing temperature, vacuum degree, and heat input. Then, extract the corresponding specific values ​​from the brazing process characteristic parameters of the pin array heat sink. Fill the specific values ​​of the brazing temperature requirement into the empty space corresponding to "brazing temperature" in the target strategy, fill the specific values ​​of the vacuum degree requirement into the empty space corresponding to "vacuum degree", and fill the specific values ​​of the heat input constraint into the empty space corresponding to "heat input". After all parameter empty spaces are filled, the complete process strategy formed is the vacuum brazing strategy of the pin array heat sink.

[0087] The beneficial effects are as follows: by screening candidate strategies based on brazing process characteristic parameters, the initial strategy range is ensured to match the current product requirements; by backtracking historical data to obtain historical performance baseline indicators and quantifying real-time constraints to obtain real-time situation constraint indicators, and by merging and calculating the fit degree according to preset weights, the fit degree evaluation takes into account both historical experience and real-time operating conditions, resulting in more accurate results; by sorting and selecting the strategy with the highest fit degree as the target strategy, the optimality of the strategy is ensured; by filling in process characteristic parameters to form a vacuum brazing strategy, the strategy is made executable. The final vacuum brazing strategy can accurately adapt to the brazing requirements of the current pin array heat sink, providing a reliable basis for the subsequent stable execution of phased process operations and ensuring the quality of welded components.

[0088] S3. Apply the vacuum brazing strategy to drive the vacuum brazing furnace to perform the staged process operations of preheating and degassing, brazing filler metal melting and spreading, diffusion bonding and controllable cooling.

[0089] In this embodiment of the invention, the application of the vacuum brazing strategy to drive the vacuum brazing furnace to perform staged process operations including preheating and degassing, solder melting and spreading, diffusion bonding, and controlled cooling includes:

[0090] The vacuum brazing strategy is analyzed to extract the target control parameters and stage switching conditions of the pin array heat sink;

[0091] The target control parameters and the stage switching conditions are encoded into the brazing adjustment command for the pin array heat sink.

[0092] The brazing adjustment command is sent to the vacuum brazing furnace to drive the vacuum brazing furnace to sequentially perform the staged process operations of preheating and degassing the pin array heat sink, melting and spreading the brazing filler metal, diffusion connection and controlled cooling.

[0093] First, we analyze the structure of the vacuum brazing strategy. This strategy clearly outlines the operational requirements for each brazing stage of the pin array heat sink. From this, we identify the specific values ​​used to control the process state of each stage. These values ​​include the heating temperature and holding time of the preheating and degassing stage, the temperature range and vacuum level of the solder melting and spreading stage, the isothermal duration and pressure of the diffusion bonding stage, and the cooling rate of the controllable cooling stage. These values ​​together constitute the target control parameters of the pin array heat sink. Simultaneously, we extract the criteria for switching between stages from the strategy. For example, the preheating and degassing stage must reach the set temperature and maintain it for a specified time before switching to the solder melting and spreading stage. The solder melting and spreading stage must ensure that the solder completely covers the substrate surface before switching to the diffusion bonding stage. The diffusion bonding stage must meet the isothermal duration requirement before switching to the controllable cooling stage. These criteria constitute the stage switching conditions for the pin array heat sink.

[0094] First, determine the instruction format that the vacuum brazing furnace control system can recognize. This format includes a stage identifier field, a control parameter field, and a switching condition field. The stage identifier field is used to indicate the currently executed process stage, the control parameter field is used to fill in the specific control values ​​for each stage, and the switching condition field is used to describe the criteria for stage switching. Then, fill the extracted target control parameters into the control parameter field according to the corresponding stage. For example, fill the heating temperature and holding time of the preheating and degassing stage into the control parameter field corresponding to the "preheating and degassing" stage identifier, and fill the temperature range and vacuum holding value of the brazing filler metal melting and spreading stage into the control parameter field corresponding to the "brazing filler metal melting and spreading" stage identifier, and so on. At the same time, fill the switching conditions of each stage into the corresponding stage switching condition field. Finally, integrate the instruction content of all stages into a complete instruction set according to the process execution order. This instruction set is the brazing adjustment instruction for the pin array radiator.

[0095] Through a dedicated data transmission line supporting the vacuum brazing furnace, the prepared brazing adjustment instructions are sent to the main control system of the furnace body. After receiving the instructions, the main control system immediately activates the instruction parsing module to interpret the stage sequence, target control parameters for each stage, and stage switching conditions in the instructions. First, according to the instruction content of the "preheating and degassing" stage, the heating device of the furnace body is activated to gradually raise the furnace temperature to the heating temperature in the target control parameters of this stage, and the furnace temperature is monitored in real time through a temperature sensor. When the temperature reaches the set value, the timing module is activated to start calculating the holding time until the holding time requirement of this stage is met, completing the preheating and degassing operation. Subsequently, based on the determination of the stage switching conditions, the main control system automatically invokes the instructions of the "solder melting and spreading" stage, adjusts the heating device to make the furnace temperature reach the temperature range of this stage, and at the same time controls the vacuum system to maintain the target vacuum degree. After confirming through the observation window or internal camera that the solder is completely melted and evenly spread on the surface of the pin array radiator substrate, this stage of operation is completed. Then, enter the "diffusion bonding" stage according to the stage switching conditions, control the furnace body to maintain the target temperature and pressure values of this stage, and continuously meet the constant temperature duration requirement to achieve the diffusion bonding of the substrate and the solder. Finally, execute the instructions of the "controlled cooling" stage, control the cooling system to gradually reduce the furnace temperature at the target cooling rate until the furnace temperature drops to room temperature, and finally complete the staged process operations of preheating and degassing, solder melting and spreading, diffusion bonding, and controlled cooling of the pin array radiator in sequence.

[0096] The beneficial effects are as follows: By analyzing the vacuum brazing strategy, the target control parameters and stage switching conditions are accurately extracted to ensure that the basis for subsequent process operations completely matches the brazing requirements of the pin array radiator, avoiding parameter omission or deviation; Encoding the parameters and switching conditions into brazing adjustment instructions to make the instructions conform to the identification standards of the vacuum brazing furnace, ensuring the accuracy of information transmission and preventing operation errors caused by instruction format problems; Sending the instructions to the furnace body and driving it to execute each stage of operation in sequence, ensuring that each process link is strictly carried out according to the preset requirements, effectively improving the standardization and stability of the pin array radiator brazing process, and providing a reliable guarantee for obtaining a post-weld component with qualified quality and performance.

[0097] S4. Conduct a situation analysis on the real-time process data stream of the staged process operations to obtain the deviation data of the real-time process data stream;

[0098] In the embodiment of the present invention, the conducting a situation analysis on the real-time process data stream of the staged process operations to obtain the deviation data of the real-time process data stream includes:

[0099] Conduct a spatio-temporal correlation reconstruction on the real-time process data stream of the staged process operations to obtain the dynamic process situation diagram of the real-time process data stream;

[0100] Analyze the evolution deviation trend of the dynamic process situation diagram, and determine the abnormal mode of the evolution deviation trend to obtain the brazing risk category of the evolution deviation trend;

[0101] The deviation trend of the evolution is quantified to obtain the working condition deviation degree of the evolution deviation trend;

[0102] The deviation of the operating condition and the brazing risk category are integrated into the deviation data of the real-time process data stream.

[0103] The analysis examines the evolutionary deviation trend of the dynamic process status diagram and determines the abnormal mode of the evolutionary deviation trend to obtain the brazing risk category of the evolutionary deviation trend, including:

[0104] The dynamic process situation diagram is deconstructed using multi-dimensional features to obtain the key trend feature vector of the dynamic process situation diagram.

[0105] Anomaly similarity analysis is performed on the key trend feature vectors to obtain the evolutionary deviation trend of the key trend feature vectors;

[0106] Based on the abnormal patterns in the historical process knowledge base, the matching degree of the evolutionary deviation trend is evaluated to determine the brazing risk category of the evolutionary deviation trend.

[0107] First, real-time process data streams generated during each stage of the process operation are collected. This data includes temperature, vacuum, and pressure values ​​collected at different time points by temperature sensors, vacuum sensors, and pressure sensors at the top, middle, and bottom of the vacuum brazing furnace, as well as the specific start and end times of each stage. These data are then organized chronologically, and the corresponding parameter values ​​from different monitoring points at the same time point are correlated and integrated. Then, with time as the horizontal axis and each process parameter as the vertical axis, curves showing the changes of each parameter over time are plotted on a chart for each of the four stages. This creates a chart that dynamically reflects the spatiotemporal changes of process parameters at each stage; this chart is the dynamic process status diagram of the real-time process data stream.

[0108] Key dimensions affecting the stability of the brazing process are extracted from the dynamic process status diagram. These dimensions include the slope of temperature change in each stage (the numerical value of temperature increase or decrease per minute), the fluctuation range of vacuum (the difference between the maximum and minimum vacuum values ​​in a certain stage), the stability of pressure (whether the pressure value remains within a preset range in a certain stage), and the difference between the actual duration of each stage and the preset duration (the number of minutes by which the actual preheating and degassing time is longer or shorter than the preset time). Each dimension's characteristics are transformed into specific descriptive data. For example, the slope of temperature change is represented as "increase of 8 degrees Celsius per minute," and the fluctuation range of vacuum is represented as "fluctuating within the range of 10 Pascals." These descriptive data are then arranged in a fixed order of "temperature characteristics → vacuum characteristics → pressure characteristics → duration characteristics" to form an ordered feature set. This feature set is the key trend feature vector of the dynamic process status diagram.

[0109] The key trend feature vector corresponding to a normal brazing condition that is completely consistent with the current pin array radiator in terms of material combination and structural type is retrieved from the historical process knowledge base and used as the standard feature vector. The current key trend feature vector is compared with the standard feature vector dimension by dimension, and the difference in each dimension is calculated. For example, if the current temperature change slope is 8 degrees Celsius per minute, while the standard feature vector shows a temperature change slope of 5 degrees Celsius per minute for the same stage, the difference is 3 degrees Celsius per minute; if the current vacuum fluctuation amplitude is 10 Pascals, while the standard feature vector shows a vacuum fluctuation amplitude of 5 Pascals for the same stage, the difference is 5 Pascals. Based on the direction and specific values ​​of the differences in all dimensions, the overall direction of change of the current key trend feature vector and the degree of deviation from the standard vector are determined. This direction of change and the degree of deviation constitute the evolutionary deviation trend of the key trend feature vector.

[0110] All stored anomaly patterns are extracted from the historical process knowledge base. Each anomaly pattern corresponds to a specific evolutionary deviation trend and a clear brazing risk category. For example, "a sudden increase in temperature change slope that is consistently more than 50% higher than the standard" corresponds to "brazing filler metal overheating risk," "vacuum fluctuations exceeding the standard range and showing a continuous downward trend" corresponds to "furnace leakage risk," and "the actual duration of the diffusion bonding stage is more than 30% shorter than the preset duration" corresponds to "insufficient diffusion bonding risk." The current evolutionary deviation trend is compared with the characteristics of each anomaly pattern one by one to determine which anomaly pattern has the highest degree of overlap with the current deviation trend. For example, if the current deviation trend is "temperature change slope increases by 3 degrees Celsius per minute, exceeding the standard by 60% and continuing to rise," and it has the highest degree of overlap with the anomaly pattern corresponding to "brazing filler metal overheating risk," then the "brazing filler metal overheating risk" corresponding to this anomaly pattern is the brazing risk category of the evolutionary deviation trend.

[0111] Quantitative standards are set for deviations in each dimension. For example, the normal range for temperature change slope is set to 3 to 6 degrees Celsius per minute. If the actual slope is 9 degrees Celsius per minute, exceeding the standard upper limit by 3 degrees Celsius, the deviation percentage is calculated by subtracting the standard upper limit from the actual value, dividing by the standard upper limit, and then multiplying by 100%. Similarly, the normal range for vacuum fluctuation amplitude is set to 3 to 8 Pascals. If the actual fluctuation amplitude is 13 Pascals, exceeding the standard upper limit by 5 Pascals, the deviation percentage is calculated in the same way. Weights are preset based on the degree of influence of each dimension on brazing quality, such as a weight of 0.4 for temperature, 0.3 for vacuum, 0.2 for pressure, and 0.1 for duration. The deviation percentage for each dimension is multiplied by its corresponding weight, and then all weighted deviation percentages are summed. This sum represents the operating condition deviation degree of the evolving deviation trend.

[0112] The obtained operating condition deviation is combined with the brazing risk category to form structured data that includes both the specific risk type and the corresponding degree of deviation. This structured data is the deviation data of the real-time process data stream.

[0113] The beneficial effects are as follows: A dynamic process situation map is formed through spatiotemporal correlation reconstruction, which can intuitively and comprehensively present the spatiotemporal changes of process parameters at each stage, providing a clear and accurate foundation for subsequent situation analysis; through multi-dimensional feature deconstruction and anomaly similarity analysis, the evolutionary deviation trend of process parameters can be accurately captured, avoiding the omission of key deviation information; by combining historical anomaly pattern matching to determine the brazing risk category, the risk nature corresponding to the deviation can be clearly identified; the deviation trend is quantified to obtain the operating condition deviation degree, which can accurately grasp the severity of the deviation; finally, the integrated deviation data can comprehensively and accurately reflect the deviation status of the real-time process, providing a targeted and data-supported basis for the dynamic parameter reconstruction of the subsequent target brazing stage, ensuring the effectiveness of subsequent process adjustments, and thus improving the controllability of the brazing process of the pin array heat sink and the post-weld quality.

[0114] S5. Based on the deviation data, the target brazing stage of the pin array heat sink is dynamically reconstructed to obtain the temperature control parameters and process logic parameters of the target brazing stage.

[0115] In this embodiment of the invention, the step of dynamically reconstructing the target brazing stage of the pin array heat sink based on the deviation data to obtain the temperature control parameters and process logic parameters of the target brazing stage includes:

[0116] In the historical process knowledge base, the temperature adjustment mapping relationship and logical adaptation mapping relationship that match the deviation data are called;

[0117] Based on the temperature adjustment mapping relationship, the temperature parameters of the target brazing stage of the pin array heat sink are adapted to obtain the temperature control parameters of the target brazing stage.

[0118] Based on the logical adaptation mapping relationship, the process decision is modified for the target brazing stage of the pin array heat sink to obtain the process logic parameters of the target brazing stage.

[0119] First, clarify the operating condition deviation and brazing risk category included in the current deviation data. For example, if the operating condition deviation is 38.75% and the brazing risk category is brazing filler metal overheating risk, then enter the mapping relationship storage module of the historical process knowledge base. Each temperature adjustment mapping relationship and logical adaptation mapping relationship in this module is associated with specific deviation characteristics, which consist of a fixed brazing risk category and a corresponding operating condition deviation range. Compare the deviation characteristics of each mapping relationship within the module with the current deviation data. When the brazing risk category of a temperature adjustment mapping relationship is completely consistent with the brazing risk category of the current deviation data, and the current operating condition deviation falls within the deviation range set by that mapping relationship, then that mapping relationship is determined to be a matching temperature adjustment mapping relationship. Using the same comparison logic, find logical adaptation mapping relationships that match the brazing risk category of the current deviation data and whose operating condition deviation meets the range requirements. This completes the operation of calling the temperature adjustment mapping relationship and logical adaptation mapping relationship that matches the deviation data from the historical process knowledge base.

[0120] First, the matched temperature adjustment mapping relationship is analyzed. This mapping relationship clearly records the adjustment direction and magnitude of the target brazing stage temperature under the corresponding deviation characteristics. Then, the current initial temperature parameters of the target brazing stage are obtained, such as an initial heating temperature of 300 degrees Celsius and a holding time of 20 minutes in the preheating and degassing stage. The temperature adjustment amount is calculated according to the mapping relationship. For example, if the current deviation is 38.75%, the mapping relationship stipulates that for every 10% deviation in this range, the temperature should be reduced by 3 degrees Celsius. The calculated total adjustment amount is a reduction of 11.625 degrees Celsius. Subtracting the adjustment amount from the initial heating temperature yields 288.375 degrees Celsius. At the same time, according to the requirement in the mapping relationship that "the holding time is extended by 5 minutes when the deviation exceeds 30%", the initial holding time is adjusted to 25 minutes. The adjusted heating temperature and holding time are integrated to form the set of parameters that constitute the temperature control parameters for the target brazing stage.

[0121] First, the matched logical adaptation mapping relationship is decomposed. This mapping relationship contains the adjustment rules for the process decisions of the target brazing stage under the corresponding deviation characteristics. For example, the rule corresponding to the risk of brazing filler metal overheating is "shorten the high temperature duration and increase the temperature monitoring frequency to once every 2 minutes". Then, the current initial process decision of the target brazing stage is retrieved, such as the high temperature duration of 15 minutes and the temperature monitoring frequency of once every 5 minutes. The initial process decision is modified according to the adjustment rules, shortening the high temperature duration from 15 minutes to 10 minutes and adjusting the temperature monitoring frequency from once every 5 minutes to once every 2 minutes. At the same time, it is verified that the modified process decision does not conflict with the process connection requirements of the target brazing stage. The process decision content formed after the modification is completed is the process logic parameter of the target brazing stage.

[0122] The beneficial effects are that by retrieving temperature adjustment mapping relationships and logical adaptation mapping relationships from the historical process knowledge base and accurately matching deviation data, reliable basis for parameter reconstruction in the target brazing stage is provided based on past effective experience, avoiding blind adjustments without data support. Adapting temperature parameters according to the temperature adjustment mapping relationship ensures that the temperature control parameters in the target stage accurately correspond to the current deviation situation, effectively mitigating specific risks such as brazing filler metal overheating. Correcting process decisions based on the logical adaptation mapping relationship ensures that the process logic parameters align with the actual needs of the target stage, guaranteeing smooth process flow. The final temperature control parameters and process logic parameters accurately adapt to the deviation problems in the target brazing stage, providing scientific and feasible parameter support for subsequent cyclic brazing control, ensuring stable progress of the brazing process, and contributing to obtaining high-quality brazed components for the pin array heat sink.

[0123] S6. Based on the temperature control parameters and the process logic parameters, perform cyclic brazing control on the pin array heat sink to obtain the welded component of the pin array heat sink.

[0124] In this embodiment of the invention, based on the temperature control parameters and the process logic parameters, the pin array heat sink is subjected to cyclic brazing control to obtain the post-brazing component of the pin array heat sink, including:

[0125] The temperature control parameters and the process logic parameters are encoded to obtain the brazing optimization instructions for the pin array heat sink.

[0126] The brazing optimization command is sent to the vacuum brazing furnace to control the vacuum brazing furnace to perform the optimized brazing operation;

[0127] During the optimized brazing operation, the brazing quality of the pin array heat sink is continuously monitored. When the brazing quality meets the preset brazing product conditions, the post-welded component of the pin array heat sink is obtained.

[0128] First, clarify the specific content of the temperature control parameters, such as the target brazing stage heating temperature of 288.375 degrees Celsius and the holding time of 25 minutes. Then, clarify the specific content of the process logic parameters, such as the temperature monitoring frequency every 2 minutes and the stage switching judgment criteria: switching can only proceed if the holding time meets the standard and the temperature fluctuation does not exceed 2 degrees Celsius. Next, determine the instruction coding format that the vacuum brazing furnace control system can recognize. This format is divided into parameter coding segments and logic coding segments. The parameter coding segment is used to input the specific values ​​of the temperature control parameters, and the logic coding segment is used to input the operating rules of the process logic parameters. Fill the heating temperature and holding time into the corresponding positions in the parameter coding segment. Convert the temperature monitoring frequency and stage switching judgment criteria into machine-recognizable instruction statements according to the preset format and fill them into the logic coding segment. Finally, integrate the parameter coding segment and the logic coding segment into a complete instruction text. This instruction text is the brazing optimization instruction for the pin array heat sink.

[0129] The optimized brazing instructions are sent to the main control unit of the vacuum brazing furnace via the wired data transmission channel. Upon receiving the instructions, the main control unit immediately activates the instruction decoding module to interpret the parameter and logic encoding segments, extracting key information such as heating temperature, holding time, monitoring frequency, and switching criteria. Based on the interpretation results, the main control unit then drives the furnace's heating device, timing device, and monitoring device to work collaboratively. For example, it activates the heating device at a heating temperature of 288.375 degrees Celsius, activates the timing device for a holding time of 25 minutes, and controls the temperature sensor to collect data every 2 minutes. It strictly adheres to the stage switching criteria in executing each step of the operation, thereby controlling the vacuum brazing furnace to perform the optimized brazing operation.

[0130] During the optimized brazing operation, the built-in temperature sensor of the vacuum brazing furnace collects real-time temperature data, the vacuum sensor collects real-time vacuum data, and the appearance of the brazed joints of the pin array radiator is observed through the furnace body's observation window to check for cracks and porosity, forming multi-dimensional brazing quality monitoring data. Simultaneously, preset brazing product conditions are retrieved, which clearly stipulate that the joint appearance should be free of cracks and porosity, the heating temperature fluctuation range should not exceed ±2 degrees Celsius, and the vacuum level should not be lower than the preset standard value of 5 × 10⁻⁶. -3 The Pascal value and holding time error do not exceed 1 minute. Real-time monitoring data is continuously compared with preset conditions. When all monitoring data meet the preset conditions, the main control unit issues a stop operation command, shuts down the heating device and vacuum system, and removes the pin array radiator after the furnace temperature drops to room temperature. The resulting complete component is the welded component of the pin array radiator.

[0131] The beneficial effects are as follows: by encoding temperature control parameters and process logic parameters into brazing optimization instructions, the parameter information is accurately converted into executable operation instructions for the vacuum brazing furnace, avoiding operational errors caused by information transmission deviations; issuing instructions to the furnace body and controlling it to execute the optimized brazing operation ensures that the entire brazing process strictly follows the adapted parameters and logic, improving operational standardization and process stability; continuous monitoring of brazing quality during operation and comparison with preset product conditions can promptly identify quality anomalies and ensure that the final welded components meet standard requirements. The overall process effectively improves the quality stability and production reliability of pin array radiator brazing, providing a guarantee of high-quality components for subsequent product applications.

[0132] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0133] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A vacuum brazing method for a pin array heat sink, characterized in that, The method includes: S1. By integrating the baseline process parameters from the historical process knowledge base and the multi-dimensional characteristic parameters of the pin array heat sink, the brazing process characteristic parameters of the pin array heat sink are obtained. S2. Based on the brazing process characteristic parameters, perform process strategy matching on the pin array heat sink to confirm the vacuum brazing strategy of the pin array heat sink, including: Based on the brazing temperature requirement, vacuum requirement and heat input constraint in the brazing process characteristic parameters, candidate strategies for the pin array heat sink are selected. A comprehensive impact assessment is performed on the candidate strategies to obtain the brazing condition adaptability of the candidate strategies. Based on the brazing condition adaptability, the candidate strategies are optimized to obtain the target strategy for the pin array heat sink. The brazing process characteristic parameters are filled into the target strategy to obtain the vacuum brazing strategy of the pin array heat sink. S3. Apply the vacuum brazing strategy to drive the vacuum brazing furnace to perform the staged process operations of preheating and degassing, brazing filler metal melting and spreading, diffusion bonding and controllable cooling. S4. Perform situational analysis on the real-time process data stream of the staged process operation to obtain the deviation data of the real-time process data stream; S5. Based on the deviation data, the target brazing stage of the pin array heat sink is dynamically reconstructed to obtain the temperature control parameters and process logic parameters of the target brazing stage. S6. Based on the temperature control parameters and the process logic parameters, perform cyclic brazing control on the pin array heat sink to obtain the welded component of the pin array heat sink.

2. The vacuum brazing method for a pin array heat sink as described in claim 1, characterized in that, The baseline process parameters from the integrated historical process knowledge base and the multi-dimensional characteristic parameters of the pin array heat sink are used to obtain the brazing process characteristic parameters of the pin array heat sink, including: Retrieve baseline process parameters from the historical process knowledge base that match the material combination and structural type of the pin array heat sink; The multidimensional feature parameters of the pin array heat sink are deconstructed to obtain the brazing condition state vector of the pin array heat sink. Map the brazing condition state vector to the adjustable parameter node in the baseline process parameters; Based on the brazing joint quality requirements of the pin array heat sink, process values ​​are assigned to the adjustable parameter nodes to obtain the brazing process characteristic parameters of the pin array heat sink.

3. The vacuum brazing method for a pin array heat sink as described in claim 1, characterized in that, The comprehensive impact assessment of the candidate strategies to obtain the brazing condition adaptability of the candidate strategies includes: The historical brazing data of the candidate strategies are traced back, and the brazing joint pass rate and defect rework rate in the historical brazing data are statistically analyzed to obtain the historical performance baseline index of the candidate strategies. The candidate strategies are subjected to real-time constraint quantization to obtain real-time context constraint indices for the candidate strategies. Based on the preset brazing quality impact weights, the brazing condition adaptability of the candidate strategies is calculated by integrating the historical performance baseline index and the real-time situation constraint index.

4. The vacuum brazing method for a pin array heat sink as described in claim 3, characterized in that, The formula for calculating the brazing condition adaptability is as follows: ； In the formula, Indicates the first Brazing condition adaptability of each candidate strategy Indicates the first The first candidate strategy A historical performance baseline indicator, Indicates the first The first candidate strategy A real-time contextual constraint indicator Indicates the first Local weights of historical performance baseline indicators, Indicates the first Local weights of real-time context constraint indicators This represents the total number of historical performance baseline indicators. This represents the total number of real-time context constraint metrics. This represents the preset historical performance fusion coefficient. This represents the preset real-time scenario fusion coefficient. This indicates a summation operation.

5. The vacuum brazing method for a pin array heat sink as described in claim 1, characterized in that, The aforementioned vacuum brazing strategy drives the vacuum brazing furnace to perform staged process operations including preheating and degassing, solder melting and spreading, diffusion bonding, and controlled cooling, including: The vacuum brazing strategy is analyzed to extract the target control parameters and stage switching conditions of the pin array heat sink; The target control parameters and the stage switching conditions are encoded into the brazing adjustment command for the pin array heat sink. The brazing adjustment command is sent to the vacuum brazing furnace to drive the vacuum brazing furnace to sequentially perform the staged process operations of preheating and degassing the pin array heat sink, melting and spreading the brazing filler metal, diffusion connection and controlled cooling.

6. The vacuum brazing method for a pin array heat sink as described in claim 1, characterized in that, The process status analysis of the real-time process data stream of the staged process operation to obtain the deviation data of the real-time process data stream includes: Spatiotemporal correlation reconstruction is performed on the real-time process data stream of the aforementioned staged process operations to obtain a dynamic process status diagram of the real-time process data stream; Analyze the evolution deviation trend of the dynamic process situation diagram, and determine the abnormal mode of the evolution deviation trend to obtain the brazing risk category of the evolution deviation trend; The deviation trend of the evolution is quantified to obtain the working condition deviation degree of the evolution deviation trend; The deviation of the operating condition and the brazing risk category are integrated into the deviation data of the real-time process data stream.

7. The vacuum brazing method for a pin array heat sink as described in claim 6, characterized in that, The analysis examines the evolutionary deviation trend of the dynamic process status diagram and determines the abnormal mode of the evolutionary deviation trend to obtain the brazing risk category of the evolutionary deviation trend, including: The dynamic process situation diagram is deconstructed using multi-dimensional features to obtain the key trend feature vector of the dynamic process situation diagram. Anomaly similarity analysis is performed on the key trend feature vectors to obtain the evolutionary deviation trend of the key trend feature vectors; Based on the abnormal patterns in the historical process knowledge base, the matching degree of the evolutionary deviation trend is evaluated to determine the brazing risk category of the evolutionary deviation trend.

8. The vacuum brazing method for a pin array heat sink as described in claim 1, characterized in that, Based on the deviation data, the target brazing stage of the pin array heat sink is dynamically reconstructed to obtain the temperature control parameters and process logic parameters for the target brazing stage, including: In the historical process knowledge base, the temperature adjustment mapping relationship and logical adaptation mapping relationship that match the deviation data are called; Based on the temperature adjustment mapping relationship, the temperature parameters of the target brazing stage of the pin array heat sink are adapted to obtain the temperature control parameters of the target brazing stage. Based on the logical adaptation mapping relationship, the process decision is modified for the target brazing stage of the pin array heat sink to obtain the process logic parameters of the target brazing stage.

9. The vacuum brazing method for a pin array heat sink as described in claim 1, characterized in that, The process involves cyclically brazing the pin array heat sink based on the temperature control parameters and the process logic parameters to obtain the brazed component of the pin array heat sink, including: The temperature control parameters and the process logic parameters are encoded into the brazing optimization instructions for the pin array heat sink. The brazing optimization command is sent to the vacuum brazing furnace to control the vacuum brazing furnace to perform the optimized brazing operation; During the optimized brazing operation, the brazing quality of the pin array heat sink is continuously monitored. When the brazing quality meets the preset brazing product conditions, the post-welded component of the pin array heat sink is obtained.

Images