Aluminum alloy electric arc additive manufacturing heat accumulation regulation and control method based on fuzzy game equilibrium

Abstract

The invention discloses an aluminum alloy electric arc additive manufacturing heat accumulation regulation and control method based on fuzzy game equilibrium. The method comprises the steps that S1, preliminary manufacturing process parameters are determined; s2, in the manufacturing process, micro-scale temperature fields of a molten pool in a micro-scale area and an adjacent area of the molten pool and a macro-scale temperature field of a macro-scale overall deposition area are collected in real time; s3, performing data integration on the micro-scale temperature field and the macro-scale temperature field to form unified multi-scale thermal feedback data; s4, constructing a hierarchical fuzzy game model based on the multi-scale thermal feedback data; and S5, performing dynamic optimization on the control variables and the target functions in the games at all levels by using the hierarchical fuzzy game model, and obtaining a feedback regulation and control strategy through game equilibrium analysis. According to the method, in-layer optimal output of process parameters is achieved, good mobility and process memorability are achieved through accumulation of manufacturing historical data and regulation and control records, and stable support is provided for thermal control quality of aluminum alloy component manufacturing.

Description

Technical Field

[0001] The present invention relates to the technical field of additive manufacturing, and particularly to a method for regulating heat accumulation in arc additive manufacturing of aluminum alloy based on fuzzy game equilibrium. Background Art

[0002] With the continuous development of additive manufacturing technology, aluminum alloy has been widely used in the fields of aerospace, automotive manufacturing and high-performance structural component manufacturing due to its excellent specific strength, thermal conductivity and corrosion resistance. As one of the important metal additive manufacturing paths, arc additive manufacturing technology uses a high-energy-density arc heat source to achieve layer-by-layer deposition forming of metal wire materials, which has the advantages of high forming efficiency, high material utilization rate and low equipment cost. However, due to the high thermal conductivity, large solidification shrinkage rate and extreme sensitivity to heat input of aluminum alloy itself, problems such as uneven heat accumulation, excessive temperature gradient and molten pool instability are extremely likely to occur during the arc additive manufacturing process, which in turn leads to grain coarsening, uneven microstructure, residual stress accumulation and even crack defects, severely restricting the popularization and application of this technology in the manufacturing of high-performance aluminum alloy components.

[0003] In the prior art, researchers generally adopt a constant or preset process parameter control strategy, such as fixed current and voltage, constant wire feeding speed and cooling flow rate, and manage the heat input through limited feedforward temperature control means. However, this type of method often ignores the dynamic coupling relationship between the micro-scale molten pool heat fluctuation and the macro-scale deposition layer heat distribution during the manufacturing process, lacks a response mechanism for multi-scale heat feedback, resulting in the failure to timely reflect the thermal field changes in parameter regulation, and being unable to effectively prevent local heat accumulation and the expansion of interlayer temperature difference. In addition, some studies attempt to introduce a closed-loop feedback control method to drive parameter correction through real-time temperature monitoring, but mostly based on a single control target or a linear regression model, which can only roughly adjust individual variables and is difficult to achieve high-precision control under complex conditions of multi-process parameter coordination and multi-objective conflict regulation.

[0004] Therefore, there is an urgent need for a heat regulation method with dynamic response ability, fuzzy reasoning mechanism and cross-scale cooperation strategy to effectively deal with the thermal instability problem in the complex manufacturing process and improve the quality stability and process robustness of arc additive manufacturing of aluminum alloy. Summary of the Invention

[0005] An object of the present invention is to propose a method for regulating heat accumulation in arc additive manufacturing of aluminum alloy based on fuzzy game equilibrium. The present invention not only realizes the intra-layer optimal output of process parameters, but also supports the inheritance of subsequent layer strategies and model updates through the accumulation of manufacturing historical data and regulation records, has good transferability and process memory, and provides stable support for the thermal control quality of aluminum alloy component manufacturing.

[0006] A thermal accumulation regulation method for aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to an embodiment of the present invention includes the following steps:

[0007] S1. Determine preliminary manufacturing process parameters according to the three-dimensional model of the manufactured workpiece;

[0008] S2. During the manufacturing process, collect the micro-scale temperature field in the molten pool and its adjacent area and the macro-scale temperature field in the overall deposition area in real time;

[0009] S3. Integrate the data of the micro-scale temperature field and the macro-scale temperature field to form unified multi-scale thermal feedback data. The micro-scale temperature field reflects local thermal dynamic changes, and the macro-scale temperature field reflects overall thermal dynamic changes;

[0010] S4. Construct a hierarchical fuzzy game model based on the multi-scale thermal feedback data. The hierarchical fuzzy game model includes a micro-scale thermal feedback sub-game model for fuzzy control of the local temperature dynamics in the molten pool and its adjacent area; and a macro-scale thermal feedback main game model for fuzzy regulation of the temperature distribution in the overall deposition area;

[0011] S5. Dynamically optimize the control variables and objective functions in each level of the game using the hierarchical fuzzy game model, and obtain a feedback regulation strategy through game equilibrium analysis. Adjust the preliminary manufacturing process parameters in real time according to the feedback regulation strategy to achieve a dynamic balance between heat input and heat dissipation.

[0012] Optionally, S1 includes the following steps:

[0013] S11. Obtain the three-dimensional model M of the manufactured workpiece 3D , the three-dimensional model M 3D includes the geometric dimensions, volume distribution, deposition layer structure and manufacturing direction information of the workpiece;

[0014] S12. Based on the three-dimensional model M 3D and the thermophysical properties of the aluminum alloy material, construct a preliminary manufacturing process parameter dataset D init , the preliminary manufacturing process parameter dataset D init includes arc current, arc voltage, wire feeding speed and cooling flow rate.

[0015] Optionally, S2 includes the following steps:

[0016] S21. During the aluminum alloy arc additive manufacturing process, perform local temperature monitoring on the molten pool and its adjacent area based on a thermocouple array and a near-field infrared thermal imaging device, and construct a micro-scale temperature field dataset T micro , the micro-scale temperature field dataset T micro is composed of micro-scale depth layers z (m)It is composed of temperature values at multiple spatial position points under [conditions], and is used to reflect the instantaneous thermal dynamic changes of the molten pool and its surrounding local areas;

[0017] S22. Based on the far-field infrared thermal imaging system and the distributed optical fiber temperature sensing network, conduct global temperature monitoring on the overall deposition area, and construct a macro-scale temperature field data set T macro , the macro-scale temperature field data set T macro is composed of temperature values at multiple spatial position points under the macro-scale depth layer z (M) and is used to reflect the overall thermal distribution state of the entire manufacturing area. And the macro-scale depth layer z (M) is different from the micro-scale depth layer z (m) .

[0018] Optionally, the S3 includes the following steps:

[0019] S31. Perform unified timestamp alignment processing on the micro-scale temperature field data set and the macro-scale temperature field data set obtained at each moment, and form a set of structured multi-scale temperature data pairs by combining the micro-scale temperature field and the macro-scale temperature field corresponding to the same time point, which is used to represent the local and overall thermal distribution states at that moment;

[0020] S32. Arrange the multi-scale temperature data pairs corresponding to each time point in chronological order, and construct a multi-scale thermal feedback time series matrix during the entire manufacturing process. The multi-scale thermal feedback time series matrix records the micro-scale temperature state information and the macro-scale temperature state information at all moments during the manufacturing process, and maintains time synchronization and consistency of spatial distribution;

[0021] S33. Perform normalization processing and scale consistency conversion on the multi-scale thermal feedback time series matrix to obtain a unified multi-scale thermal feedback data set F multi , and the unified multi-scale thermal feedback data set performs spatial dimension fusion and thermal intensity standardization processing on the micro-scale temperature information and the macro-scale temperature information.

[0022] Optionally, the S4 includes the following steps:

[0023] S41. Based on the unified multi-scale thermal feedback data set F multi , at each manufacturing time step k, extract the instantaneous temperature deviation of the molten pool area, the average temperature of the overall deposition area, and the temperature gradient between the upper and lower deposition layers, and construct a thermal feedback state vector:

[0024]

[0025] Among them, represents the actual temperature at the center of the molten pool during the manufacturing process of the k-th layer The microscale temperature difference from the aluminum alloy set reference temperature , represents the average temperature value of the macroscale temperature field in the overall deposition area of the k-th layer, and represents the temperature gradient between the current layer and the previous layer;

[0026] S42. Input the thermal feedback state vector s k into the microscale temperature control sub-game model and the macroscale thermal equilibrium main game model respectively, and output the microscale control quantity and the macroscale control quantity

[0027] S43. Construct the fuzzy membership function set μ j (·), impose a regulation intensity weight on the control variables based on the thermal feedback state vector, and define the fuzzy control objective functions for the microscale and macroscale:

[0028]

[0029] Among them, represents the microscale control cost function of the k-th layer, combined with the energy application cost under the local temperature control error, represents the macroscale thermal equilibrium cost function, reflecting the thermal equilibrium cost of the deposition layer path angle adjustment and the total cooling flow operation of global cooling;

[0030] S44. Construct the cross-scale coupling scheduler C sync , and the cross-scale coupling scheduler adjusts the output weight of the controller according to the microscale control quantity and the macroscale control quantity , and defines the scheduling rules:

[0031]

[0032] Among them, is the fuzzy modulation function, with the input being the macroscale temperature gradient between the current layer and the previous layer and the output being the adjustment coefficient for the microscale control quantity . When the interlayer thermal gradient is greater than the preset value, it means there is a risk of thermal accumulation or weld fracture. γ(·) increases local cooling or reduces heat input, is the fuzzy modulation function, with the input being the microscale temperature difference between the current molten pool temperature and the target temperature and the output being the adjustment coefficient for the macroscale control quantity . When the molten pool is overheated or fluctuates violently, η(·) increases the total cooling flow operation of global cooling or changes the deposition layer path angle to inhibit the out-of-control upward or lateral thermal diffusion;

[0033] S45. Integrate the micro-scale temperature control sub-game model, the macro-scale thermal equilibrium main game model, and the cross-scale coupling scheduler to construct the corresponding complete hierarchical fuzzy game control structure for the k-th layer. The complete hierarchical fuzzy game control structure takes the unified multi-scale thermal feedback state as the input. Through the micro-scale temperature control sub-game model, the process control parameters of the arc current, wire feeding speed, and local cooling flow rate are obtained. Through the macro-scale thermal equilibrium main game model, the deposition path angle and the global cooling total cooling flow rate adjustment strategy are obtained. The cross-scale coupling scheduler dynamically coordinates the output results of the two game models according to the current molten pool temperature deviation and the inter-layer temperature gradient. By jointly minimizing the micro-scale control cost function and the macro-scale control cost function, the optimal control strategy set in the manufacturing process of the k-th layer is obtained. The control strategy set includes the arc current, wire feeding speed, local cooling flow rate, deposition path angle adjustment amount, and global cooling total cooling flow rate, which are used to drive the real-time update of the process parameters of the manufacturing layer and realize the local-global integrated dynamic regulation of the thermal accumulation state in the aluminum alloy arc additive manufacturing process.

[0034] Optionally, the micro-scale temperature control sub-game model is constructed including micro-scale game participants. The game participants include the arc current I k regulator, the wire feeding speed v k controller, and the local cooling flow rate Q k regulator. The fuzzy control input is Output micro-scale control quantity

[0035] Optionally, the macro-scale thermal equilibrium main game model is constructed including macro-scale game participants. The game participants include the deposition layer path angle θ k scheduler and the global cooling total cooling flow rate The fuzzy control input is the average temperature value and the temperature gradient Output macro-scale control quantity

[0036] Optionally, the S5 includes the following steps:

[0037] S61. Based on the k-th layer hierarchical fuzzy game control structure Call the micro-scale control cost function in the micro-scale temperature control sub-game model and the macro-scale control cost function in the macro-scale thermal equilibrium main game model Construct a joint optimization model. The goal of the joint optimization model is to solve the optimal solution of the control variable set under the current thermal feedback state:

[0038]

[0039] S62. Using the fuzzy game equilibrium analysis method, the game participants are analyzed in the current thermal feedback state vector s k The response strategy space under the convergence calculation is carried out to obtain the equilibrium solution *Indicates the optimal control solution after optimization through game equilibrium analysis;

[0040] S63. Formulate a multi-condition feedback control strategy based on the numerical relationship of each parameter in the thermal feedback state vector, which is used for determining and triggering strategies for different thermal anomaly situations in the manufacturing process;

[0041] S64. Write the feedback control strategy into the control instruction set of the kth layer, drive the process control system to update the parameters, and store them in the manufacturing process database. Adjust the preliminary manufacturing process parameters in real time according to the feedback control strategy to achieve a dynamic balance between heat input and heat dissipation.

[0042] Optionally, the feedback control strategy is specifically:

[0043] If the microscale temperature difference And the temperature gradient That is, the microscale temperature deviation exceeds the set microscale temperature difference threshold But the temperature gradient between layers does not exceed the safe range G safe , then the following feedback control strategy is output: arc current I k Set to Reduce heat input; wire feed speed v k Set to Slow down the metal deposition rate; local cooling flow rate Q k Set to Optimize heat dissipation;

[0044] If the temperature gradient And micro-scale temperature difference That is, the interlayer temperature gradient exceeds the risk threshold G risk , and the molten pool temperature deviation is still within the allowable range, the following feedback control strategy is output: deposition path angle θ k Set to Avoid heat concentration areas; global cooling total cooling flow Set to Optimize overall cooling;

[0045] If the temperature gradient And micro-scale temperature difference That is, local and overall thermal anomalies occur simultaneously, and the following joint control strategy is output: linkage activation All controlled variables: arc current I k , wire feeding speed v k , local cooling flow Q k , deposition path angle θk 、 Total cooling flow rate of global cooling All are executed according to the optimal solution;

[0046] If the average temperature value That is, the average temperature of the current overall deposition area of the k-th layer exceeds the macroscopic average temperature threshold It indicates that there is a global temperature rise risk during the manufacturing process, and the following macroscopic control strategy is output: Total cooling flow rate of global cooling Is set to where ΔQ is the flow gain value automatically generated by the system according to the degree of temperature rise; the deposition path angle θ k Is set to Guide the heat input to transfer to the area with higher heat dissipation efficiency.

[0047] The beneficial effects of the present invention are as follows:

[0048] (1) The present invention introduces a hierarchical fuzzy game modeling mechanism driven by multi-scale thermal feedback, realizes the collaborative regulation and feedback closed-loop update of local-global thermal behavior. By constructing a unified multi-scale thermal feedback data set and establishing a hierarchical fuzzy game model including a micro-scale temperature control sub-game model and a macro-scale thermal equilibrium main-game model, it can capture and express the dynamic change law from micro-local thermal perturbation to overall temperature distribution in real time, significantly enhancing the system's response ability to thermal anomalies. In the model structure, different scales realize the mutual induction and adjustment of control quantities through a cross-scale coupling scheduler, thus establishing a multi-layer game feedback control closed-loop with self-adaptability.

[0049] (2) The present invention constructs a fuzzy control cost function driven by micro-scale temperature deviation and macro-scale temperature gradient, introduces a fuzzy membership function set in multi-variable regulation to perform weighted penalty and fuzzy evaluation on the regulation cost, effectively coordinates the mutual restraint relationship among parameters such as arc current, wire feeding speed and cooling flow rate. By constructing a joint optimization target and introducing a fuzzy game equilibrium solution solving strategy, the system can achieve a global optimal regulation point between production efficiency and thermal equilibrium quality, significantly improving the parameter coordination ability.

[0050] (3) The present invention formulates strategy output rules under three types of working conditions according to different thermal field states (local overheating, excessive interlayer thermal gradient or both occurring simultaneously) during the manufacturing process, combined with the real-time numerical values of key indicators in the thermal feedback state vector, and forms a feedback regulation strategy set to guide the dynamic adjustment of manufacturing parameters of the current layer. This mechanism not only realizes the intra-layer optimal output of process parameters, but also supports the inheritance of subsequent layer strategies and model updates through the accumulation of manufacturing history data and regulation records, with good transferability and process memory, providing stable support for the thermal control quality of aluminum alloy component manufacturing. Brief Description of the Drawings

[0051] The accompanying drawings are used to provide a further understanding of the present invention and form a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the accompanying drawings:

[0052] Figure 1 It is a flowchart of a method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium proposed by the present invention. Specific embodiments

[0053] Now, the present invention will be further described in detail with reference to the accompanying drawings. These drawings are all simplified schematic diagrams, only showing the basic structure of the present invention in a schematic way, so they only show the components related to the present invention.

[0054] Refer to Figure 1 , a method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium, includes the following steps:

[0055] S1. Determine the preliminary manufacturing process parameters according to the three-dimensional model of the manufactured workpiece;

[0056] S2. During the manufacturing process, collect the micro-scale temperature field in the molten pool and its adjacent area and the macro-scale temperature field in the overall deposition area of the macro-scale in real time;

[0057] S3. Integrate the data of the micro-scale temperature field and the macro-scale temperature field to form unified multi-scale thermal feedback data. The micro-scale temperature field reflects the local thermal dynamic changes, and the macro-scale temperature field reflects the overall thermal dynamic changes;

[0058] S4. Construct a hierarchical fuzzy game model based on the multi-scale thermal feedback data. The hierarchical fuzzy game model includes a micro-scale thermal feedback sub-game model for fuzzy control of the local temperature dynamics in the molten pool and its adjacent area; and a macro-scale thermal feedback main game model for fuzzy regulation of the temperature distribution in the overall deposition area;

[0059] S5. Dynamically optimize the control variables and objective functions in each level of the game using the hierarchical fuzzy game model, and obtain the feedback regulation strategy through game equilibrium analysis. Adjust the preliminary manufacturing process parameters in real time according to the feedback regulation strategy to achieve the dynamic balance between heat input and heat dissipation.

[0060] In this embodiment, S1 includes the following steps:

[0061] S11. Obtain the three-dimensional model M of the manufactured workpiece 3D , the three-dimensional model M 3D includes the geometric dimensions, volume distribution, deposition layer structure and manufacturing direction information of the workpiece;

[0062] S12. Based on the three-dimensional model M 3D and the thermophysical properties of the aluminum alloy material, construct a preliminary manufacturing process parameter dataset D init The preliminary manufacturing process parameter dataset D init includes arc current, arc voltage, wire feeding speed, and cooling flow rate.

[0063] In this embodiment, S2 includes the following steps:

[0064] S21. During the aluminum alloy arc additive manufacturing process, based on the thermocouple array and the near-field infrared thermal imaging device, perform local temperature monitoring on the molten pool and its adjacent areas to construct a micro-scale temperature field dataset T micro The micro-scale temperature field dataset T micro consists of temperature values at multiple spatial position points located under the micro-scale depth layer z (m) and is used to reflect the instantaneous thermal dynamic changes in the molten pool and its surrounding local areas;

[0065] S22. Based on the far-field infrared thermal imaging system and the distributed optical fiber temperature sensing network, perform global temperature monitoring on the overall deposition area to construct a macro-scale temperature field dataset T macro The macro-scale temperature field dataset T macro consists of temperature values at multiple spatial position points located under the macro-scale depth layer z (M) and is used to reflect the overall thermal distribution state of the entire manufacturing area. Moreover, the macro-scale depth layer z (M) is different from the micro-scale depth layer z (m)

[0066] In this embodiment, S3 includes the following steps:

[0067] S31. Perform unified timestamp alignment processing on the micro-scale temperature field dataset and the macro-scale temperature field dataset obtained at each moment, and form a set of structured multi-scale temperature data pairs by combining the micro-scale temperature field and the macro-scale temperature field corresponding to the same time point, which is used to represent the local and overall thermal distribution states at that moment;

[0068] S32. Arrange the multi-scale temperature data pairs corresponding to each time point in chronological order to construct a multi-scale thermal feedback time series matrix during the entire manufacturing process. The multi-scale thermal feedback time series matrix records the micro-scale temperature state information and the macro-scale temperature state information at all moments during the manufacturing process and maintains time synchronization and spatial distribution consistency;

[0069] S33. Perform normalization processing and scale consistency conversion on the multi-scale thermal feedback time series matrix to obtain a unified multi-scale thermal feedback dataset F multi ​, the unified multi-scale thermal feedback dataset fuses micro-scale temperature information and macro-scale temperature information in the spatial dimension and performs thermal intensity normalization.

[0070] In this embodiment, S4 includes the following steps:

[0071] S41. Based on the unified multi-scale thermal feedback dataset F multi , at each manufacturing time step k, extract the instantaneous temperature deviation of the molten pool area, the average temperature of the overall deposition area, and the temperature gradient between the upper and lower deposition layers, and construct a thermal feedback state vector:

[0072]

[0073] Among them, represents the actual temperature at the center of the molten pool during the manufacturing process of the k-th layer and the set reference temperature of the aluminum alloy The micro-scale temperature difference, represents the average temperature value of the macro-scale temperature field in the overall deposition area of the k-th layer, represents the temperature gradient between the current layer and the previous layer;

[0074] S42. Input the thermal feedback state vector s k into the micro-scale temperature control sub-game model and the macro-scale thermal equilibrium main game model respectively, and output the micro-scale control quantity and the macro-scale control quantity

[0075] S43. Construct a fuzzy membership function set μ j (·), based on the thermal feedback state vector, apply an adjustment intensity weight to the control variable, and define the fuzzy control objective function for the micro-scale and macro-scale:

[0076]

[0077] Among them, represents the micro-scale control cost function of the k-th layer, combined with the energy application cost under the local temperature control error, represents the macro-scale thermal equilibrium cost function, reflecting the thermal equilibrium cost of the deposition layer path angle adjustment and the total cooling flow operation of the global cooling;

[0078] S44. Construct a cross-scale coupling scheduler C sync , the cross-scale coupling scheduler adjusts the output weight of the controller according to the micro-scale control quantity and the macro-scale control quantity , and defines the scheduling rule:

[0079]

[0080] Among them, is the fuzzy modulation function, and the input is the macroscopic scale temperature gradient between the current layer and the previous layer The output is the adjustment coefficient for the micro-scale control quantity When the interlayer thermal gradient is greater than the preset value, it means that there is a risk of thermal accumulation or weld fracture. γ(·) increases local cooling or reduces heat input. Fuzzy modulation function, the input is the micro-scale temperature difference between the current molten pool temperature and the target temperature The output is the adjustment coefficient for the macroscopic control quantity When the molten pool is overheated or fluctuates violently, η(·) increases the total cooling flow rate operation of global cooling or changes the deposition layer path angle to inhibit the out-of-control upward or lateral thermal diffusion;

[0081] S45. Integrate the micro-scale temperature control sub-game model, the macroscopic thermal equilibrium main game model and the cross-scale coupling scheduler to construct the corresponding complete hierarchical fuzzy game control structure for the k-th layer The complete hierarchical fuzzy game control structure takes the unified multi-scale thermal feedback state as the input, obtains the process control parameters of the arc current, wire feeding speed and local cooling flow rate through the micro-scale temperature control sub-game model, obtains the deposition path angle and the global cooling total cooling flow rate adjustment strategy through the macroscopic thermal equilibrium main game model, and the cross-scale coupling scheduler dynamically coordinates the output results of the two game models according to the current molten pool temperature deviation and the interlayer temperature gradient. By jointly minimizing the micro-scale control cost function and the macroscopic scale control cost function, the optimal control strategy set in the manufacturing process of the k-th layer is obtained. The control strategy set includes the arc current, wire feeding speed, local cooling flow rate, deposition path angle adjustment amount and global cooling total cooling flow rate, which are used to drive the real-time update of the process parameters of the manufacturing layer and realize the local-global integrated dynamic regulation of the thermal accumulation state in the aluminum alloy arc additive manufacturing process.

[0082] In this embodiment, the micro-scale temperature control sub-game model The construction of includes micro-scale game participants. The game participants include the arc current I k Adjusting party, wire feeding speed v k Controlling party and local cooling flow rate Q k Adjusting party, and the fuzzy control input is Output micro-scale control quantity

[0083] In this embodiment, the macroscopic thermal equilibrium main game model The construction of includes macroscopic game participants. The game participants include the deposition layer path angle θ k Scheduling party and global cooling total cooling flow rate The input of the fuzzy control is the average temperature value and the temperature gradient Output the macroscopic control quantity

[0084] In this embodiment, S5 includes the following steps:

[0085] S61. Based on the hierarchical fuzzy game control structure of the k-th layer Call the microscale control cost function in the microscale temperature control sub-game model and the macroscopic control cost function in the macroscopic thermal equilibrium main game model Construct a joint optimization model, and the goal of the joint optimization model is to solve the optimal solution of the control variable set under the current thermal feedback state:

[0086]

[0087] S62. Use the fuzzy game equilibrium analysis method to calculate the convergence of the response strategy space of the game participants under the current thermal feedback state vector s k to obtain the equilibrium solution * represents the optimal control solution optimized through game equilibrium analysis;

[0088] S63. Based on the numerical relationship of each parameter in the thermal feedback state vector, formulate a multi-condition feedback control strategy for the determination and strategy triggering of different thermal anomaly situations in the manufacturing process;

[0089] S64. Write the feedback control strategy into the control instruction set of the k-th layer, drive the process control system to update the parameters, and store them in the manufacturing process database. According to the feedback control strategy, adjust the preliminary manufacturing process parameters in real time to achieve the dynamic balance between heat input and heat dissipation.

[0090] In this embodiment, the feedback control strategy is specifically as follows:

[0091] If the microscale temperature difference and the temperature gradient That is, the microscale temperature deviation exceeds the set microscale temperature difference threshold but the interlayer temperature gradient does not exceed the safe range G safe , then output the following feedback control strategy: the arc current I k is set to reduce the heat input; the wire feeding speed v k is set to slow down the metal deposition rate; the local cooling flow rate Q k is set to optimize the heat dissipation capacity;

[0092] If the temperature gradient And micro-scale temperature difference That is, the interlayer temperature gradient exceeds the risk threshold G risk , and the molten pool temperature deviation is still within the allowable range, the following feedback control strategy is output: deposition path angle θ k Set to Avoid heat concentration areas; global cooling total cooling flow Set to Optimize overall cooling;

[0093] If the temperature gradient And micro-scale temperature difference That is, local and overall thermal anomalies occur simultaneously, and the following joint control strategy is output: linkage activation All controlled variables: arc current I k , wire feeding speed v k , local cooling flow Q k , deposition path angle θ k , Global cooling total cooling flow All are executed according to the optimal solution;

[0094] If the average temperature That is, the average temperature of the entire deposition area of the current kth layer exceeds the macroscopic average temperature threshold This indicates that there is a global temperature rise risk in the manufacturing process, so the following macro control strategy is output: Global cooling total cooling flow Set to Where ΔQ is the flow gain value automatically generated by the system according to the degree of temperature rise; the deposition path angle θ k Set to Directs heat input toward areas where heat dissipation is more efficient.

[0095] Embodiment 1:

[0096] From October 2024 to December 2024, an aviation structural parts manufacturer conducted process research and thermal control optimization experiments based on arc additive manufacturing technology for a new generation of lightweight aluminum alloy frame beam components. The frame beam component is a large curved structure with a length of about 850mm and a wall thickness of 5mm. A total of 62 layers need to be deposited. The component is made of 5A06 aluminum alloy wire. This material is prone to local heat accumulation, interlayer grain coarsening and thermal cracks during the arc additive manufacturing process, which directly affects the mechanical properties and dimensional accuracy. It is a typical structure sensitive to thermal accumulation control.

[0097] Under traditional process conditions, the enterprise used a fixed-parameter additive path, setting the arc current at 135 A, the voltage at 21 V, the wire feeding speed at 2.8 mm / s, and the cooling air speed at 10 m / s. During continuous deposition, the system could not sense the evolution of the interlayer temperature, and only relied on preset cooling intervals and thermal imaging manual inspections to control the heat input. However, in multiple test batches, the temperature in the area from the 28th layer to the 40th layer of the component remained above 410 °C for a long time, far exceeding the recommended critical temperature of 385 °C for grain stability of aluminum alloy, resulting in microcracks and arching deformation near the 30th layer, seriously affecting the structural stability.

[0098] To solve this problem, the experimental team introduced the present invention for deployment testing on the actual production line. The system used a high-precision near-field thermocouple matrix (resolution 0.5 mm) to be embedded around the molten pool area, and combined with a FLIR T1040 infrared thermal imaging device to monitor the overall deposition area, collecting micro-scale temperature field and macro-scale temperature field data. When manufacturing each layer, the system obtained the molten pool temperature, average layer temperature, and interlayer temperature difference information in real time at a frequency of 2 Hz, forming a multi-scale thermal feedback data set.

[0099] During the manufacturing process, the system used the constructed micro-scale temperature control sub-game model to adjust the arc current, wire feeding speed, and local cooling. At the same time, it combined with the macro-scale main game model to dynamically optimize the deposition path angle and global cooling strategy, and used a cross-scale coupling scheduler to correct the weights of the control variables. In the critical heat accumulation area from the 29th layer to the 40th layer, the system automatically identified an abnormal state where the interlayer temperature gradient was greater than 3.5 K / mm, and immediately reduced the arc current from 135 A to 121 A, the wire feeding speed from 2.8 mm / s to 2.3 mm / s, while increasing the local cooling air speed to 13.5 m / s, and fine-tuning the deposition path angle to disperse the heat flow direction, controlling the trend of total heat accumulation.

[0100] Under the action of the thermal control strategy, the average temperature of the component from the 30th layer to the 40th layer was maintained at about 376 °C, effectively avoiding the phenomenon of local grain coarsening. The surface size deviation of the final product structure was controlled within ±0.18 mm, the maximum interlayer thermal gradient was 3.1 K / mm, significantly lower than 4.9 K / mm under the traditional process, and there were no crack defects in the X-ray inspection, meeting the aviation structure certification standards.

[0101] To verify the performance improvement of the method of the present invention, the project team compared 10 components manufactured by the traditional method with 10 components manufactured by the method of the present invention. The specific comparison data of the thermal field control accuracy and structural quality are shown in Table 1 below:

[0102] Table 1 Comparison of 10 components manufactured by the traditional method and 10 components manufactured by the method of the present invention

[0103] Project indicators Traditional method (average value) Method of the present invention (average value) Highest temperature of microscale molten pool (°C) 425 381 Maximum value of interlayer temperature gradient (K / mm) 4.9 3.1 Interlayer thermal equilibrium recovery time (s) 22.4 9.7 Maximum deviation of surface size (mm) ±0.48 ±0.18 Incidence rate of internal microcracks (%) 30％ 0％ Average grain size (μm) 18.7 13.2 System control delay response (s) No automatic response 1.8s Process adaptive parameter adjustment frequency (times / layer) Manual 1 time / 10 layers Automatic 1.6 times / layer Average forming efficiency (g / min) 3.9 3.6 Qualified rate (meeting all detection standards) 60％ 100％

[0104] In addition, after each layer of manufacturing, the system will output the control and write it into the historical database. By means of time-series feedback, the robustness of the control strategy for subsequent layers is improved, endowing the manufacturing process with learning ability and memory, and completing closed-loop intelligent control.

[0105] This embodiment 1 fully demonstrates the practical feasibility and superiority of the method of the present invention in solving the problems of out-of-control heat accumulation, excessive temperature gradient, and difficult coupling of multi-parameter regulation in aluminum alloy arc additive manufacturing. The experimental results show that the method can significantly improve the temperature control accuracy, reduce the risk of hot cracks, and improve the structural consistency and quality qualification rate without significantly sacrificing the forming efficiency, and is applicable to the complex thermal environment manufacturing tasks of high-performance aluminum alloy additive components.

[0106] The present invention introduces a hierarchical fuzzy game modeling mechanism driven by multi-scale thermal feedback to realize the collaborative regulation and feedback closed-loop update of local-global thermal behavior. By constructing a unified multi-scale thermal feedback data set and establishing a hierarchical fuzzy game model including a micro-scale temperature control sub-game model and a macro-scale thermal equilibrium main game model, the dynamic change law from micro-local thermal perturbation to overall temperature distribution can be captured and expressed in real time, significantly enhancing the system's response ability to thermal anomalies. In the model structure, the control quantities are mutually induced and adjusted between different scales through a cross-scale coupling scheduler, thus establishing a multi-layer game feedback control closed-loop with self-adaptability.

[0107] The present invention constructs a fuzzy control cost function driven by micro-scale temperature deviation and macro-scale temperature gradient, introduces a fuzzy membership function set in multi-variable regulation to perform weighted penalty and fuzzy evaluation on the regulation cost, effectively coordinates the mutual restraint relationship among parameters such as arc current, wire feeding speed, and cooling flow rate. By constructing a joint optimization target and introducing a fuzzy game equilibrium solution solving strategy, the system can achieve a global optimal regulation point between production efficiency and thermal equilibrium quality, significantly improving the parameter coordination ability.

[0108] The present invention formulates strategy output rules under three types of working conditions according to different thermal field states (local overheating, excessive interlayer thermal gradient, or both occurring simultaneously) during the manufacturing process, combined with the real-time numerical values of key indicators in the thermal feedback state vector, and forms a feedback regulation strategy set to guide the dynamic adjustment of the manufacturing parameters of the current layer. This mechanism not only realizes the intra-layer optimal output of process parameters, but also supports the inheritance and model update of subsequent layer strategies through the accumulation of manufacturing history data and regulation records, and has good transferability and process memory, providing stable support for the thermal control quality of aluminum alloy component manufacturing.

[0109] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, shall be covered by the protection scope of the present invention.

Claims

1. A thermal accumulation regulation method for aluminum alloy arc additive manufacturing based on fuzzy game equilibrium, characterized in that It includes the following steps: S1. Determine the preliminary manufacturing process parameters according to the three-dimensional model of the workpiece to be manufactured; S2. During the manufacturing process, collect in real time the micro-scale temperature field in the molten pool and its adjacent area and the macro-scale temperature field in the overall deposition area at the macro scale; S3. Integrate the data of the micro-scale temperature field and the macro-scale temperature field to form unified multi-scale thermal feedback data. The micro-scale temperature field reflects the local thermal dynamic changes, and the macro-scale temperature field reflects the overall thermal dynamic changes; S4. Construct a hierarchical fuzzy game model based on the multi-scale thermal feedback data. The hierarchical fuzzy game model includes a micro-scale thermal feedback sub-game model for fuzzy control of the local temperature dynamics in the molten pool and its adjacent area, and a macro-scale thermal feedback main game model for fuzzy regulation of the temperature distribution in the overall deposition area; S5. Dynamically optimize the control variables and objective functions in each level of the game using the hierarchical fuzzy game model, and obtain the feedback regulation strategy through game equilibrium analysis. Adjust the preliminary manufacturing process parameters in real time according to the feedback regulation strategy to achieve the dynamic balance between heat input and heat dissipation.

2. The method for regulating thermal accumulation in arc additive manufacturing of aluminum alloy based on fuzzy game equilibrium according to claim 1, characterized in that The S1 includes the following steps: S11. Obtain the three-dimensional model M of the manufactured workpiece 3D , 3D model M 3D Contains information on the workpiece’s geometric dimensions, volume distribution, deposited layer structure, and manufacturing direction; S12. Based on the three-dimensional model M 3D and the thermophysical properties of the aluminum alloy material, construct the preliminary manufacturing process parameter dataset D init , the preliminary manufacturing process parameter dataset D init includes arc current, arc voltage, wire feeding speed, and cooling flow rate.

3. A method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to claim 2, characterized in that, The S2 includes the following steps: S21. During the aluminum alloy arc additive manufacturing process, local temperature monitoring of the molten pool and its adjacent area is carried out based on a thermocouple array and a near-field infrared thermal imaging device, and a micro-scale temperature field data set T is constructed. micro , the micro-scale temperature field data set T micro consists of temperature values at multiple spatial position points located below the micro-scale depth layer z (m) and is used to reflect the instantaneous thermal dynamic changes in the molten pool and its surrounding local areas; S22. Based on the far-field infrared thermal imaging system and the distributed optical fiber temperature sensing network, the overall deposition area is globally monitored for temperature, and a macroscopic-scale temperature field dataset T is constructed. macro The macroscopic-scale temperature field dataset T macro consists of temperature values at multiple spatial position points located below the macroscopic-scale depth layer z (M) , and is used to reflect the overall thermal distribution state of the entire manufacturing area. Moreover, the macroscopic-scale depth layer z (M) is different from the micro-scale depth layer z (m) .

4. A method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to claim 1, characterized in that, The S3 includes the following steps: S31. Perform unified timestamp alignment processing on the micro-scale temperature field dataset and the macro-scale temperature field dataset obtained at each moment, and form a set of structured multi-scale temperature data pairs corresponding to the micro-scale temperature field and the macro-scale temperature field at the same time point to represent the local and overall thermal distribution states at that moment; S32. Arrange the multi-scale temperature data pairs corresponding to each time point in chronological order to construct a multi-scale thermal feedback time series matrix for the entire manufacturing process. The multi-scale thermal feedback time series matrix records the micro-scale temperature state information and the macro-scale temperature state information at all moments during the manufacturing process, and maintains time synchronization and spatial distribution consistency; S33. Normalize the multi-scale thermal feedback time series matrix and perform scale consistency conversion to obtain a unified multi-scale thermal feedback dataset F multi , and the unified multi-scale thermal feedback dataset performs spatial dimension fusion and thermal intensity standardization processing on micro-scale temperature information and macro-scale temperature information.

5. A method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to claim 4, characterized in that, The S4 includes the following steps: S41. Based on the unified multi-scale thermal feedback dataset F multi , at each manufacturing time step k, extract the instantaneous temperature deviation of the molten pool region, the average temperature of the overall deposition region, and the temperature gradient between the upper and lower deposition layers, and construct a thermal feedback state vector: Among them, represents the actual temperature at the center of the molten pool during the manufacturing process of the k-th layer and the microscale temperature difference from the set reference temperature of the aluminum alloy ; represents the average temperature value of the macroscopic temperature field in the overall deposition area of the k-th layer, and represents the temperature gradient between the current layer and the previous layer; S42. Input the thermal feedback state vector s k into the micro-scale temperature control sub-game model and the macro-scale thermal equilibrium main game model respectively, and output the micro-scale control quantity and the macro-scale control quantity S43. Construct a set of fuzzy membership functions μ j (·), impose a regulation intensity weight on the control variable based on the thermal feedback state vector, and define the fuzzy control objective functions for the microscale and the macroscale: Among them, represents the micro-scale control cost function of the k-th layer, combining the energy application cost under local temperature control error, represents the macro-scale thermal equilibrium cost function, reflecting the thermal equilibrium cost of the deposition layer path angle adjustment and the global cooling total cooling flow operation; S44. Construct a cross-scale coupling scheduler C sync , and the cross-scale coupling scheduler adjusts the output weight of the controller according to the micro-scale control quantity and the macro-scale control quantity to define the scheduling rule: Among them, is a fuzzy modulation function, and the input is the macroscopic scale temperature gradient between the current layer and the previous layer The output is the adjustment coefficient for the microscale control quantity When the interlayer thermal gradient is greater than the preset value, it means that there is a risk of thermal accumulation or weld fracture. γ(·) increases local cooling or reduces heat input Fuzzy modulation function, the input is the microscale temperature difference between the current molten pool temperature and the target temperature The output is the adjustment coefficient for the macroscopic control quantity When the molten pool is overheated or fluctuates violently, η(·) increases the total cooling flow rate of global cooling or changes the deposition layer path angle to inhibit the out-of-control upward or lateral heat diffusion; S45. Integrate the micro-scale temperature control sub-game model, the macro-scale thermal equilibrium main-game model, and the cross-scale coupling scheduler to construct the corresponding complete hierarchical fuzzy game control structure for the k-th layer.

6. A method for regulating thermal accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to claim 5, characterized in that The micro-scale temperature control sub-game model The construction of which includes micro-scale game participants, and the game participants include the arc current I k Regulating party, wire feeding speed v k Controlling party and local cooling flow rate Q k For the regulating party, the fuzzy control input is Output micro-scale control quantity 7. A method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to claim 5, characterized in that, The described macro thermal equilibrium main game model The construction of which includes macro game participants, and the game participants include the deposition layer path angle θ k The dispatcher and the total global cooling flow rate The fuzzy control input is the average temperature value And the temperature gradient Output the macro control quantity 8. A method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to claim 5, characterized in that The S5 includes the following steps: S61. Based on the k-th layer hierarchical fuzzy game control structure Respectively call the microscale control cost function in the microscale temperature control sub-game model And the macro control cost function in the macro thermal equilibrium main game model Construct a joint optimization model, and the goal of the joint optimization model is to solve the optimal solution of the control variable set under the current thermal feedback state: S62. Using the fuzzy game equilibrium analysis method to calculate the convergence of the response strategy space of game participants under the current thermal feedback state vector s k to obtain the equilibrium solution * represents the optimal control solution optimized through game equilibrium analysis; S63. Develop a multi-condition feedback regulation strategy based on the numerical relationships of the parameters in the thermal feedback state vector for the determination and strategy triggering of different thermal anomaly situations during the manufacturing process; S64. Write the feedback regulation strategy into the regulation instruction set of the k-th layer, drive the process control system to update the parameters, and store them in the manufacturing process database. Adjust the preliminary manufacturing process parameters in real time according to the feedback regulation strategy to achieve the dynamic balance between heat input and heat dissipation.

9. A method for regulating heat accumulation in aluminum alloy arc additive manufacturing based on fuzzy game equilibrium according to claim 8, characterized in that, The specific feedback regulation strategy is: If there is a micro-scale temperature difference and a temperature gradient that is, the micro-scale temperature deviation exceeds the set micro-scale temperature difference threshold but the interlayer temperature gradient does not exceed the safe range G safe , then the following feedback control strategy is output: the arc current I k is set to reduce the heat input; Wire feeding speed v k Set to Reduce the metal deposition rate; local cooling flow rate Q k Set to Optimize the heat dissipation capacity; If the temperature gradient and the microscale temperature difference i.e., the interlayer temperature gradient exceeds the risk threshold G risk , and the temperature deviation of the molten pool is still within the allowable range, then the following feedback control strategy is output: the deposition path angle θ k is set to avoid the heat concentration area; the total cooling flow rate of the global cooling is set to optimize the overall cooling; If the temperature gradient And micro-scale temperature difference That is, local and overall thermal anomalies occur simultaneously, and the following joint control strategy is output: linkage activation All controlled variables: arc current I k , wire feeding speed v k , local cooling flow Q k , deposition path angle θ k , Global cooling total cooling flow All are executed according to the optimal solution; If the average temperature value that is, the average temperature of the overall deposition area of the current k-th layer exceeds the macroscopic average temperature threshold it indicates that there is a global temperature rise risk during the manufacturing process, and the following macroscopic control strategy is output: the total cooling flow rate of global cooling is set to where ΔQ is the flow rate gain value automatically generated by the system according to the degree of temperature rise; the deposition path angle θ k is set to to guide the heat input to transfer to the area with higher heat dissipation efficiency.

Images