Additive manufacturing powder bed preheating system and method based on hot air-zoned radiation synergy

Abstract

This invention discloses a powder bed preheating system and method for additive manufacturing based on hot air-zoned radiation synergy, including a preheating chamber, a powder bed, a hot air circulation system, a zoned radiation heating system, and a layered collaborative controller. The beneficial effects of this invention are: it adopts an "upper-layer steady-state optimization + lower-layer dynamic compensation" architecture: the upper layer optimizes the hot air inlet temperature and velocity with energy efficiency and temperature uniformity as objectives; the lower layer, based on real-time temperature feedback, adjusts the zoned radiation power through a "feedback + feedforward" decoupling algorithm to compensate for local temperature disturbances, integrating the speed of hot air flow with the precision of zoned radiation. This solves the problems of response lag, spatial coupling, and lack of laser preheating synergy, achieving rapid, uniform, and dynamic control of the powder bed temperature field, improving the quality of additive manufacturing, and is suitable for high-precision manufacturing of various polymer materials.

Description

Technical Field

[0001] This invention relates to a powder bed preheating system, specifically an additive manufacturing powder bed preheating system and method based on hot air-zoned radiation synergy, belonging to the field of additive manufacturing technology. Background Technology

[0002] Additive manufacturing technology, as one of the core technologies of the Fourth Industrial Revolution, plays an increasingly important role in high-end fields such as aerospace, biomedicine, and automotive manufacturing. Among them, laser powder bed melting (LPBF) technology, with its high precision and ability to manufacture complex structures, has become the mainstream process for additive manufacturing of polymer materials. In these processes, the precise control of the powder bed temperature field is a key factor determining the quality, mechanical properties, and surface finish of the formed parts.

[0003] Currently, powder bed preheating technology faces three interrelated core technical challenges:

[0004] Challenge 1: Dynamic Response Lag. The thermal conductivity of polymer powder materials is typically in the range of 0.1-0.3 W / (m·K), which is 2-3 orders of magnitude lower than that of metallic materials. Existing infrared radiation preheating systems mainly rely on heat radiation transfer and contact conduction between powder particles, resulting in a thermal response time constant as high as 5-10 minutes. In the dynamic process of rapid laser scanning (scanning speed 1-5 m / s) and layer-by-layer processing (layer thickness 0.1-0.2 mm), this lag response cannot compensate for local temperature fluctuations in a timely manner, leading to temperature field runaway.

[0005] Challenge Two: Spatial Coupling Control Complexity. When forced convection (such as inert gas protection) is introduced into the system, the airflow field and temperature field exhibit a strong coupling effect. Heat transfer along the flow path of hot air results in upstream and downstream temperature differences of up to 10-20°C, forming a systemic spatial temperature gradient. Traditional zoned independent control strategies ignore this physical coupling, leading to mutual interference between control loops and making it difficult to achieve global temperature uniformity.

[0006] Challenge 3: Lack of coordinated laser-preheating control. In actual printing processes, the temperature in the laser sintering zone can reach above the material's melting point (200-400℃), while the preheating zone needs to be maintained below the melting point (150-250℃). Current technology lacks a real-time coordination mechanism between laser position and preheating control, leading to extreme high-temperature data from the laser heat-affected zone contaminating the preheating control feedback signal, severely impacting control accuracy and stability.

[0007] The technological development trends and challenges are as follows: With the widespread application of high-performance polymers (such as PEEK and PEI) and biodegradable materials (such as PLA and PCL) in additive manufacturing, more stringent requirements have been placed on temperature field control. These materials have narrow thermal decomposition temperature windows (typically ±5℃) and are extremely sensitive to temperature fluctuations. Existing preheating technologies have become a key bottleneck restricting their industrial application.

[0008] Therefore, there is an urgent need to develop a new generation of powder bed preheating control technology with rapid response, spatial decoupling, and real-time collaborative capabilities. Summary of the Invention

[0009] The purpose of this invention is to provide an additive manufacturing powder bed preheating system and method based on hot air-zoned radiation synergy in order to solve at least one of the above-mentioned technical problems.

[0010] The present invention achieves the above objectives through the following technical solution: an additive manufacturing powder bed preheating system based on hot air-zoned radiation synergy, comprising a preheating outer box and a powder bed, wherein the powder bed is disposed at the bottom of the preheating outer box, and the preheating outer box is provided with a hot air circulation system and a zoned radiation heating system, wherein the hot air circulation system heats the powder bed with hot air in a horizontal direction, and the zoned radiation heating system heats the powder bed with hot air in a vertical direction;

[0011] A layered coordinating controller is fixedly connected to the upper end of the preheating outer box. The layered coordinating controller includes an upper controller and a lower controller. The upper controller of the layered coordinating controller determines the basic operating point of the hot air circulation system based on the preheating temperature of the powder bed and the preset optimization target. The lower controller of the layered coordinating controller dynamically adjusts the power of each heating zone in the zoned radiation heating system based on the preheating temperature of the powder bed and the preset optimization target.

[0012] As a further embodiment of the present invention: the hot air circulation system includes an air duct and a temperature sensor. The two sides of the preheating outer box are respectively provided with a hot air inlet and a hot air outlet. An air inlet hood is sealed to the outside of the preheating outer box where the hot air inlet is located, and an air outlet hood is sealed to the outside of the preheating outer box where the hot air outlet is located. The air duct is located on the outside of the preheating outer box, and both ends of the air duct are connected to the air inlet hood and the air outlet hood, respectively. A fan and a heater are installed on the air duct. The temperature sensor is fixedly connected to the inner wall of the preheating outer box on the side where the hot air inlet is located, and the fan, heater and temperature sensor are all electrically connected to the layered collaborative controller.

[0013] As a further embodiment of the present invention: the partitioned radiation heating system includes thermal radiation lamps and infrared thermal imagers fixedly connected to the inner wall of the top of the preheating outer box. The thermal radiation lamps are arranged in three groups at equal intervals. The thermal radiation lamps and infrared thermal imagers are both located directly above the powder bed, and both the thermal radiation lamps and infrared thermal imagers are electrically connected to the layered collaborative controller.

[0014] An additive manufacturing powder bed preheating method based on hot air-zoned radiation synergy includes an additive manufacturing powder bed preheating system, and the preheating method includes the following steps:

[0015] Step 1: Upper Layer Preheating Optimization: Based on the set target preheating temperature T ref The basic operating point of the hot air circulation system is determined by the upper-level controller, including the hot air inlet temperature T. air,in Wind speed v air ;

[0016] Step 2, Lower-level dynamic compensation: The lower-level controller executes a real-time compensation cycle, calculates and updates the power of each heating zone based on the real-time temperature feedback from the infrared thermal imager, so as to accurately track the target preheating temperature.

[0017] As a further aspect of the present invention: In step one, the basic working point is obtained by solving a multi-objective optimization problem, the objective function of which is:

[0018]

[0019] in, The total power consumption of the system is Let V be the steady-state temperature variance.

[0020] As a further aspect of the present invention: In step two, for any non-uppermost heating zone j, its compensated heating power P j The calculation of (t) includes: a feedback control term P based on the temperature deviation of the local area. feedback,j This is used to compensate for the combined thermal effects, including multi-lamp radiative coupling; and a feedforward control term P based on the temperature deviation of at least one upstream heating zone i. feedforward,j It is used to decouple the upstream and downstream thermal effects caused by the cooling of hot air along the process.

[0021] As a further aspect of the present invention: the temperature of heating zone i located deviates from the target value ΔT i At that time, according to the heat transfer equation along the hot air path, the air temperature change upon reaching downstream zone j is:

[0022]

[0023] To maintain a constant temperature in downstream section j, the required compensation power is:

[0024]

[0025] Among them: A j η is the effective control area of ​​partition j. lamp Let be the radiation-convective heat transfer efficiency of the lamp tube; therefore, the theoretical expression for the decoupling compensation coefficient is:

[0026]

[0027] The formula shows that the compensation coefficient decreases exponentially with the distance between intervals, which is completely consistent with the physical law of heat transfer along the hot air path, and provides theoretical guidance for experimental calibration.

[0028] As a further aspect of the present invention: feedback control term P feedback,j An improved PID controller is used, featuring anti-integral saturation and derivative-first functions:

[0029]

[0030] Wherein, the error is defined as: e j (t)=T ref -T j (t); PID parameters adopt a piecewise tuning strategy: Start-up phase (0-60s): K p,j =1.2, K i,j =0.02, K d,j =0.08, ensuring rapid response; Steady-state phase (>60s): K p,j =0.8, K i,j =0.05, K d,j =0.15, ensuring precise control.

[0031] As a further aspect of the present invention: feedforward control term (P) feedforward,j The formula for calculating ) is:

[0032]

[0033] Among them, T i (t) represents the real-time temperature of upstream partition i, C ij To determine the steady-state decoupling compensation coefficients obtained through system identification experiments, C′ ij C′ is the dynamic compensation coefficient. ij =0.1·C ij ·τ air , τ air This is the delay time for hot air transmission.

[0034] The beneficial effects of this invention are:

[0035] 1) This invention effectively solves the core problems in existing additive manufacturing powder bed preheating technology, such as dynamic response lag, complex spatial coupling control, and lack of laser-preheating synergy, through the innovative dual-mode preheating concept of "hot air-zoned radiation synergy" and the hierarchical control architecture of "upper layer steady-state optimization + lower layer dynamic compensation".

[0036] 2) This invention breaks through the limitations of traditional single heat source preheating, introduces forced convection to greatly improve the system response speed, can quickly compensate for local temperature fluctuations and avoid temperature field runaway; secondly, through the hierarchical control architecture, it achieves decoupling of global energy efficiency optimization and local fast response, solves the "curse of dimensionality" problem in multi-heat source collaborative control, and transforms hot air cooling along the process from a harmful disturbance into a controllable process. Through the feedforward compensation algorithm, it achieves precise decoupling of upstream and downstream thermal effects, and significantly improves the uniformity of spatial temperature.

[0037] 3) This invention innovatively introduces dynamic ROI technology, combined with laser position synchronization and dynamic masking technology, to avoid interference from the high-temperature laser zone on preheating control, achieving true synergy between the laser and the preheating system, ensuring control accuracy and stability. Based on mature industrial devices and standard communication protocols, it has low technical risk; it has a wide range of applications, suitable for various polymer materials, especially for high-precision manufacturing of heat-sensitive materials; it has a high degree of standardization, with parameterized control algorithm design, enabling rapid adaptation to different equipment and process requirements, facilitating technology promotion; it has significant economic benefits, reducing scrap rate and energy consumption, shortening the investment payback period, providing strong technical support for high-end additive manufacturing applications, and promoting the further application of additive manufacturing technology in high-end fields such as aerospace, biomedicine, and automotive manufacturing. Attached Figure Description

[0038] Figure 1 This is a front view schematic diagram of the hardware structure of the powder bed preheating system of the present invention;

[0039] Figure 2 This is a top view schematic diagram of the hardware structure of the powder bed preheating system of the present invention;

[0040] Figure 3 This is a schematic diagram of the preheating method of the present invention.

[0041] In the diagram: 1. Preheating outer box; 11. Hot air inlet; 12. Hot air outlet; 13. Air inlet hood; 14. Air outlet hood; 2. Powder bed; 3. Air duct; 31. Fan; 32. Heater; 4. Thermal radiation lamp; 5. Infrared thermal imager; 6. Temperature sensor; 7. Layered collaborative controller. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Example 1, as Figures 1 to 2 As shown, an additive manufacturing powder bed preheating system based on hot air-zoned radiation synergy includes a preheating outer box 1 and a powder bed 2. The powder bed 2 is set at the bottom of the preheating outer box 1. The preheating outer box 1 is equipped with a hot air circulation system and a zoned radiation heating system. The hot air circulation system heats the powder bed 2 with hot air in the horizontal direction, and the zoned radiation heating system heats the powder bed 2 with hot air in the vertical direction.

[0044] A layered coordinating controller 7 is fixedly connected to the upper end of the preheating outer box 1. The layered coordinating controller 7 includes an upper controller and a lower controller. The upper controller of the layered coordinating controller 7 determines the basic operating point of the hot air circulation system according to the preheating temperature of the powder bed 2 and the preset optimization target. The lower controller of the layered coordinating controller 7 dynamically adjusts the power of each heating zone in the zoned radiation heating system according to the preheating temperature of the powder bed 2 and the preset optimization target.

[0045] Example 2, in addition to all the technical features included in Example 1, also includes:

[0046] The hot air circulation system includes an air duct 3 and a temperature sensor 6. The two sides of the preheating outer box 1 are respectively provided with a hot air inlet 11 and a hot air outlet 12. An air inlet hood 13 is sealed to the outside of the preheating outer box 1 where the hot air inlet 11 is provided, and an air outlet hood 14 is sealed to the outside of the preheating outer box 1 where the hot air outlet 12 is provided. The air duct 3 is located on the outside of the preheating outer box 1, and both ends of the air duct 3 are connected to the air inlet hood 13 and the air outlet hood 14, respectively. A fan 31 and a heater 32 are installed on the body of the air duct 3. The temperature sensor 6 is fixedly connected to the inner wall of the preheating outer box 1 on the side where the hot air inlet 11 is provided, and the fan 31, the heater 32 and the temperature sensor 6 are all electrically connected to the layered collaborative controller 7.

[0047] The zoned radiant heating system includes thermal radiation lamps 4 and infrared thermal imagers 5, which are fixedly connected to the inner wall of the top of the preheating outer box 1. The thermal radiation lamps 4 are arranged in three groups at equal intervals. Both the thermal radiation lamps 4 and the infrared thermal imagers 5 are located directly above the powder bed 2, and both the thermal radiation lamps 4 and the infrared thermal imagers 5 are electrically connected to the layered co-controller 7.

[0048] It should be noted that the layered collaborative controller 7 uses Advantech's IPC-610 series industrial PC. This choice is based on the following engineering considerations: Environmental adaptability: The industrial PC has an operating temperature range of -20℃ to 60℃ and vibration resistance, making it suitable for additive manufacturing workshop environments; Real-time performance guarantee: It supports real-time operating system expansion (such as RTX or QNX) to ensure deterministic execution of control tasks; Expandability: It provides multiple PCIe slots to support subsequent functional expansion.

[0049] An NIPCIe-6321 data acquisition card is installed in the hierarchical collaborative controller 7. This card provides 16 analog outputs (±10V, 16-bit resolution) and 32 digital I / Os, which is sufficient to support the needs of multi-zone control. An 8-channel solid-state power regulation module is also installed, with each channel output power ranging from 0-5kW and a response time of <10ms, which meets the requirements of fast power regulation.

[0050] Fan 31 is selected with a rated air volume of 500m³ / h. 3 An industrial centrifugal fan with a capacity of / h is used in conjunction with a 3kW PTC ceramic heater. The PTC heater's self-limiting temperature characteristic (maximum temperature 250℃) provides inherent safety protection. The air duct 3 is a Φ100mm high-temperature resistant stainless steel duct with a polished inner wall to reduce flow resistance. Flexible corrugated pipes are used at key connections to absorb thermal expansion.

[0051] The thermal radiation lamp 4 uses quartz infrared heating lamps. Three sets of quartz infrared heating lamps are installed sequentially at a height of 300mm above the powder bed 2, along a preset airflow direction (set as the positive X-axis), forming upstream, midstream, and downstream zones. Each set of quartz infrared heating lamps contains four 1kW lamps, using a 2-series, 2-parallel electrical connection method, ensuring both power density and improved reliability. The angle between the lamps and the powder bed surface is designed to be 45° to optimize the uniformity of radiation energy distribution.

[0052] The infrared thermal imager used is an FLIRA655sc (640×480 pixels, 50Hz frame rate, temperature measurement accuracy ±2℃), mounted directly above the powder bed surface at a distance of 800mm. This distance ensures a 120mm×90mm field of view, completely covering the effective processing area. A protective window is added to the front of the camera to prevent dust from contaminating the lens.

[0053] A real-time industrial Ethernet based on the EtherCAT protocol is established between the hierarchical collaborative controller 7 and the main control system of the additive manufacturing equipment. The communication cycle is set to 1ms, and the synchronization accuracy of the laser position data reaches ±0.1mm, which meets the accuracy requirements of dynamic ROI control.

[0054] Example 3, as Figure 3As shown, a preheating method for additive manufacturing powder bed using hot air-zoned radiation synergy includes a preheating system, and the preheating method includes the following steps:

[0055] S1. After hardware integration is completed, system integration testing is performed to verify the independent functions and collaborative working capabilities of each subsystem. At this point, the system possesses basic data acquisition, power output, and communication capabilities, providing a reliable experimental platform for subsequent system analysis.

[0056] S2. In-depth analysis and mathematical modeling of the system's thermophysical properties. Based on the hardware platform built in S1, the goal of this step is to gain a deep understanding of the system's thermophysical behavior, establish an accurate mathematical model, and provide a theoretical foundation for control algorithm design. This step is a crucial transition from "hardware construction" to "control design."

[0057] S21. The establishment of the system energy balance equation: Based on the basic principles of heat transfer, the transient energy balance equation for any infinitesimal element (x, y) on the powder bed surface is as follows:

[0058]

[0059] Among them, the convective heat transfer term: q″ conv =h(x,y)·[T air [(x,y,t)-T(x,y,t)];

[0060] Radiative heat absorption term:

[0061] Conductive heat dissipation items:

[0062] Radiant heat dissipation items:

[0063] This equation forms the theoretical basis for the design of the control system of this invention, where each term corresponds to a specific physical process and a controllable variable.

[0064] S22. Engineering analysis of key heat transfer mechanisms: Based on the energy balance equation, three core mechanisms affecting control performance were identified:

[0065] Mechanism 1: The revolutionary improvement in response speed caused by forced convection; the dynamic response characteristic time of a traditional pure radiation preheating system is:

[0066]

[0067] Where L = 50 mm is the characteristic length, α eff =8.5×10 -6 m 2 / s represents the effective thermal diffusivity of the powder.

[0068] With the introduction of forced convection, convective heat transfer becomes the dominant heat transfer mechanism, and the system response time is reduced to:

[0069]

[0070] Where h is the convective heat transfer coefficient. The response speed is improved by approximately 50 times, which is the physical basis for the rapid temperature regulation achieved in this invention.

[0071] Mechanism 2: Spatial coupling effect of hot air heat transfer along the flow path. When hot air flows through the surface of the powder bed, its temperature decay along the flow path follows a one-dimensional energy conservation equation:

[0072]

[0073] Where: P wet For wet perimeter, Air mass flow rate.

[0074] The analytical solution to this equation is:

[0075]

[0076] CFD simulation verification shows that under typical operating conditions (v air =0.5m / s, T air,in =200℃), with a temperature drop of 15-20℃ along the flow path, forming a significant upstream and downstream temperature difference. This spatial coupling effect is the physical basis of the feedforward control strategy of this invention.

[0077] Mechanism 3: Inter-coupling analysis of radiation fields from multiple lamps. Based on the theory of radiative heat transfer, the radiative interaction between adjacent zones can be analyzed using the coupling coefficient ξ. ij Quantification:

[0078]

[0079] Among them, F ij Let ξ be the angle coefficient from partition i to partition j. Experimental results show that the coupling coefficient ξ between adjacent partitions... 12 ≈0.18, meaning that when the power of the upstream zone changes by 1kW, the radiative heat flux density received by the downstream zone changes by approximately 180W / m. 2 This predictable coupling provides theoretical guidance for the parameter tuning of feedback controllers.

[0080] A complete mathematical model of the system was established, and three key physical mechanisms and their quantitative relationships were identified. These model and mechanism analyses provide clear theoretical guidance for subsequent parameter calibration and controller design.

[0081] S3. Systematic calibration of key controller parameters. Parameter calibration is the bridge connecting the theoretical model and actual control. This step, based on the theoretical analysis in S2, obtains the key parameters required by the control algorithm through rigorous experimental procedures. The core objective of calibration is to quantify the "spatial coupling effect" identified in S2 and obtain the decoupling compensation coefficient C. ij .

[0082] S31. Based on the physical model established in S2, when the temperature of upstream partition i deviates from the target value ΔT i At that time, according to the heat transfer equation along the hot air path, the air temperature change upon reaching downstream zone j is:

[0083]

[0084] To maintain a constant temperature in downstream section j, the required compensation power is:

[0085]

[0086] Among them, A j η is the effective control area of ​​partition j. lamp The radiative-convective heat transfer efficiency of the lamp tube.

[0087] Therefore, the theoretical expression for the decoupling compensation coefficient is:

[0088]

[0089] The formula shows that the compensation coefficient decreases exponentially with the distance between intervals, which is completely consistent with the physical law of heat transfer along the hot air path, and provides theoretical guidance for experimental calibration.

[0090] S32. To ensure the accuracy and reproducibility of calibration results, a strict standard operating procedure (SOP) must be established:

[0091] Phase A: Establishing the system steady-state baseline, starting all hardware systems, running the preheating program to 180℃ and maintaining it for 30 minutes to ensure the system reaches complete thermal equilibrium; adjusting all feedforward gains C in the control software. ij Set to zero to enable only PID feedback control; record the base power P_j,base for each zone, which will serve as a reference point for subsequent calculations; verify temperature stability: the temperature fluctuation in each zone should be less than ±1℃ for 10 minutes.

[0092] Phase B: Systematic Disturbance Test. For each upstream partition i, perform the following standardized disturbance test: Adjust the target temperature of partition i from 180℃ to 175℃ (ΔT = -5℃); keep the target temperatures of other partitions unchanged and observe the system response; record the power change process of each downstream partition j until a new steady state is reached (criteria: power change rate < 0.1% / min, lasting for 5 minutes); record the power P of each partition under the new steady state. j,new .

[0093] Phase C: Parameter calculation and verification, calculation of compensation coefficient: C ij =(P j,new P j,base ) / ΔT; Perform reverse verification: restore the temperature of partition i to 180℃ and verify whether the power of each partition returns to the baseline value; repeat the test 3 times and take the average value as the final parameter to ensure the statistical reliability of the results.

[0094] S33. After completing the testing of all partition pairs, construct the complete compensation coefficient matrix C. Based on physical principles, this matrix has the following properties: Upper triangularity: C ij = 0 (when i ≥ j), because downstream does not affect upstream; distance attenuation: |C ij | decreases as |ij| increases, with the strongest influence from adjacent partitions; symmetric constraint: under the same geometric conditions, C_12≈C_23.

[0095] These physical constraints are used to verify the rationality of the measured parameters and make necessary corrections to ensure their physical consistency. A rigorously calibrated and verified compensation coefficient matrix C is obtained, which accurately quantifies the spatial coupling characteristics of the system and provides a reliable parameter basis for the feedforward compensation of the lower-level controller.

[0096] S4. Based on the hardware platform, theoretical model, and calibration parameters established in the first three steps, this step implements the software-based control algorithm. The software architecture design must balance real-time performance, reliability, and maintainability, embodying the core concept of "layered collaboration."

[0097] S41. The software architecture adopts the design concept of "layered decoupling and concurrent execution": Time scale separation: the upper-level controller handles slow-changing processes (minute level), and the lower-level controller handles fast-changing processes (millisecond level); Functional modularization: each functional module is designed independently and communicates through standard interfaces, which facilitates debugging and maintenance; Real-time guarantee: key control tasks adopt hard real-time scheduling to ensure time determinism.

[0098] S42, Visual Processing Thread (Highest Priority): Triggering Mechanism: Directly triggered by the hardware interrupt of the infrared camera, ensuring zero-latency response; Execution Logic: Obtain raw thermal image data (14 bits) from the camera SDK buffer, perform bad pixel correction and temperature calibration conversion, and store the processed temperature matrix into a thread-safe circular buffer; Performance Indicators: Single frame processing time <2ms, ensuring real-time performance at a 50Hz sampling rate.

[0099] Main control loop thread (second highest priority): Scheduling mechanism: precisely woken up by a high-precision hardware timer (accuracy ±10μs) with a period of 20ms; Execution flow: strictly follows the five-step sequence of "data acquisition → ROI processing → control calculation → output update → safety check"; Time budget: total execution time <15ms, with a 5ms safety margin.

[0100] User Interface and Management Thread (Normal Priority): Functional Responsibilities: Handles operator interaction, parameter configuration, data logging, and upper-level optimization calculations; Non-real-time Characteristics: The execution delay of this thread does not affect control performance and can be preempted by higher-priority threads.

[0101] S43. Engineering Implementation of Upper-Level Controller (Steady-State Optimization)

[0102] S431. The upper-level controller solves the following multi-objective constrained optimization problem:

[0103]

[0104] Constraints:

[0105]

[0106] The formulas for calculating each term of the objective function are as follows: Total power consumption: Temperature variance:

[0107] The weighting coefficients are determined based on application requirements: for high-precision applications, w uniformity =0.7, w energy =0.3; for energy-saving applications, it can be adjusted to w uniformity =0.5, w energy =0.5.

[0108] S432. Considering computational efficiency and reliability, an implementation scheme combining offline optimization and online table lookup is adopted:

[0109] Offline data generation stage: in two-dimensional parameter space (T) {air,in} ,v {air}A 21×17 uniform grid (357 working points) was established on the surface; for each grid point, the corresponding P was obtained through CFD simulation or experimental testing. total and Var(T steady ); Calculate the objective function value J, and generate triplet data (T) {air,in} ,v {air} ,J); Store data in binary format, supporting fast memory-mapped access.

[0110] Online table lookup optimization phase: When the upper-level controller is triggered, the entry with the smallest J value is searched in the lookup table in memory; a binary search algorithm is used, with a search time complexity of O(logN) and a typical execution time of <100μs; the optimal operating point (T) is output. {air,in} ,v {air} ^), and then issued to the executing agency via the DAQ card.

[0111] S44, Refined Implementation of Lower-Level Controller (Dynamic Compensation)

[0112] S441. Dynamic ROI generation is the key innovation of this invention, and its implementation includes the following steps:

[0113] Laser position synchronization: The position feedback (x) of the laser galvanometer system is read at 1ms intervals via the EtherCAT interface. laser ,y laser Considering communication delay and system response time, a look-ahead prediction of the laser position is performed: (x pred ,y pred )=(x laser ,y laser )+v laser ×t delay Predicted time t delay =5ms, derived from the calibration of system response characteristics.

[0114] Dynamic mask generation: A circular mask region with a radius of r = 15 mm is generated, centered on the predicted laser position; pixels within the mask in the temperature matrix are marked as sintering areas, and the rest are preheating control areas; the mask radius is selected based on laser power and scanning speed: r = f(P laser ,v scan ), to ensure complete coverage of the high-temperature affected area.

[0115] Zoned Temperature Calculation: Within the preheating control zone, the average temperature of each zone is calculated according to its geometric boundaries; an area-weighted average method is used to improve calculation accuracy: T j =Σ(T pixel ×A pixel ) / Σ(A pixel Applying a first-order low-pass filter to smooth the temperature signal: T j,filtered=α×T j +(1-α)×T j,prev The filter coefficient α = 0.8.

[0116] S442. For each partition j, its control output consists of a feedback term and a feedforward term:

[0117] Feedback control: An improved PID controller is used, featuring anti-integral saturation and derivative-first functions.

[0118]

[0119] Wherein, the error is defined as: e j (t)=T ref -T j (t).

[0120] The PID parameters are tuned using a piecewise tuning strategy: Start-up phase (0-60s): K p,j =1.2, K i,j =0.02, K d,j =0.08 (fast response) steady-state stage (>60s): K p,j =0.8, K i,j =0.05, K d,j =0.15 (precise control);

[0121] Feedforward compensation term: Based on spatial coupling theory and calibration parameter matrix, the mathematical expression for feedforward compensation is:

[0122]

[0123] Among them, C ij The steady-state compensation coefficient, C, is obtained through experimental calibration in step 3. ′ ij : Dynamic compensation coefficient, C′ ij =0.1·C ij ·τ air , τ air This is the hot air transmission delay time;

[0124] The physical meaning of this feedforward term is: to predictively adjust the downstream power based on the upstream temperature deviation and its rate of change, and actively counteract the cooling effect of hot air along the path.

[0125] Output limiting and safety protection: Total output power is limited to the range of [0, 100%]; Anti-integral saturation: When the output is saturated, the integral term accumulation stops; Temperature upper limit protection: When the temperature of any zone exceeds 220℃, the heating output of that zone is immediately cut off.

[0126] S45. After software development is completed, system-level integration testing should be conducted according to engineering specifications to ensure the reliability and performance indicators of the entire system.

[0127] Functional integrity testing: Verify the independent functional correctness of each subsystem (hot air, zoned heating, temperature monitoring, communication); test the collaborative working capability of the hierarchical controllers to ensure error-free data transmission and command execution between upper and lower level controllers; verify the dynamic ROI generation function to ensure the accuracy of laser position synchronization and mask generation.

[0128] Performance indicator verification: Control accuracy test: Under standard operating conditions, the temperature control accuracy of each zone should be within ±2℃; Response time test: The time from the change of the target temperature to the system stabilization should be <30s; Stability test: After 24 hours of continuous operation, the temperature fluctuation should be <±1℃, and there should be no system failure.

[0129] Safety protection test: Verify over-temperature protection: When an over-temperature condition is manually set, the system should cut off the heating output within 2 seconds; Verify communication fault protection: When the communication link is disconnected, the system should enter a safe shutdown mode; Verify watchdog function: When a software anomaly is simulated, the hardware watchdog should automatically power off for protection.

[0130] A fully functional and reliable control software system was completed, passed rigorous engineering testing and verification, and achieved the final transformation from theoretical design to engineering application.

[0131] Through the systematic implementation of the above four steps, this invention successfully constructed a preheating control system for additive manufacturing powder beds based on hot air-zonal radiation synergy. The entire implementation process embodies a rigorous engineering methodology:

[0132] 1. From hardware selection to software implementation, each step has clear technical basis and performance indicators, forming a complete technical implementation chain. The hardware platform of S1 provides the experimental basis for the system analysis of S2; the theoretical model of S2 guides the parameter calibration of S3; and the calibration parameters of S3 support the implementation of the control algorithm of S4.

[0133] 2. Through detailed hardware configuration, parameter calibration, and software architecture design, it has been proven that this invention is not only theoretically advanced but also highly operable in engineering practice. All key technical aspects have specific implementation plans and performance assurance measures.

[0134] 3. Compared with traditional preheating methods, this system has achieved significant improvements in several key indicators: Response speed: improved from 300s to 30s, a 10-fold increase; Temperature uniformity: improved from ±5℃ to ±2℃, a 150% increase; Energy efficiency ratio: reduced system energy consumption by 15-20% through optimized control strategies; Reliability: reduced system failure rate to less than one in ten thousand through multiple safety protection mechanisms.

[0135] The implementation plan of this project lays a solid technical foundation for the industrial application of temperature field control technology in additive manufacturing and has important value for promotion and application.

[0136] Working Principle: The hot air circulation system generates hot air at a specific temperature and velocity through an adjustable-speed fan and an adjustable-power heater, which flows over the powder bed surface to achieve rapid and fundamental heat transfer. The zoned radiant heating system consists of multiple independently controllable heating elements, arranged in several zones above the powder bed for compensatory and differentiated heating in specific areas. The temperature monitoring device monitors the surface temperature distribution of the powder bed in real time non-contact and transmits the data to the hierarchical collaborative controller. The upper-level steady-state optimization of the hierarchical collaborative controller, on a slow time scale, solves the optimization problem based on the target preheating temperature and energy efficiency target to determine the global optimal basic operating point of the hot air circulation system, including the hot air inlet temperature and wind speed. The lower-level dynamic compensation, on a fast time scale, uses a multivariable decoupled feedforward control strategy based on the given basic operating point in the upper level to rapidly adjust the power of each zoned radiant heating element according to real-time temperature feedback, accurately tracking the target temperature and suppressing local temperature disturbances. In the specific control process, the dynamic ROI technology generates a circular mask area to mark the high-temperature "sintering zone" of the laser through laser position synchronization and look-ahead prediction. It calculates the zone temperature based solely on the temperature data of the "preheating control zone" and uses area-weighted averaging and low-pass filtering to process the temperature signal. The feedback control term uses an improved PID controller with anti-integral saturation and derivative-ahead functions, and adjusts according to the temperature deviation of the local area. The feedforward control term is based on spatial coupling theory and a calibrated compensation coefficient matrix. It predictively adjusts the downstream power according to the upstream zone temperature deviation and its rate of change, actively offsetting the cooling effect of hot air along the process. At the same time, it ensures the stable operation of the system through output limiting and safety protection mechanisms. Ultimately, it achieves rapid, uniform, and dynamic control of the powder bed temperature field, improving the stability of the additive manufacturing process and the forming quality of the parts.

[0137] 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 invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0138] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A preheating system for additive manufacturing powder bed based on hot air-zoned radiation synergy, comprising a preheating outer chamber (1) and a powder bed (2), characterized in that: The powder bed (2) is located at the bottom of the preheating outer box (1). The preheating outer box (1) is equipped with a hot air circulation system and a zoned radiation heating system. The hot air circulation system heats the powder bed (2) with hot air in the horizontal direction, and the zoned radiation heating system heats the powder bed (2) with hot air in the vertical direction. The upper end of the preheating outer box (1) is fixedly connected to a layered collaborative controller (7). The layered collaborative controller (7) includes an upper controller and a lower controller. The upper controller of the layered collaborative controller (7) determines the basic working point of the hot air circulation system according to the preheating temperature of the powder bed (2) and the preset optimization target. The lower controller of the layered collaborative controller (7) dynamically adjusts the power of each heating zone in the zoned radiation heating system according to the preheating temperature of the powder bed (2) and the preset optimization target. The hot air circulation system includes an air duct (3) and a temperature sensor (6). The zoned radiation heating system includes a thermal radiation lamp (4) and an infrared thermal imager (5) fixedly connected to the inner wall of the top of the preheating outer box (1). The thermal radiation lamps (4) are arranged in three groups at equal intervals. The thermal radiation lamps (4) and the infrared thermal imager (5) are both located directly above the powder bed (2). The thermal radiation lamps (4) and the infrared thermal imager (5) are both electrically connected to the layered collaborative controller (7).

2. The additive manufacturing powder bed preheat system of claim 1, wherein: The preheating outer box (1) has a hot air inlet (11) and a hot air outlet (12) on both sides of the box. The preheating outer box (1) with the hot air inlet (11) is sealed with an air inlet hood (13). The preheating outer box (1) with the hot air outlet (12) is sealed with an air outlet hood (14). The air guide pipe (3) is set on the outside of the preheating outer box (1). The two ends of the air guide pipe (3) are connected to the air inlet hood (13) and the air outlet hood (14) respectively. A fan (31) and a heater (32) are installed on the pipe body of the air guide pipe (3). The temperature sensor (6) is fixedly connected to the inner wall of the box on the side with the hot air inlet (11) of the preheating outer box (1). The fan (31), the heater (32) and the temperature sensor (6) are all electrically connected to the layered collaborative controller (7).

3. A preheating method for additive manufacturing powder bed based on hot air-zoned radiation synergy, comprising the additive manufacturing powder bed preheating system as described in any one of claims 1-2, characterized in that: The preheating method includes the following steps: Step 1: Upper Layer Preheating Optimization: Based on the set target preheating temperature The basic operating point of the hot air circulation system is determined by the upper-level controller, including the hot air inlet temperature. and wind speed In step one, the basic working point is obtained by solving a multi-objective optimization problem, the objective function of which is: ； wherein, Ptotai is the total power consumption of the system, is the steady state temperature variance; Step 2, Lower-level dynamic compensation: The lower-level controller executes a real-time compensation cycle, calculates and updates the power of each heating zone based on the real-time temperature feedback from the infrared thermal imager (5), so as to accurately track the target preheating temperature.

4. The preheating method of claim 3, wherein: In step one, the basic working point is obtained by solving a multi-objective optimization problem, the objective function of which is: ； wherein, Ptot is the total system power consumption, σT is the steady state temperature variance.

5. The preheating method according to claim 3, characterized in that: In step two, for any non-uppermost heating zone j, its compensated heating power The calculation includes: a feedback control term based on the temperature deviation in this region. It is used to compensate for the overall thermal effects, including radiation coupling from multiple lamps; and a feedforward control term based on a temperature deviation of at least one upstream heating zone (i) for decoupling the upstream and downstream thermal effects caused by the in-line cooling of the hot air.

6. The preheating method of claim 5, wherein: Located in the heating zone Temperature deviation from target value At that time, according to the heat transfer equation along the hot air path, it reaches the downstream zone. The air temperature change is as follows: ； To maintain the temperature constant in the downstream partition The required compensation power is: ； wherein: is the effective control area of the partition is the effective control area of the partition is the radiation-convection heat transfer efficiency of the lamp; thus, the theoretical expression of the decoupling compensation factor is: ； The formula shows that the compensation coefficient decreases exponentially with the distance between intervals.

7. The preheating method of claim 5, wherein: The feedback control term With improved PID controller, with anti-integral saturation and differential anticipation function: ； The error is defined as follows: ; The PID parameters are tuned using a segmented tuning strategy: Start-up phase (0-60s): , , Ensure rapid response; steady-state phase (>60s): , , To ensure precise control.

8. The preheating method of claim 5, wherein: The feedforward control term The calculation formula is: ； in, upstream partition Real-time temperature, The steady-state decoupling compensation coefficients are obtained through system identification experiments. For dynamic compensation coefficients, , This is the delay time for hot air transmission.

Images