Multi-laser power monitoring method for additive manufacturing field

Abstract

The invention is suitable for the technical field of additive manufacturing, and provides a multi-laser power monitoring method for the field of additive manufacturing, which comprises the following steps: power calibration in a non-processing state, periodic monitoring in a printing process and power abnormity closed-loop adjustment. In the invention, the hardware cost is obviously reduced; the maintenance cost is greatly reduced; the laser state has no interference; the material adaptability is high; the monitoring precision is high; abnormal response is timely; alignment is verified through the voltage signal amplitude, manual micrometer calibration (2 hours per time in the existing scheme) is not needed, the alignment time is shortened to 10 minutes per time, and the operation efficiency is improved by 12 times; carrying out data visualization; and the laser quantity is adaptive.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of additive manufacturing, and particularly relates to a multi-laser power monitoring method for the field of additive manufacturing. BACKGROUND

[0002] Laser power stability is the key to determining the quality of the component: too high power will cause excessive melting of the material, splashing and thermal deformation, and too low power will cause insufficient melting of the powder, insufficient density (usually <95%), and even delamination between layers. Large-size component printing has a long cycle (≥48 hours) and high cost (single component material cost ≥100,000 yuan), and if the power of a single laser is abnormal and not discovered in time, the entire component will be scrapped, resulting in significant economic losses. Therefore, online monitoring of multiple laser powers is a core requirement for large-size additive manufacturing, but existing monitoring solutions have the following unresolved defects:

[0003] Cost and space conflict is prominent: existing solutions generally use independent configuration of "one laser-one monitoring module", and for 25 lasers, the cost of a single monitoring module is ≥5000 yuan, and the total monitoring cost is ≥125,000 yuan, which is far beyond the budget of small and medium-sized equipment; at the same time, the top of the large-size equipment needs to integrate a galvanometer and optical adjustment components, and the reserved space is ≤50mm, so 25 monitoring modules cannot be installed at the same time, resulting in a quality risk for some lasers;

[0004] Optical path changes affect printing accuracy: some solutions add a beam splitter (splitting ratio 90%:10%) in the laser light path to divide the laser into a main beam (for printing) and a secondary beam (for monitoring). This design will cause a 5%-15% loss of main beam power, and installation errors of the beam splitter (even ≤0.05mm) will shift the laser focal point, causing uneven energy density in the printing area, and the component density deviation can reach ±3%, which cannot meet the requirement of ≥99.5% density for aerospace components;

[0005] Indirect monitoring has poor accuracy and adaptability: another type of solution indirectly infers laser power by collecting the radiation intensity of the molten pool, but the molten pool radiation is disturbed by multiple factors: first, material differences, the radiation coefficient of titanium alloy and stainless steel differs by ≥20%, and the radiation intensity deviation is significant under the same power; second, process parameter influence, the radiation intensity fluctuates by ≥5% when the scanning speed changes by 100mm / s and the layer thickness changes by 0.05mm. This type of solution needs to build a complex model such as CNN-LSTM for fitting (training period ≥72 hours), and needs to be retrained when the material is changed, and the power inference error is ≥±5W, which is much higher than the ≤2W deviation allowed for large-size components;

[0006] Real-time and device computing power do not match: the existing real-time monitoring scheme (response time ≤10 ms) needs to continuously occupy the device computing power, while the large-size device printing needs to process multiple laser scanning paths (single laser path data ≥100 MB / layer), temperature control and other data synchronously, and continuous monitoring will cause computing power overload, scanning lag (≥1 s), layer interlocking defects, and decrease of 10%-15% of the tensile strength of the component, and more than 20% reduction of fatigue life.

[0007] Therefore, a multi-laser power monitoring method for the field of additive manufacturing is needed to solve the above problems. SUMMARY

[0008] The purpose of the embodiments of the present application is to provide a multi-laser power monitoring method for the field of additive manufacturing to solve the problems raised in the background.

[0009] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0010] A multi-laser power monitoring method for the field of additive manufacturing, comprising the following steps:

[0011] S1, power calibration in a non-processing state: performing preset power point output, standard power measurement, voltage acquisition and linear relationship fitting on each laser in the multi-laser array (≥16 lasers, matrix arrangement) of the additive manufacturing device, establishing the "power-voltage" reference of each laser, and the calibration process does not change the original light path of the laser;

[0012] S2, periodic monitoring during printing: pausing the additive manufacturing device after printing a preset number of layers, acquiring the power data of each laser by grouping monitoring units (containing photodetectors and attenuators) according to the "1 unit-5 lasers" grouping logic, and calculating the real-time power mean and power deviation;

[0013] S3, power abnormality closed-loop adjustment: if the power deviation exceeds the preset threshold, start secondary verification to exclude accidental errors, and still abnormal, fine-tune the laser drive current until the deviation meets the requirements; store the monitoring data throughout the process for quality traceability.

[0014] Further technical solutions, the "preset power points" in step S1 include 100W, 200W, 300W, 400W four power values, covering the printing power requirements of mainstream materials (titanium alloy, stainless steel, nylon) in additive manufacturing;

[0015] The "standard power measurement" uses a standard power meter with an accuracy of ±0.1W, and each power point of each laser is measured 5 times, and the mean value is taken as the actual power value.

[0016] Further technical solutions, step S1 "linear relationship fitting" using the least squares method, get each laser linear equation P = aiUi (ai for the conversion factor), fitting error ≤±0.1W;

[0017] If the calibration deviation calculated by the formula ΔP = | preset power - actual power average | > 0.1W, the laser needs to be calibrated and the step S1 is repeated until the deviation ≤0.1W and ai is stored.

[0018] Further technical solutions, step S2 "preset layer number" is 100 layers, suitable for the printing rhythm of additive manufacturing equipment (100 layers takes 4-6 hours);

[0019] The photodetector of the "grouped monitoring unit" is an InGaSe detector (wavelength response 800nm-1700nm) or an infrared detector (wavelength response 8-12μm), with a power measurement range of 50W-500W, and a reflective attenuator with an attenuation rate of 80%-90% to avoid damage to the detector.

[0020] Further technical solutions, step S2 "monitoring unit and laser substrate center alignment" verification method: control the corresponding laser to emit laser, observe the monitoring unit voltage signal, adjust the automatic displacement table (resolution 0.01mm) to make all monitoring unit voltage signal amplitude deviation ≤5%, ensure that the alignment error ≤0.1mm.

[0021] Further technical solutions, step S2 "power data acquisition" specific way: control the corresponding group of lasers to output standard process power P si , single light output 1s (ensure signal stability), each laser repeats 5 times power data p ij (i represents the laser number; j represents the test number), according to the formula Calculate the real-time power average, according to ΔP' i = |P si - P' i |(i = 1, 2, …, 25) to calculate the power deviation.

[0022] Further technical solutions, step S3 "preset threshold" is 2W;

[0023] "Secondary verification" needs to repeat the power data acquisition and calculation of step S2, if the secondary verification deviation is still >2W, the data processing module calculates the current adjustment amount (based on the "power-current" linear coefficient of the laser) according to ΔP' i Value, through the drive circuit to fine-tune the current.

[0024] Further technical solutions, step S3 "data storage" needs to record the monitoring data of every 100 layers, including the pij P' i ΔP' i And adjust the record, the storage medium is an industrial grade SD card (capacity ≥ 8 GB), the data retention time is ≥ 5 years, and the quality traceability requirement of the additive manufacturing industry is met.

[0025] Compared with the prior art, the beneficial effects of the present application are:

[0026] The hardware cost of the present application is significantly reduced: by adopting the "1 unit-5 laser" grouping logic, the number of monitoring modules of 25 lasers is reduced from 25 to 5, the hardware cost is reduced from 125,000 yuan to 25,000 yuan, and the cost is reduced by 80%; when applied in batches (≥10 devices), the total cost can be saved by more than 1,000,000 yuan, and the equipment investment threshold of small and medium-sized manufacturers is reduced;

[0027] The maintenance cost is greatly reduced: the monitoring module is integrated on the scraper, and only the scraper needs to be disassembled (without disassembling the top light path of the equipment) during maintenance, the maintenance time is shortened from 4 hours / time of the existing scheme to 1 hour / time, the annual maintenance cost is reduced from 20,000 yuan per unit to 0.8 yuan per unit, and the cost is reduced by 60%;

[0028] The present application has no interference with the laser state: the monitoring unit is deployed at the rear end of the laser light (close to the substrate side), does not interfere with the original light path, avoids the power loss (5%-15%) and focal point deviation (≤0.1mm) caused by the beam splitter, ensures that the substrate laser power deviates from the theoretical value by ≤0.5W, and the component density is stable at ≥99.5% (the existing scheme is ≤97.2%);

[0029] Strong material adaptability: no need to adjust the light path according to the material, only need to update the conversion coefficient ai of the data processing module (such as titanium alloy ai=200W / V, stainless steel ai=180W / V), the adaptation time is ≤30 minutes, compared with the existing indirect monitoring scheme (retraining model for 72 hours), the adaptation efficiency is improved by 864 times;

[0030] The present application has high monitoring accuracy: through "5 times average + linear calibration", the monitoring error is ≤±0.3W, which is much better than the ±5W error of the existing indirect scheme, and can meet the precision requirements of aerospace core components (density ≥99.5%, fatigue strength ≥950MPa);

[0031] Abnormal response in time: power abnormalities can be found in time every 100 layers, avoiding the component scrap caused by abnormality lasting for 4-6 hours, the qualified rate of large-size component printing is increased from 70% to more than 95%, the single component cost loss is reduced from 100,000 yuan / time to 5,000 yuan / time, and the economic loss is significantly reduced;

[0032] The present application, automatic alignment: alignment time is shortened to 10 minutes / time, and operation efficiency is improved by 12 times through voltage signal amplitude verification without manual calibration by micrometer (2 hours / time in the prior art scheme);

[0033] Data visualization: each laser deviation (red pre-alarm anomaly) is displayed in real time on the touch screen without manual copying of data (30 minutes / time in the prior art scheme), manual workload is reduced by 90%, and manual recording errors are avoided;

[0034] The present application, laser number adaptation: if the device is 16 lasers (4 rows and 4 columns), the monitoring unit can be adjusted to 4, and only the grouping parameters are updated (from 5 groups to 4 groups), and the extension time is less than or equal to 1 hour without the need to replace hardware;

[0035] Laser type adaptation: if it is a CO2 laser (10.6 μm), only the detector needs to be replaced to an infrared type (cost less than or equal to 2000 yuan per unit), without the need to change other links, and the adaptation meets the non-metal (nylon, resin) printing demand.

[0036] In order to more clearly illustrate the structural features and effects of the present application, the present application will be described in detail below in combination with the drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 It is a laser arrangement diagram of a 9-laser additive manufacturing device at the top of the present application;

[0038] Figure 2 It is a principle diagram of a multi-laser power monitoring scheme A of the present application;

[0039] Figure 3 It is a principle diagram of a multi-laser power monitoring scheme B of the present application;

[0040] Figure 4 It is a monitoring layout diagram in a cabin of a 25-laser additive manufacturing device of the present application;

[0041] Figure 5 It is a schematic diagram of Table 1 of the present application;

[0042] Figure 6 It is a schematic diagram of Table 2 of the present application;

[0043] Figure 7 It is a schematic diagram of Table 3 of the present application.

[0044] Figure 8 It is a flowchart of the present application. DETAILED DESCRIPTION

[0045] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0046] The specific implementation of the present application is described in detail below in combination with specific embodiments.

[0047] Example 1: 25 laser titanium alloy components (aerospace turbine blades) monitoring

[0048] As shown in the drawings, the embodiment of the present application provides a multi-laser power monitoring method for the field of additive manufacturing, comprising the following steps: Figures 1-8 Step S1, non-processing calibration

[0049] S11, control 25 lasers to output 100W, 200W, 300W, 400W in turn, and each power point is stable for 5s;

[0050] S12, measured by Ophir Nova II standard power meter (±0.1W), No. 1 laser average: 100.1W (100W set), 200.2W (200W set), 300.3W (300W set), 400.2W (400W set);

[0051] S13, collect voltage value: 0.5005V (100W), 1.001V (200W), 1.5015V (300W), 2.001V (400W), and fit P=200U (ai=200W / V, error 0.05W);

[0052] S14, ΔP1=0.1W≤0.1W, store ai; the remaining 24 a is 199.8-200.2W / V.

[0053] Step S2, periodic monitoring

[0054] S21, install the monitoring unit to the scraper, adjust the displacement table so that the voltage amplitude deviation is ≤3%, and the alignment error is 0.08mm;

[0055] S22, import turbine blade model, start printing (P=350W, scanning speed 1200mm / s)

[0056] S23, print 100 layers (4.5 hours, powder laying 15kg), pause the equipment;

[0057]

[0058] ​S24, move the scraper to the center of laser No. 1-5, control the output of 350W, single light emission for 1s, collect No. 1 laser for 5 times: 349.8W, 350.2W, 349.9W, 350.1W, 350.0W;

[0059] S25, calculate P'1=350.0W, ΔP'1=0.0W; complete the monitoring of No. 6-25, and the P' of No. 23 laser is 347.5W (ΔP'=2.5W>2W);

[0060] Step S3, closed-loop adjustment and traceability;

[0061] S31, recheck No. 23 laser, P'=347.6W (ΔP'=2.4W>2W);

[0062] S32, calculate the adjustment amount (power-current coefficient 0.5W / mA, need to adjust 4.8mA), and the current of the driving circuit is increased from 3.5A to 3.5048A;

[0063] S33, reacquire No. 23 laser P'=349.9W (ΔP'=0.1W≤2W), and resume printing;

[0064] S34, repeat the monitoring every 100 layers, complete the printing after 48 hours, and store 12 groups of data (1200 layers) in the SD card.

[0065] In this embodiment, the multi-laser device is SLMSolutions1200 type (metal powder bed fusion process, substrate 2050mm×2050mm);

[0066] Laser array: 25 IPGYLR-400 fiber lasers (1064nm, 100-400W, 5 rows and 5 columns, 400mm×400mm per scanning area);

[0067] Monitoring unit: 5 Thorlabs PDA10DT-EC InGaSe detectors (800-1700nm, 50-500W) + at least 2 pieces of 98% attenuation film, aluminum alloy dustproof shell (translucent quartz window≥95% transmittance);

[0068] Adjustment assembly: HIWINK KH20 automatic displacement table (0.01mm resolution, ≤0.02mm repeat positioning error) + SKD11 scraper (load≥10kg);

[0069] Data processing: Xilinx Artix-7 FPGA (calculation) + STM32H7 ARM (instruction), 16GB industrial SD card;

[0070] Component parameters: titanium alloy TC4 turbine blade (1000mm x 600mm x 300mm, standard process power P = 350W, powder laying 0.05mm)

[0071] Precision and quality effect: the deviation of 25 lasers is less than or equal to 0.5W, the turbine blade density is 99.7% (Archimedes method), the tensile strength is 950MPa (in line with HB7762-2022), the fatigue life is 10 7 times of secondary circulation (existing scheme 8 x 10 6 times), detected by a third party in the aerospace field;

[0072] Cost and efficiency effect: monitoring cost 25,000 yuan (existing scheme 125,000 yuan), reduced by 80%; printing qualified rate from 70% to 98%, avoiding 1 time of scrapping (saving 150,000 yuan of titanium alloy cost); monitoring time 375s / 100 layers, accounting for 5.2% of the total cycle, no efficiency loss;

[0073] Maintenance and adaptation effect: ΔP' = 2.2W caused by pollution of No. 2 detector in later period, locate fault through data tracing for 30 minutes, return to normal after cleaning (existing scheme 4 hours); when replacing stainless steel printing, update ai = 180W / V, adaptation time 25 minutes, very high efficiency.

[0074] Example 2: 16 laser nylon component (aviation support) monitoring

[0075] The difference between this embodiment and example 1 is:

[0076] Step S1, non-processing calibration;

[0077] S11, control 16 lasers to output 50W, 100W, 150W, and stabilize for 5s at each power point;

[0078] S12, measure with Coherent PowerMax-Pro standard power meter (±0.1W), No. 1 laser average: 50.0W (50W), 100.1W (100W), 150.1W (150W);

[0079] S13, collect voltage values: 0.625V (50W), 1.251V (100W), 1.876V (150W), and fit P = 80U (ai = 80W / V, error 0.08W);

[0080] S14, ΔP1≤0.1W, store ai; the remaining 15 ai are 79.8-80.2W / V.

[0081] Step S2, periodic monitoring;

[0082] S21, adjust the displacement table to make the voltage amplitude deviation ≤5%, the alignment error is 0.09mm;

[0083] S22, import the support model, and start printing (P=120W, scanning speed 800mm / s);

[0084] S23, print 100 layers (3 hours, powder laying 8kg), and pause the equipment;

[0085] S24, move the scraper to the center of the laser No. 1-4, collect 5 sets of data, and the laser No. 12 has P'=117.8W (ΔP'=2.2W>2W).

[0086] Step S3, closed-loop adjustment and traceability;

[0087] S31, recheck the laser No. 12, P'=117.7W (ΔP'=2.3W>2W);

[0088] S32, calculate the adjustment amount (power-current coefficient 0.2W / mA, 15mA needs to be adjusted), and the current is increased from 2.0A to 2.015A;

[0089] S33, reacquire the laser No. 12 P'=119.9W (ΔP'=0.1W), and restore printing;

[0090] S34, repeat the monitoring every 100 layers, and complete the printing after 36 hours, and store 9 sets of data (900 layers).

[0091] In this embodiment, the multi-laser equipment is Stratasys Fortus 900mc type (nylon powder sintering process, base plate 1500mm*1500mm);

[0092] Laser array: 16 Synrad 48-5 CO2 lasers (10.6μm, 50-150W, 4 rows and 4 columns, 300mm*300mm per scanning area);

[0093] Monitoring unit: 4 Thorlabs PDA50B-EC infrared detectors (8-12μm, 50-200W) + at least 2 pieces of 98% attenuation film, ceramic dustproof shell (temperature resistance ≥200℃);

[0094] Adjustment component: THKKR30 automatic displacement table (0.01mm resolution) + aluminum alloy scraper (load ≥8kg);

[0095] Data processing: Xilinx Spartan-7 FPGA+STM32F4 ARM, 8GB SD card;

[0096] Component parameters: nylon 6 aviation bracket (1200mm*800mm*200mm, standard process power P

[0097] =120W, powder laying 0.1mm);

[0098] Material adaptation effect: infrared detector perfectly adapts to CO2 laser (10.6um), tensile strength of nylon bracket 65MPa (complying with ISO527-2012), surface bubble rate 0.5% (existing scheme 3%), meeting the requirements of aviation lightweight;

[0099] Environmental adaptation effect: printing cabin 150℃, dust 5mg / m 3 , ceramic shell makes the sensitivity attenuation of the detector less than or equal to 5%, and the monitoring error is less than or equal to 0.3W (existing scheme is greater than or equal to 2W) for 36 hours;

[0100] Cost and expansion effect: monitoring cost 18,000 yuan (existing scheme 80,000 yuan), reduced by 77.5%; when expanded to 20 laser equipment, one unit is added + update grouping parameters (45 minutes), and the adaptation cost is less than or equal to 2,000 yuan, meeting the demand of multi-specification equipment.

[0101] Working principle and use process of the application:

[0102] Based on the whole-link logic of "physical signal conversion-accurate data calculation-execution closed-loop regulation-full-process data tracing", the fine control of multi-laser power is realized, and the specific working process is divided into four stages:

[0103] First stage: non-processing state calibration - establish "power-voltage" reference;

[0104] The core of this stage is to eliminate the individual differences of each laser, to provide "one-to-one" accurate reference for subsequent monitoring, and to avoid reference error transmission:

[0105] Laser emission and signal conversion, including laser output and signal acquisition;

[0106] Laser output: multiple laser arrays output laser according to preset power (100W, 200W, 300W, 400W), which is calibrated to parallel light by a collimating mirror, and is vertically irradiated to the center of the substrate after focusing by a galvanometer and a field lens (such as No. 1 laser corresponding to (x1, y1)), so as to ensure that the laser energy is concentrated without deviation;

[0107] Signal acquisition: the detector of the grouping monitoring unit receives the attenuated laser (90% attenuation, only 10% into the detector) and converts the optical signal into an analog voltage signal (power and voltage are linearly positively correlated, such as 300W corresponding to 1.5V); after filtering (eliminate high-frequency noise) and amplification (signal amplitude 0.1-5V), the voltage signal is transmitted to the ADC interface of the data processing module (sampling rate 1kHz, ensuring that the signal is not distorted);

[0108] Linear fitting and calibration verification, including data processing, calibration judgment and reference storage;

[0109] Data processing: the FPGA chip performs least squares fitting on the "power-voltage" data of each laser, such as 4 sets of data (100W / 0.5005V, 200W / 1.001V, etc.) of No. 1 laser fitting P=200U, and calculating the fitting error (≤±0.1W);

[0110] Calibration judgment: calculate the deviation according to ΔP=|preset power-actual power average|, if ΔP>0.1W (such as 100.2W actually under 100W setting of a certain laser), send fine adjustment instruction (current +0.002A) to the drive circuit, re-execute output, measurement and fitting until ΔP≤0.1W;

[0111] Reference storage: the ARM chip stores the ai parameters of each laser in the format of "number-ai-time" to the SD card (such as "1-200W / V-20240510"), ai parameters can only be updated through password verification to prevent unauthorized modification.

[0112] Alignment reference establishment, including position adjustment and reference recording;

[0113] Position adjustment: the automatic displacement table drives the monitoring unit to move, controls the corresponding group of lasers (such as 1-5) to output 300W, and collects voltage signals in real time; if a unit has no signal (misaligned) or amplitude deviation >5% (offset), adjust the X / Y axis of the displacement table (step 0.01mm) until the amplitude deviation ≤5%;

[0114] Reference recording: store the displacement table coordinates corresponding to each laser (such as No. 1 laser X=200mm, Y=200mm) as the alignment reference for subsequent monitoring, avoiding repeated adjustment.

[0115] Second stage: periodic monitoring during printing process—grouping acquisition and deviation calculation;

[0116] This stage deeply integrates the method and equipment printing rhythm, realizes multi-laser full coverage monitoring without affecting efficiency:

[0117] Printing pause and unit movement, including rhythm triggering;

[0118] Rhythm trigger: the data processing module receives the device "layer signal" in real time, sends a "pause instruction" when printing to 100 layers, and the device stops powder laying and scanning; at the same time, a "monitoring instruction" is triggered, the alignment reference is called by the automatic displacement table, and the monitoring unit is moved to the center position of the substrate of the first group of lasers (1-5) with a moving accuracy of ≤0.02mm;

[0119] Power collection and data calculation, including laser light output, signal conversion and mean and deviation calculation;

[0120] Laser light output: the data processing module sends "light output instruction" to the driving circuit, and the first group of lasers outputs standard process power P si (eg 350W), single light output 1s (to ensure that the probe signal is saturated), and each laser repeats the light output 5 times to avoid transient interference;

[0121] Signal conversion: each time the light is output, the probe converts the laser signal into a voltage signal and transmits it to the data processing module, and the single power p ij (eg voltage 1.75V, ai=200W / V, then p j =350W) is calculated through ai parameters;

[0122] Mean and deviation calculation: the power mean is calculated according to P' =∑1 5 p j / 5 (to offset random errors), and ΔP' i =|P si -P' i | (at least 2 pieces of 98% decay pieces);

[0123] Group cycle coverage: after the first group of monitoring is completed, the automatic displacement table moves the monitoring unit to the center of the substrate of the second group of lasers (6-10), and repeats the "light output-collection-calculation" steps to complete the monitoring of all groups (3-5 groups) in turn, ensuring that the 25 lasers are not missed, and the total time of single monitoring is 375s (25×1s×5 times ÷ 5 groups + moving time);

[0124] Third stage: power abnormality closed-loop adjustment-error elimination and precise correction;

[0125] This stage ensures that the power deviation is controllable through the "secondary verification-current fine tuning-reverification" logic, avoiding abnormal influence on the quality of the component:

[0126] Deviation judgment and secondary verification include normal and abnormal situations;

[0127] Normal condition: If all laser ΔP'≤2W, the data processing module sends the "continue printing instruction", the monitoring unit resets to the side of the cabin (without affecting the powder laying), and the equipment resumes printing; at the same time, the monitoring data (P', ΔP', time) of this time are stored in the SD card according to the time stamp;

[0128] Abnormal condition: If a laser ΔP'>2W (such as laser No. 23 ΔP'=2.5W), start the secondary verification - repeat the "light emission - collection - calculation" steps of the laser, to exclude accidental errors caused by dust instantaneous shielding (≤0.1s) and power grid instantaneous fluctuation (≤0.2W); if the secondary verification ΔP'≤2W, it is determined as accidental error, and the printing is resumed; if it is still >2W, it is determined as laser power abnormality, and the adjustment process is triggered;

[0129] Current fine tuning and re-verification, including adjustment amount calculation, adjustment execution and re-verification;

[0130] Adjustment amount calculation: the data processing module calculates the adjustment amount according to the linear coefficient of laser "power - current" (such as 0.5W / mA), such as ΔP'=2.5W, the current needs to be adjusted =2.5W÷0.5W / mA=5mA;

[0131] Adjustment execution: send "current adjustment instruction" to the drive circuit, such as increasing the current of laser No. 23 from 3.5A to 3.505A, and waiting for 10s (to ensure that the power is stable);

[0132] Re-verification: execute the "light emission - collection - calculation" steps of the laser again until ΔP'≤2W, complete the correction and resume printing, and record the adjustment parameters (current before adjustment, adjustment amount, power after adjustment) to the SD card;

[0133] The fourth stage: cyclic monitoring and data tracing, including cyclic execution and data tracing;

[0134] Cyclic execution: repeat the "pause - monitoring - judgment - adjustment" process every 100 layers of printing until the component is printed, to ensure that the laser power is stable during the whole printing period;

[0135] Data tracing: the whole process data (including calibration parameters, monitoring data of every 100 layers, adjustment records) stored in the SD card can be queried through the human-computer interaction module, if there is a quality problem (such as insufficient density) in a certain area of the component, the laser power fluctuation (such as a long-term deviation of 1.8W) of the corresponding layer can be traced back to quickly locate the fault cause, and provide basis for process optimization; the data retention time is ≥5 years, which meets the quality tracing requirements in the fields of aviation, aerospace and shipbuilding;

[0136] In conclusion, the method realizes the "low cost, high precision, no optical path change" monitoring of multiple laser powers in the field of additive manufacturing through full-link closed-loop logic, solves the core pain points of the prior art, and provides core technical support for large-size, high-value component manufacturing.

[0137] The above merely describes preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A multi-laser power monitoring method for the field of additive manufacturing, characterized in that, Comprising the following steps: S1, power calibration in non-processing state: preset power point output, standard power measurement, voltage acquisition and linear relationship fitting are performed on each laser of the multi-laser array (≥16 lasers, matrix arrangement) of the additive manufacturing equipment, the "power-voltage" reference of each laser is established, and the calibration process does not change the original light path of the laser; S2, periodic monitoring during printing process: after printing a preset number of layers, the additive manufacturing equipment is paused, the power data of each laser is collected by the grouping monitoring unit (containing photodetector and attenuator) according to the "1 unit-5 lasers" grouping logic, and the real-time power mean and power deviation are calculated; S3, power abnormality closed-loop adjustment: if the power deviation exceeds the preset threshold, secondary verification is started to exclude accidental errors, and if the deviation is still abnormal, the laser drive current is adjusted until the deviation meets the requirements; the monitoring data is stored throughout the process for quality traceability.

2. The multi-laser power monitoring method for the field of additive manufacturing according to claim 1, characterized in that, The "preset power point" in step S1 includes 100W, 200W, 300W, and 400W, covering the printing power requirements of mainstream materials (titanium alloy, stainless steel, nylon) in additive manufacturing; The "standard power measurement" uses a standard power meter with an accuracy of ±0.1W, and each power point of each laser is measured 5 times, and the mean value is taken as the actual power value.

3. The multi-laser power monitoring method for the field of additive manufacturing according to claim 1, characterized in that, The "linear relationship fitting" in step S1 uses the least squares method to obtain the linear equation P=aiUi (ai is the conversion coefficient) of each laser, and the fitting error is ≤±0.1W; If the calibration deviation calculated by the formula ΔP=|preset power-actual power mean| is >0.1W, the laser needs to be calibrated and the S1 step is repeated until the deviation is ≤0.1W and ai is stored.

4. The multi-laser power monitoring method for the field of additive manufacturing according to claim 1, characterized in that, The "preset number of layers" in step S2 is 100 layers, which is suitable for the printing rhythm of additive manufacturing equipment (100 layers takes 4-6 hours); The photodetector of the "grouping monitoring unit" is an InGaSe detector (wavelength response 800nm-1700nm) or an infrared detector (wavelength response 8-12μm), with a power measurement range of 50W-500W, and a reflective attenuator with an attenuation rate of 80%-90% to avoid damage to the detector.

5. The multi-laser power monitoring method for the field of additive manufacturing according to claim 1, characterized in that, The verification method of "monitoring unit aligned with laser substrate center" in step S2 is: control the corresponding laser to emit laser, observe the voltage signal of the monitoring unit, adjust the automatic displacement table (resolution 0.01mm) to make the voltage signal amplitude deviation of all monitoring units ≤5%, and ensure that the alignment error is ≤0.1mm.

6. The multi-laser power monitoring method for the field of additive manufacturing according to claim 1, characterized in that, The specific way of "power data collection" in step S2 is: controlling the corresponding group of laser to output standard process power P si , single light output 1s (ensure signal stability), repeat collecting power data p ij of each laser for 5 times i (i represents the number of lasers; j represents the test number), calculate the real-time power average according to the formula , and calculate the power deviation according to ΔP' i = |P si - P' i | (i=1, 2, …, 25).

7. The multi-laser power monitoring method for the field of additive manufacturing according to claim 1, characterized in that, The "preset threshold" in step S3 is 2W; "Secondary check" needs to repeat the power data collection and calculation of step S2, if the secondary check deviation is still > 2W, the data processing module according to ΔP i Numerical calculation of current adjustment amount (based on the linear coefficient of laser "power-current"), through the drive circuit to fine-tune the current.

8. The multi-laser power monitoring method for the field of additive manufacturing according to claim 1, characterized in that, The "data storage" in step S3 needs to record the monitoring data of every 100 layers, including the p ij , P' i , ΔP' i and adjustment records. The storage medium is an industrial-grade SD card (capacity ≥ 8 GB), and the data retention time is ≥ 5 years, which meets the quality traceability requirements of the additive manufacturing industry.

Images