Wind speed self-adaptive dynamic adjusting method and system for SLM (selective laser melting) metal additive equipment

Abstract

The invention relates to the technical field of metal additive manufacturing, and provides a wind speed self-adaptive dynamic adjusting method for SLM (selective laser melting) metal additive equipment, which comprises the following steps of: detecting wind speed by arranging sensors at the positions of an air blowing port and an air suction port, and feeding back pressure by a pressure sensor arranged in a printing cabin along the central axis direction of a wind field; global wind field characteristics are captured, wind speed uniformity is used as a control target, fan control and regulation are performed based on feedforward-feedback closed-loop compound control and wind field gradient pressure difference as a constraint, wind speed fluctuation is quickly responded to eliminate steady-state errors, local airflow separation and eddy current caused by layer loss stability due to pressure gradient are avoided, and the stability of a fan is improved. Rapid and reliable regulation and control of wind field uniformity are achieved, and the problem that the forming quality of parts is affected by wind speed fluctuation and attenuation of various wind field structures in the printing process of metal additive manufacturing equipment and interference of a uniform wind field in a printing cabin in the adjusting process is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of metal additive manufacturing, in particular to the technology of uniform wind field, and more specifically to a method and system for adaptive dynamic adjustment of wind speed for SLM metal additive manufacturing equipment. Background Art

[0002] During the process of forming a printed part by a metal additive manufacturing equipment, especially a Selective Laser Melting (SLM) equipment, spatter and soot are often generated. The spatter will greatly affect the forming quality of the printed part, and the soot will block the laser lens, directly affecting the stability of the printing process. Therefore, when designing an SLM equipment, a circulating air flow module (wind field structure) needs to be introduced, and symmetrically distributed air blowing ports and air suction ports are designed in the forming chamber, so that the protective gas can flow stably in the forming chamber and take away the generated spatter and soot, reducing the influence on the metal deposition layer to improve the quality of the component.

[0003] With the continuous expansion of the metal additive manufacturing market, at present, in order to form a stable air flow field, a lot of research and development has been done on the air source and internal structure of the wind field structure of the SLM equipment, including the design of multi-air source pipelines, the design of the rectifying flow channels of the air blowing and suction structure, etc., in order to obtain an air flow parallel to the printing width in the forming chamber. However, it is very difficult to ensure the uniformity of the wind speed at the air blowing ports and the balance of the air flow intensity only through mechanical design, because most of the traditional fan operation frequency converters adopt fixed frequency control. Coupled with the long-term operation of the equipment, the pressure difference of the pipeline filter element increases, the wind speed will decay and there are fluctuations. The operator needs to modify the fan frequency irregularly according to the operation situation of the equipment to reach the expected wind speed. The operation is frequent, greatly increasing the input of the labor cost of the equipment. Moreover, the fluctuating wind speed will also interfere with the uniform wind field, thus affecting the forming quality of the parts, and will also increase the research and development investment in the mechanical structure (such as the design of multi-air source pipelines, the design of the rectifying flow channels of the air blowing and suction structure).

[0004] Therefore, researching and developing a method and device for adaptive dynamic adjustment of wind speed suitable for metal additive manufacturing equipment to facilitate wind field uniformity is of great significance for improving the printing quality, verifying the design effectiveness of the air blowing and suction structure, ensuring the quality of the printed formed parts, and the stability and quality consistency of the printing process. Summary of the Invention

[0005] The present invention aims to solve the problems of wind speed fluctuation, attenuation of various wind field structures during the printing process of metal additive manufacturing equipment, and interference with the uniform wind field in the printing chamber during the adjustment process, which in turn affect the part forming quality. By setting sensors at the air blowing outlet and air suction outlet positions to detect the wind speed and arranging pressure sensors along the central axis of the wind field in the printing chamber for pressure feedback, the global wind field characteristics are captured. Taking the wind speed uniformity as the control target, based on the feedforward-feedback closed-loop compound control and using the wind field gradient pressure difference as a constraint for fan control regulation, the wind speed fluctuation is quickly responded to eliminate the steady-state error, and the local air flow separation and eddy current caused by the pressure gradient leading to layer instability are avoided, realizing the rapid and reliable regulation of the wind field uniformity.

[0006] According to the first aspect of the object of the present invention, a wind speed adaptive dynamic adjustment method for an SLM metal additive manufacturing equipment is proposed, including:

[0007] An anemometer is respectively set at the air blowing outlet and air suction outlet positions of the printing chamber to measure the incoming air lateral wind speed V 进 and the outgoing air lateral wind speed V 出 ;

[0008] A front pressure sensor and a rear pressure sensor are respectively set on both sides of the filter element in the air inlet pipe between the fan and the air blowing outlet of the printing chamber to detect the air inlet filter element pressure difference ΔP filter ;

[0009] Three equally spaced micro pressure difference sensors are arranged along the central axis of the wind field in the printing chamber to respectively monitor the wind field pressure and determine the real-time wind field pressure difference gradient accordingly;

[0010] Based on the air inlet filter element pressure difference ΔP filter A feedforward filter element compensation model is used to correct the target wind speed, and with the wind speed uniformity as the control target, a wind speed uniformity closed-loop control model is established to respond to the wind speed fluctuation and output the wind speed response fan(t);

[0011] Based on the wind field pressure difference gradient, the wind speed response fan(t) is forcibly constrained; and

[0012] Based on the constrained wind speed response fan(t), the fan frequency is dynamically adjusted.

[0013] As an optional implementation manner, the three equally spaced micro pressure difference sensors are arranged with an interval of 10 mm each, and the pressure gradient is calculated using the monitored pressure values, with the unit of Pa / m:

[0014]

[0015] wherein, P3 and P1 respectively represent the pressure detection values of the micro pressure difference sensors downstream and upstream of the wind field.

[0016] As an alternative implementation, the feed-forward filter compensation model is set as:

[0017] Δv = α * ΔP filter + β

[0018] where α and β represent the pressure difference compensation amount and the basic wind speed compensation amount respectively;

[0019] The target wind speed v′ is corrected based on the feed-forward filter compensation model target as:

[0020] v′ target = v target + Δv

[0021] where v target represents the ideal wind speed value set according to the printing process.

[0022] As an alternative implementation, with the wind speed uniformity as the control target, a closed-loop control model for wind speed uniformity is established to respond to wind speed fluctuations and output the wind speed response fan(t), including:

[0023] Combining the feed-forward filter compensation model to establish a closed-loop control model for wind speed uniformity, as follows:

[0024]

[0025] where e represents the deviation between the actual wind speed at the center of the wind field and the ideal wind speed value set according to the printing process; K p and K i represent the proportional control coefficient and the integral control coefficient respectively; k f represents the linear proportional coefficient of the fan frequency and the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

[0026] As an alternative implementation, with the wind speed uniformity as the control target, a closed-loop control model for wind speed uniformity is established to respond to wind speed fluctuations and output the wind speed response fan(t), including:

[0027] Combining the feed-forward filter compensation model to establish a closed-loop control model for wind speed uniformity, as follows:

[0028]

[0029] where e’ represents the deviation between the standard deviation of the actual wind speed at the center of the wind field and the preset wind speed standard deviation threshold; K p and K i represent the proportional control coefficient and the integral control coefficient respectively; k f represents the linear proportional coefficient of the fan frequency and the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

[0030] As an optional implementation manner, the forced constraint on the wind speed response fan(t) based on the wind field pressure difference gradient includes:

[0031] Based on the wind field pressure difference gradient And judge the wind field pressure difference gradient Whether it exceeds the preset pressure difference gradient constraint threshold. If it exceeds the preset pressure difference gradient constraint threshold, the wind speed response fan(t) of the fan is forced to be reduced according to the gradient preset constraint η.

[0032] As an optional implementation manner, the gradient preset constraint is set to 5% - 10%.

[0033] As an optional implementation manner, among the model parameters of the feedforward filter compensation model, the actual measured wind speed of the filter under different blockage degrees is used to obtain the parameter values of the feedforward filter compensation model by recursive least squares fitting, including the pressure difference compensation amount and the basic wind speed compensation amount:

[0034] The basic wind speed compensation amount represents the tiny compensation amount when the filter is initially unblocked;

[0035] The pressure difference compensation amount represents the dynamic compensation amount required for each kPa pressure difference in the filter blockage state.

[0036] As an optional implementation manner, the wind speed adaptive dynamic regulation method further includes:

[0037] According to the preset continuous aging period of the filter, based on the monitoring data within the aging period and the recursive least squares method, dynamically adapt to the change of the filter state, and update the model parameters of the feedforward filter compensation model.

[0038] As an optional implementation manner, the wind speed adaptive dynamic regulation method further includes:

[0039] Monitor the wind speed uniformity index at a preset period for uniformity verification, and trigger an alarm when the wind speed uniformity index exceeds the preset threshold, and record the working condition data;

[0040] According to the alarm, activate the recursive least squares method to dynamically adapt to the change of the filter state, and update the model parameters of the feedforward filter compensation model.

[0041] Compared with the prior art, the significant advantages of the wind speed adaptive dynamic regulation method for SLM metal additive manufacturing equipment of the present invention are:

[0042] (1) In a metal additive manufacturing device, a filter element is used to capture splashing metal particles. However, it will gradually become blocked with the increase of usage time, resulting in a reduction in the cross-sectional area of the air duct, an increase in flow resistance. The present invention converts the pressure difference into a wind speed compensation amount through a linear relationship, combines feedforward control to quickly offset the blockage interference, compensates for the wind speed attenuation in advance, realizes the rapid response of the fan frequency control, and quickly cancels the main interference.

[0043] (2) In the feedforward-feedback closed-loop composite control system designed by the present invention, the feedforward control calculates the compensation wind speed Δ according to the pressure difference ΔP of the filter element filter and corrects the wind speed target value to quickly respond to and offset the actual wind speed drop caused by the blockage of the filter element, and increases the rotation speed to compensate for the resistance interference caused by the blockage during the use of the filter element; at the same time, in the feedback control, starting from the ideal wind speed v targe continuously eliminates the residual error after the feedforward compensation. Thus, the measurable interference (pressure difference of the filter element) can be quickly compensated by the feedforward control, and the wind speed fluctuation deviation caused by the unmodeled disturbances (such as temperature fluctuations, mechanical wear) can be suppressed in cooperation with the feedback control, realizing the uniform control of the wind field;

[0044] (3) In the feedforward-feedback closed-loop collaborative control, the gradient pressure difference in the center of the wind field is further used for constraint to avoid the local airflow separation and vortex generated by the pressure gradient leading to the instability of the laminar flow, and truly realize the rapid and reliable regulation of the wind field uniformity;

[0045] (4) The wind speed adaptive dynamic regulation method for the SLM metal additive equipment proposed by the present invention can update the dynamic model through adaptive learning. According to the aging cycle of the filter element usage, the model parameters of the feedforward filter element compensation model are dynamically updated and calibrated in advance (through RLS), and according to the abnormal state of the wind speed uniformity index, for example, when the preset deviation allowance is exceeded, the model parameters of the feedforward filter element compensation model are activated to be dynamically updated and calibrated by RLS. After the update, the model parameters are directly written into the filter element compensation module to achieve seamless switching. While taking into account the accuracy and real-time performance, the calculation load is reduced and the system oscillation caused by frequent updates is avoided.

[0046] It should be understood that all combinations of the foregoing concepts and additional concepts described in more detail below can be regarded as part of the inventive subject matter of the present disclosure as long as such concepts do not conflict with each other. In addition, all combinations of the claimed subject matter are regarded as part of the inventive subject matter of the present disclosure.

[0047] The foregoing and other aspects, embodiments, and features of the teachings of the present invention can be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention, such as the features and / or beneficial effects of exemplary embodiments, will be apparent in the following description, or will be learned through the practice of the specific embodiments according to the teachings of the present invention. Description of the Drawings

[0048] The accompanying drawings are not intended to be drawn to scale. In the accompanying drawings, each identical or nearly identical component shown in each figure may be denoted by the same reference numeral. For the sake of clarity, not every component is labeled in each figure. Now, embodiments of various aspects of the present invention will be described by way of example and with reference to the accompanying drawings.

[0049] Figure 1 is a schematic diagram of the air duct structure of an SLM metal additive manufacturing device according to an embodiment of the present invention.

[0050] Figure 2 is a schematic diagram of a wind speed adaptive dynamic regulation system of an SLM metal additive device according to an embodiment of the present invention.

[0051] Figure 3 is a schematic diagram of the control principle of a wind speed adaptive dynamic regulation method of an SLM metal additive device according to an embodiment of the present invention.

[0052] Figure 4a and 4b are respectively the frequency and wind speed effect diagrams of the traditional fixed-frequency wind speed frequency according to an embodiment of the present invention.

[0053] Figure 5a and 5b is the frequency and wind speed effect diagram of wind speed adaptive dynamic regulation according to an embodiment of the present invention.

[0054] Figure 6a and 6b are respectively the wind speed distribution cloud comparison diagrams of a 10 mm forming height according to an embodiment of the present invention, where 6a represents the wind speed distribution of a fixed-frequency wind field and 6b represents the wind speed distribution of a variable-frequency wind field. Detailed Embodiments

[0055] To better understand the technical content of the present invention, specific embodiments are hereby given and described in conjunction with the accompanying drawings as follows.

[0056] In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which numerous illustrative embodiments are shown. The embodiments of this disclosure are not necessarily intended to cover all aspects of the present invention. It should be understood that the various concepts and embodiments introduced above, as well as those described in more detail below, can be implemented in any of many ways, because the concepts and embodiments disclosed in the present invention are not limited to any implementation manner. Additionally, some aspects of the present invention can be used alone, or in any suitable combination with other aspects disclosed in the present invention.

[0057] {Embodiment 1}

[0058] Combined with the attached Figures 1-3As shown, a method for wind speed adaptive dynamic regulation of an SLM metal additive manufacturing device according to an embodiment of the present invention includes:

[0059] An anemometer is respectively arranged at the air blowing port and the air suction port of the printing chamber 200 to measure the incoming air lateral wind speed V 进 and the outgoing air lateral wind speed V 出 ;

[0060] A front pressure sensor and a rear pressure sensor are respectively arranged on both sides of the filter element inside the air inlet pipe 300 between the fan 100 and the air blowing port of the printing chamber 200 to detect the air inlet filter element differential pressure ΔP filter ;

[0061] Three equally spaced micro differential pressure sensors are arranged along the central axis of the wind field in the printing chamber to respectively monitor the wind field pressure and determine the real-time wind field differential pressure gradient based on this;

[0062] Based on the air inlet filter element differential pressure ΔP filter The feedforward filter element compensation model is used to correct the target wind speed, and with the wind speed uniformity as the control target, a wind speed uniformity closed-loop control model is established to respond to wind speed fluctuations and output the wind speed response fan(t);

[0063] Based on the wind field differential pressure gradient, the wind speed response fan(t) is forcibly constrained; and

[0064] Based on the constrained wind speed response fan(t), the fan frequency is dynamically regulated.

[0065] As an optional implementation manner, the three equally spaced micro differential pressure sensors are arranged with an interval of 10 mm each, and the pressure gradient is calculated using the monitored pressure values, with the unit of Pa / m:

[0066]

[0067] wherein, P3 and P1 respectively represent the pressure detection values of the micro differential pressure sensors downstream and upstream of the wind field.

[0068] As an optional implementation manner, the feedforward filter element compensation model is set as:

[0069] Δv = α * ΔP filter + β

[0070] wherein, α and β respectively represent the differential pressure compensation amount and the basic wind speed compensation amount;

[0071] Based on the feedforward filter element compensation model, the corrected target wind speed v′ target is:

[0072] v′ target = v target + Δv

[0073] Among them, v target represents the ideal wind speed value set according to the printing process.

[0074] As an alternative implementation, with the wind speed uniformity as the control target, a closed-loop control model for wind speed uniformity is established to respond to wind speed fluctuations and output the wind speed response fan(t), including:

[0075] Combining with the feedforward filter compensation model to establish a closed-loop control model for wind speed uniformity, as follows:

[0076]

[0077] Among them, e represents the deviation between the actual wind speed at the center of the wind field and the ideal wind speed value set according to the printing process; K p and K i respectively represent the proportional control coefficient and the integral control coefficient; k f represents the linear proportional coefficient between the fan frequency and the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

[0078] As an alternative implementation, with the wind speed uniformity as the control target, a closed-loop control model for wind speed uniformity is established to respond to wind speed fluctuations and output the wind speed response fan(t), including:

[0079] Combining with the feedforward filter compensation model to establish a closed-loop control model for wind speed uniformity, as follows:

[0080]

[0081] Among them, e represents the deviation between the standard deviation of the actual wind speed at the center of the wind field and the preset wind speed standard deviation threshold; K p and K i respectively represent the proportional control coefficient and the integral control coefficient; k f represents the linear proportional coefficient between the fan frequency and the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

[0082] As an alternative implementation, based on the wind field pressure difference gradient, the wind speed response fan(t) is forced to be constrained, including:

[0083] Based on the wind field pressure difference gradient and judging the wind field pressure difference gradient whether it exceeds the preset pressure difference gradient constraint threshold, if it exceeds the preset pressure difference gradient constraint threshold, then the wind speed response fan(t) of the fan is forced to be reduced according to the gradient preset constraint η.

[0084] As an alternative implementation, the gradient preset constraint is set to 5% - 10%.

[0085] As an alternative embodiment, among the model parameters of the feedforward filter compensation model, the actual measured wind speed of the filter at different degrees of blockage is used, and the parameter values of the feedforward filter compensation model, including the differential pressure compensation amount and the basic wind speed compensation amount, are obtained by fitting and solving using the recursive least squares method:

[0086] The basic wind speed compensation amount represents the small compensation amount when the filter is initially unblocked;

[0087] The differential pressure compensation amount represents the dynamic compensation amount required for each kPa differential pressure in the blocked state of the filter.

[0088] As an alternative embodiment, the wind speed adaptive dynamic adjustment method further includes:

[0089] According to the preset continuous aging cycle of the filter, based on the monitoring data within the aging cycle and the recursive least squares method, dynamically adapt to the change of the filter state, and update the model parameters of the feedforward filter compensation model.

[0090] As an alternative embodiment, the wind speed adaptive dynamic adjustment method further includes:

[0091] Monitor the wind speed uniformity index at a preset cycle for uniformity verification, and trigger an alarm when the wind speed uniformity index exceeds the preset threshold, and record the working condition data; and according to the alarm, activate the recursive least squares method to dynamically adapt to the change of the filter state, and update the model parameters of the feedforward filter compensation model.

[0092] {Embodiment 2}

[0093] In this embodiment, we further elaborate on the present invention with specific examples.

[0094] I. System initialization and configuration description

[0095] Combined with Figure 1 、 Figure 2 As shown, state detection sensors are arranged in the air inlet pipe 300 of the SLM device (the pipe between the outlet of the fan 100 and the inlet of the printing chamber 200) and inside the printing chamber 300 to sample the air inlet and the pressure state in the process chamber in real time.

[0096] In the embodiment of the present invention, anemometers (Vin, Vout) are respectively arranged at the air blowing port and the air suction port of the printing chamber to measure the lateral air inlet speed V 进 and the lateral air outlet speed V 出 . As a preferred embodiment, high-precision anemometers are respectively arranged at the air blowing port position and the air suction port position to monitor the air flow speeds at the inlet and outlet.

[0097] Meanwhile, a front pressure sensor (P 前 ) and a rear pressure sensor (P 后 ) are respectively arranged on both sides of the filter element in the air inlet pipe between the blower and the air outlet of the printing chamber to detect the differential pressure ΔP filter of the air inlet filter element. The difference in pressure after flowing through the filter element is used as the differential pressure ΔP filter of the air inlet filter element.

[0098] Three equally spaced micro differential pressure sensors (equally spaced, with a spacing of 10 mm) arranged along the central axis of the wind field in the printing chamber are used to respectively monitor the wind field pressure and calculate the pressure gradient, and thus determine the real-time wind field differential pressure gradient (ΔP / Δx, unit: Pa / m):

[0099]

[0100] Among them, P3 and P1 respectively represent the pressure detection values of the micro differential pressure sensors downstream and upstream of the wind field.

[0101] Thus, the global wind field characteristics are captured through the symmetric layout of the sensors to avoid local interference.

[0102] In an alternative embodiment, as Figure 2 shown, a control system 1000 is also provided in the embodiment of the present invention. For example, an embedded control system (the main control of the SLM device) is connected to the blower 100 and each state detection sensor, and the wind speed is adaptively and dynamically controlled according to the feedback of the detection values of the state detection sensors. With the wind speed stability as the control target, the frequency of the blower is controlled to keep the wind speed stable.

[0103] After the system is started, the communication status of the anemometer and the pressure sensor is first detected, and the zero point of the sensor is calibrated. The detection values of the sensor are detected in a windless state to eliminate the inherent deviation of the sensor and ensure the accuracy of the initial data, avoiding systematic errors caused by zero drift.

[0104] As an alternative embodiment, the sampling rate of the sensor is 50 Hz. Within a sampling window length of 1 s, the sliding average filtering is used to determine the sliding filtered mean value of the sampling value as the detection value output of the sensor.

[0105] II. Preset parameter configuration

[0106] Ideal wind speed value: v target is configured, and in this example, it is configured as 2.5 m / s (set according to the material heat capacity and layer thickness);

[0107] Allowable fluctuation range: Δv max = ±0.2 m / s, corresponding to the wind speed uniformity index (i.e., the uniformity coefficient C v ), C v≤3%;

[0108] Pressure gradient threshold: ΔP / Δx ≤ 10 Pa / m (to prevent air flow separation).

[0109] III. Feedforward compensation

[0110] In the present invention, feedforward compensation is used to correct the pressure difference change caused by the aging of the filter element during use, and the resulting wind speed attenuation.

[0111] Feedforward control directly measures the interference source (filter element pressure difference), and based on a pre-established feedforward filter element compensation model, adjusts the control quantity (target wind speed) in advance, so as to perform compensation before the interference affects the actual wind speed, achieve fast response, and avoid the lag of traditional feedback control.

[0112] In an embodiment of the present invention, feedforward control is performed through a feedforward filter element compensation model.

[0113] The control rate design of feedforward compensation is as follows:

[0114] Δv = α * ΔP filter + β

[0115] In the formula, α and β respectively represent the pressure difference compensation amount and the basic wind speed compensation amount.

[0116] And based on this, the target wind speed v' is further corrected target as:

[0117] v' target = v target + Δv

[0118] In the formula, v target represents the ideal wind speed value set according to the printing process.

[0119] In this embodiment, α and β are calibrated by linear regression of the filter element aging experiment. In this example, the settings are as follows:

[0120] α = 0.03, that is, 0.03 m / s of wind speed needs to be compensated for every kPa of pressure difference;

[0121] β = 0.05, as the basic compensation amount, to offset the initial resistance of the filter element.

[0122] It should be understood that in practice, when the initial pressure difference ΔP of the filter element filte r = 0 (there is still resistance when not blocked), β needs to be corrected to a positive value (such as β = 0.05 set in this embodiment) to ensure reasonable compensation logic.

[0123] Thus, the feed-forward filter element compensation model can quickly respond to the measurable change in the differential pressure of the filter element, that is, when the filter element gets blocked and the flow resistance increases, through feed-forward control for quick response to compensate for the attenuation of the wind speed in advance, and feedback for fine adjustment of the wind speed uniformity, with a response speed less than or equal to 50 ms.

[0124] In this embodiment, the process of calibrating the parameters of the feed-forward model through linear regression includes:

[0125] Measure the actual wind speed v at different degrees of filter element blockage (0 - 5 kPa) real , and the test data examples are as follows (taking five groups of data as an example):

[0126] <![CDATA[△P flter (kPa)]]> <![CDATA[v real (m / s)]]> <![CDATA[v target (m / s)]]> 0 2.50 2.50 1 2.45 2.50 2 2.38 2.50 3 2.30 2.50 4 2.21 2.50

[0127] Define the compensated wind speed as follows: Δv = v targe t - v real ;

[0128] Fit the wind speed compensation model: Δv = α * ΔP filter + β;

[0129] Use the recursive least squares algorithm to solve for the parameters α and β:

[0130]

[0131] After substituting the test data at the degree of filter element blockage, solve to obtain the calibrated values of the parameters α and β.

[0132] It should be understood that the parameter α reflects the sensitivity of the filter element blockage, that is, the wind speed loss caused by unit differential pressure. The parameter β represents the compensation for the initial resistance of the filter element, that is, the differential pressure ΔP filter = 0).

[0133] IV. Feedback control

[0134] In this step, taking the design of a PI controller as an example, combined with the differential pressure ΔP of the inlet filter element filter Use the feed-forward filter element compensation model to correct the target wind speed, and taking the wind speed uniformity as the control target, establish a closed-loop control model for wind speed uniformity, which is the closed-loop of the Xi'an wind speed uniformity control to respond to the wind speed fluctuation and output the wind speed response fan(t).

[0135] As one of the implementation manners, the establishment of a closed-loop control model for wind speed uniformity with the wind speed uniformity as the control target, responding to the wind speed fluctuation and outputting the wind speed response fan(t) includes:

[0136] Combine the feed-forward filter element compensation model to establish a closed-loop control model for wind speed uniformity as follows:

[0137]

[0138] where e’ represents the deviation between the actual wind speed standard deviation at the wind field center and the preset wind speed standard deviation threshold; K p and K i represent the proportional control coefficient and the integral control coefficient respectively; k f represents the linear proportional coefficient of the fan frequency to the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

[0139] In this example, the control law is configured to control the wind speed uniformity coefficient C v (characterized by the wind speed standard deviation) at 3%, that is:

[0140] f fan (t) = 0.8·(C v - 3%) + 0.2∫0 t (C v - 3%)dt

[0141] Closed-loop control parameter selection:

[0142] The proportional term dominates K p = 0.8, for rapid response to wind speed fluctuations;

[0143] The integral term control coefficient K i = 0.2, to eliminate the steady-state error (the integral time constant T i = K p / K i = 4 si = 4s).

[0144] As another implementation, with the wind speed uniformity as the control target, a closed-loop control model for wind speed uniformity is established to respond to wind speed fluctuations and output the wind speed response fan(t), including:

[0145] Combining with the feedforward filter compensation model to establish a closed-loop control model for wind speed uniformity, as follows:

[0146]

[0147] where e represents the deviation between the actual wind speed at the wind field center and the ideal wind speed value set according to the printing process; K p and K i represent the proportional control coefficient and the integral control coefficient respectively; k f represents the linear proportional coefficient of the fan frequency to the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

[0148] As an optional implementation, based on the wind field pressure difference gradient, the wind speed response fan(t) is forcibly constrained, including:

[0149] Based on the previously calculated wind field pressure difference gradient And determine the pressure difference gradient of the wind field Check whether it exceeds the preset pressure difference gradient constraint threshold. If it exceeds the preset pressure difference gradient constraint threshold, forcefully reduce the wind speed response fan(t) of the wind turbine according to the gradient preset constraint η.

[0150] For example, according to the previously configured gradient constraint threshold of 10 Pa / m, when the calculated pressure difference gradient of the wind field exceeds 10 Pa / m, constrain and reduce the wind turbine frequency: η * fan(t). Among them, η is pre-configured to be 5% - 10%, that is, forcefully reduce the wind turbine frequency by 5% - 10% to prevent airflow separation.

[0151] Furthermore, according to the linear relationship between the frequency and wind speed of wind turbine control, dynamically adjust the wind turbine frequency based on the constrained wind speed response fan(t).

[0152] V. Performance verification

[0153] In this embodiment, monitor the wind speed uniformity index according to a preset period for uniformity verification, and trigger an alarm and record the operating condition data when the wind speed uniformity index exceeds the preset threshold.

[0154] Among them, the system is configured to calculate the wind speed uniformity coefficient C according to a set period v And conduct uniformity verification.

[0155] For example, calculate the wind speed uniformity coefficient C once every 5 minutes v :

[0156]

[0157] In the formula, N represents the total number of wind field center wind speed samplings within a period (such as 5 minutes), represents the average value of all wind speed values obtained from the wind field center wind speed sampling within a period (such as 5 minutes), v i represents the wind speed value of the i-th sampling, i = 1, 2, 3,..., N.

[0158] If C v > 5%, then trigger an alarm and record the operating condition data.

[0159] VI. Feedforward model adaptive learning and model update

[0160] In this embodiment, when C v > 5%, that is, when abnormal fluctuations caused by long-term drifts such as filter element aging and powder type change occur, the system controls to activate the parameter adaptive update algorithm (RLS algorithm), and uses the recursive least squares method to dynamically adapt to the change of the filter element state and update the model parameters of the feedforward filter element compensation model.

[0161] Therefore, when the filter element ages or different models of filter elements are replaced / the powder type is changed, by dynamically updating the feedforward parameters α and β of the flow resistance characteristics of the reaction filter element, the change of the filter element characteristics can be continuously tracked, and the compensation accuracy can be maintained. At the same time, the sampling linear model has a low calculation amount (only multiplication and addition), is suitable for embedded real-time control, and can be applied to different filter element types, only by recalibrating the parameters.

[0162] For example, the set limiting compensation equation is:

[0163] Δv = α·ΔP filter +β = φ T θ;

[0164] φ = [ΔP filter ,1] T ;

[0165] θ[α,β] T ;

[0166] where φ is the eigenvector and θ is the parameter to be estimated.

[0167] Parameter update is performed through error calculation and covariance matrix update, and the specific design is as follows:

[0168] Error calculation:

[0169] where, y k is the measured Δv (compensation wind speed error) at the k-th step.

[0170] Covariance matrix update:

[0171] Further parameter update: θ k = θ k-1 +P k ψ k e k ;

[0172] where, λ represents the forgetting factor (0 < λ ≤ 1), which is used to reduce the weight of old data, and usually takes the value of λ = 0.95 - 0.99 (when λ = 1, it means complete memory, which is applicable to a steady-state environment, such as when the filter element model is fixed and the aging is slow); P k represents the covariance matrix, and the initial value is set to P0 = δ -1 I (δ is a small constant, such as 0.01, with a low confidence in the initial parameters and allowing rapid update).

[0173] As an optional embodiment, the value of the forgetting factor λ is as follows:

[0174] Λ = 1 - (parameter change speed / data sampling rate);

[0175] For example, if the filter element characteristics change by 5% per hour (which can be pre-tested) and the sampling interval is 11 seconds, then λ≈1 - 5% / 3600≈0.9999.

[0176] During the process of recalibrating the aforementioned parameters, for a relatively large initial value of the covariance matrix, for example, P0 = 100I indicates a high uncertainty in the initial parameters, which accelerates the initial convergence. As data accumulates, the covariance matrix gradually shrinks, the parameter update amplitude decreases, and the system tends to be stable.

[0177] In another embodiment, the wind speed adaptive dynamic regulation method proposed by the present invention further includes:

[0178] According to a preset continuous aging cycle of the filter element, for example, the RLS algorithm is dynamically activated every 4h, and based on the monitoring data within the aging cycle and the recursive least squares method, it dynamically adapts to the change of the filter element state and updates the model parameters of the feedforward filter element compensation model.

[0179] {Embodiment 3}

[0180] In this embodiment, in combination with Attached Figure 4a 、 4b 、5a, 5b and Figure 6a 、 6b shown, the wind speed adaptive dynamic regulation method proposed by the present invention and the effect of traditional fixed-frequency control of the fan are further elaborated and compared.

[0181] Before the wind speed adaptive dynamic regulation method proposed by the present invention is adopted, the regulation effect diagrams of the fan frequency and wind speed as shown in Figure 4a 、 4b are obtained. As the operation time of the equipment accumulates, the pressure difference of the system filter element becomes larger, the wind speed attenuation and fluctuation are larger, and it is difficult to meet the wind speed maintenance and stability requirements for equipment printing.

[0182] After the wind speed adaptive dynamic regulation method proposed by the present invention is adopted, the wind speed adaptive dynamic regulation effect diagrams as shown in Figure 5a 、 5b are obtained. Combining the comparison of the relationship between the fan frequency and wind speed regulation shown in the figure, as the operation time of the equipment accumulates, the system adaptively adjusts the real-time changing frequency dynamically, reaches the target wind speed and maintains the wind speed stability.

[0183] Furthermore, after adopting the wind speed adaptive dynamic regulation method, we conduct a wind field simulation and obtain as shown in Figure 6a 、 6bCompare the velocity distribution cloud at a height of 10 mm from the cabin body substrate. The substrate area is evenly divided into 5×5 regions, and the average velocity values in each region are extracted respectively. During the printing process, the black smoke that appears is generally between 10 mm and 20 mm from the substrate. From the velocity distribution cloud map at a height of 10 mm from the substrate, from top to bottom, the wind speed generally shows an increasing state, and there is a sharp decrease in the wind speed in the top row region (as shown in Figure 6). In addition, from left to right, the wind speed generally shows a fluctuating situation.

[0184] According to the above analysis, the longitudinal distribution difference of the average wind speed in the entire air inlet section of the fixed-frequency wind field structure is too large, and the transverse distribution difference is slightly smaller (as Figure 6a ). In addition, the wind speed at a height of 10 mm from the substrate at the air suction port of the fixed-frequency wind field structure is too low. Under the action of the wind field of the fixed-frequency wind field structure, a large amount of soot will accumulate in the area of the substrate near the air suction port. The uneven wind speed across the entire printing area will also cause soot accumulation during large-layer-thickness printing. Compared with the wind speed distribution of the fixed-frequency wind field structure (as Figure 6b ), the wind speed distribution of the variable-frequency wind field structure is more uniform, and the frequency does not need to be changed with the accumulation of the operation time of the device. The system will switch to the required wind speed at any time according to the need and maintain the wind speed stable. Thus, through the wind speed adaptive dynamic regulation method proposed by the present invention, through the composite control of feedforward-closed-loop feedback, combined with gradient pressure difference for constraint to avoid layer instability caused by pressure gradient, real-time variable frequency control of the fan is achieved, truly realizing fast and reliable regulation of the wind field uniformity, helping to balance the air flow entering the channel opening, improving the wind speed uniformity and air flow intensity of the printing area, helping to effectively remove soot at various positions during the processing, and improving the quality control and consistency of printing forming.

[0185] Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Those with ordinary knowledge in the technical field to which the present invention pertains can make various modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to what is defined by the claims.

Claims

1. A method for wind speed adaptive dynamic adjustment of an SLM metal additive manufacturing device, characterized in that, Including: An anemometer is respectively arranged at the air blowing port and the air suction port of the printing cabin to measure the crosswise air inlet speed V 进 and the crosswise air outlet speed V 出 ; A front pressure sensor and a rear pressure sensor are respectively arranged on both sides of the filter element in the air inlet pipeline between the fan and the air outlet of the printing cabin to detect the differential pressure ΔP of the air inlet filter element filter ; Three equally spaced micro differential pressure sensors arranged along the central axis of the wind field in the printing chamber, which respectively monitor the wind field pressure and determine the real-time wind field differential pressure gradient based on this; Based on the differential pressure ΔP of the intake air filter element filter Adopt a feedforward filter element compensation model to correct the target wind speed, and take the wind speed uniformity as the control target to establish a closed-loop control model of wind speed uniformity, respond to the wind speed fluctuation and output the wind speed response fan(t); Based on the wind field differential pressure gradient, forcibly constrain the wind speed response fan(t); And Dynamically adjust the fan frequency based on the constrained wind speed response fan(t).

2. The wind speed adaptive dynamic adjustment method for the SLM metal additive manufacturing equipment according to claim 1, wherein The three equally spaced micro differential pressure sensors are arranged with an interval of 10 mm each, and the pressure gradient is calculated using the monitored pressure values, with the unit of Pa / m: Among them, P3 and P1 respectively represent the pressure detection values of the micro differential pressure sensors downstream and upstream of the wind field.

3. The wind speed adaptive dynamic regulation method for SLM metal additive manufacturing equipment according to claim 1, characterized in that The feedforward filter compensation model is set as: Δv = α * ΔP filter + β Among them, α and β respectively represent the differential pressure compensation amount and the basic wind speed compensation amount; Modify the target wind speed v′ based on the feedforward filter element compensation model target It is: v′ target = v target + Δv Among them, v target represents the ideal wind speed value set according to the printing process.

4. The wind speed adaptive dynamic adjustment method for the SLM metal additive manufacturing equipment according to claim 3, wherein Taking the wind speed uniformity as the control target, establishing a closed-loop control model for wind speed uniformity, responding to the wind speed fluctuation and outputting the wind speed response fan(t), including: Establishing a closed-loop control model for wind speed uniformity in combination with the feedforward filter compensation model, as follows: where e represents the deviation between the actual wind speed at the center of the wind field and the ideal wind speed value set according to the printing process; K p and K i represent the proportional control coefficient and the integral control coefficient respectively; k f represents the linear proportional coefficient of the fan frequency to the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

5. The wind speed adaptive dynamic adjustment method for the SLM metal additive manufacturing equipment according to claim 3, wherein Taking the wind speed uniformity as the control target, establishing a closed-loop control model for wind speed uniformity, responding to the wind speed fluctuation and outputting the wind speed response fan(t), including: Establishing a closed-loop control model for wind speed uniformity in combination with the feedforward filter compensation model, as follows: where e’ represents the deviation between the actual wind speed standard deviation at the wind field center and the preset wind speed standard deviation threshold; K p and K i respectively represent the proportional control coefficient and the integral control coefficient; k f represents the linear proportional coefficient of the fan frequency to the inlet wind speed, reflecting the efficiency of the conversion of the rotational kinetic energy of the motor into the kinetic energy of the air flow.

6. The wind speed adaptive dynamic adjustment method for SLM metal additive manufacturing equipment according to claim 4 or 5, characterized in that Based on the wind field differential pressure gradient, forcibly constrain the wind speed response fan(t), including: Based on the wind field pressure difference gradient And judge the wind field pressure difference gradient Whether it exceeds the preset pressure difference gradient constraint threshold. If it exceeds the preset pressure difference gradient constraint threshold, the wind speed response fan(t) of the wind turbine is forced to decrease according to the gradient preset constraint η.

7. The wind speed adaptive dynamic adjustment method for SLM metal additive manufacturing equipment according to claim 6, characterized in that, The gradient preset constraint is set to 5% - 10%.

8. The wind speed adaptive dynamic adjustment method for SLM metal additive manufacturing equipment according to any one of claims 1-7, characterized in that Among the model parameters of the feedforward filter compensation model, the actual measured wind speeds of the filter under different clogging degrees are used, and the parameter values of the feedforward filter compensation model, including the differential pressure compensation amount and the basic wind speed compensation amount, are obtained by recursive least squares fitting; The basic wind speed compensation amount represents the small compensation amount when the filter is initially unclogged; The differential pressure compensation amount represents the dynamic compensation amount required for each kPa differential pressure in the clogged state of the filter.

9. The wind speed adaptive dynamic adjustment method for the SLM metal additive manufacturing equipment according to any one of claims 1-7, characterized in that The wind speed adaptive dynamic adjustment method further includes: According to the preset continuous aging cycle of the filter, based on the monitoring data during the aging cycle and the recursive least squares method, dynamically adapt to the change of the filter state and update the model parameters of the feedforward filter compensation model.

10. The wind speed adaptive dynamic adjustment method for SLM metal additive manufacturing equipment according to any one of claims 1-7, characterized in that The wind speed adaptive dynamic adjustment method further includes: Monitoring the wind speed uniformity index at a preset cycle for uniformity verification, and triggering an alarm when the wind speed uniformity index exceeds the preset threshold, and recording the working condition data; According to the alarm, activate the recursive least squares method to dynamically adapt to the change of the filter state and update the model parameters of the feedforward filter compensation model.

Images