A method for optimizing process parameters of a metal powder injection molding

Abstract

The application provides a metal powder injection molding process parameter optimization method, and relates to the technical field of industrial production process control and optimization. The method extracts the standard deviation of particle size distribution and the peak particle size as the powder flowability characteristic parameters by segmenting the particle size distribution curve of the current batch of powder for statistical analysis. The historical filling resistance data of the mold cavity are spatially decoupled according to the thread root region and the polished rod section region to generate a mold cavity partition resistance characteristic map containing partition resistance conflict amount. Then, the powder particle size distribution deviation is converted into an injection pressure correction amount, and the partition resistance conflict amount is converted into a partition compensation amount of the holding pressure time, so as to realize the differential parameter configuration of the thread root and the polished rod section. In this way, the problem of full-mold-cavity density uniformity loss caused by the coupling of powder batch microscopic characteristic fluctuation and mold cavity space filling resistance heterogeneity can be effectively solved, and the demand conflict between the thread root densification and the polished rod section anti-flashing can be coordinated.

Description

Technical Field

[0001] This invention relates to the field of industrial production process control and optimization technology, and more specifically, to a method for optimizing process parameters of metal powder injection molding. Background Technology

[0002] Metal Powder Injection Molding (MIM) technology loads a high volume fraction of metal powder into a polymer binder to form a feedstock with suitable rheological properties. Through injection molding, debinding, and sintering, it enables the mass production of near-net-shape metal parts with complex geometries. In the field of precision medical devices, this technology is widely used to manufacture products with stringent biocompatibility and mechanical performance requirements, such as titanium alloy orthopedic implant screws. The molding quality of these parts, especially the density uniformity of the thread root and the smooth section, directly affects the fatigue life and clinical safety of the implant. Process parameter control, as the core of the MIM molding process, essentially involves coordinating the setting of multi-dimensional variables such as injection pressure, holding time, mold temperature, and cooling rate. This ensures that the feedstock melt, under non-isothermal and non-Newtonian flow conditions, fully fills the mold cavity with complex geometric boundaries and significant differences in spatial resistance distribution, and effectively compensates for material cooling shrinkage during the holding pressure stage to suppress molding defects such as micropores, flash, and dimensional deviations. Early process parameter setting relied heavily on empirical trial molding and orthogonal experimental design. In recent years, with the introduction of mold flow analysis software and statistical process control (SPC) methods, parameter optimization strategies based on simulation prediction and batch statistics have become the industry mainstream. These methods typically establish a statistical mapping model between global parameters such as injection pressure and holding time and key product quality indicators based on the process window of historically stable batches, and implement unified parameter compensation through offline analysis of the mold cavity pressure curve. However, these methods exhibit significant limitations when dealing with the two inherent uncertainties in MIM production.

[0003] The core flaw in existing technical solutions lies in their failure to effectively decouple the microscopic characteristic fluctuations of powder batches from the heterogeneity of mold cavity filling resistance, thus falling into a control dilemma of "global parameter conflicting regions." Specifically, on the one hand, the particle size distribution of different batches of feed powder of the same grade often fluctuates due to the perturbations of the atomization process, causing deviations in the powder's bulk density, apparent viscosity, and melt flowability. Existing offline detection and batch mean comparison methods can only eliminate batches with out-of-tolerance values, but cannot quantify the real-time statistical characteristics of particle size distribution and map them into online corrections for injection pressure, resulting in a mismatch between the flow front velocity and pressure transmission characteristics when fixed parameters are applied to new powder batches. On the other hand, for multi-feature common mold parts such as orthopedic screws, the local filling resistance at the thread root is significantly higher than that of the smooth section due to the rapid change in the flow channel cross-section and the deep and narrow cavity effect, resulting in drastically different requirements for holding pressure and compensation time and pressure gradient. Existing methods apply a single holding pressure curve to the entire mold cavity, failing to provide differentiated control based on resistance variations and density requirements in different areas. This often results in micropores at the thread root due to insufficient feeding, while flash forms on the smooth section due to excessive holding pressure, leading to unpredictable fluctuations in the scrap rate per batch. These limitations are further amplified under batch consistency requirements for medical-grade products, constituting a key bottleneck for the in-depth application of MIM technology in high-end implantable devices. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for optimizing metal powder injection molding process parameters. This method can, to some extent, solve the problem of insufficient control precision of process parameters over the product quality distribution within a batch, resulting in disordered fluctuations in the micropore rate at the thread root and the flash amount in the smooth rod section within a single batch. Consequently, it is impossible to achieve coordinated assurance of uniformity of the entire mold cavity density and batch stability through a unified parameter set.

[0005] According to one aspect of the present invention, a method for optimizing process parameters of metal powder injection molding is provided, comprising:

[0006] The particle size distribution curve of the current batch of powder is segmented and statistically processed according to the particle size interval. The peak particle size and the standard deviation of the particle size distribution curve are extracted as the flowability characteristic parameters of the current batch of powder.

[0007] The historical filling resistance data of the mold cavity is divided into regions according to the spatial location of the thread root region and the smooth rod section region. The statistical characteristic of the filling resistance of each region is calculated to generate a mold cavity partition resistance characteristic map.

[0008] The pressure correction amount is calculated based on the powder flowability characteristic parameters and the injection pressure reference value, and the holding time partition compensation amount is calculated based on the difference in filling resistance of each region in the mold cavity partition resistance characteristic map. The parameter adjustment command containing the pressure correction amount and the holding time partition compensation amount is output.

[0009] The parameters of the injection molding equipment are configured according to the parameter adjustment instructions. Differentiated holding time compensation is applied to the thread root region and the smooth rod section region respectively. The injection pressure is adjusted according to the pressure correction amount to complete the process parameter optimization configuration for the current batch.

[0010] Furthermore, the current batch powder flowability characteristic parameters include peak particle size, particle size distribution standard deviation, and historical stable batch baseline corresponding to the current batch;

[0011] The peak particle size is determined based on the proportion of particle volume distribution or particle size probability density in each particle size range.

[0012] The standard deviation of particle size distribution is determined based on the representative particle size of each particle size interval, the proportion of normalized particle volume distribution, and the width of the particle size interval.

[0013] The historical stable batch baseline includes the peak particle size stability center value, the particle size distribution standard deviation stability center value, the peak particle size stability scale, and the particle size distribution standard deviation stability scale.

[0014] Furthermore, the historical filling resistance data of the mold cavity has a region attribution label, which is determined based on the three-dimensional geometric structure information of the mold cavity and the spatial coordinates of the filling resistance measurement points;

[0015] The three-dimensional geometric structure information of the mold cavity includes the axial starting position of the thread segment, the axial ending position of the thread segment, the radial position of the deepest point of the thread profile, the thread profile height, the axial range of the polished rod segment, and the radial position of the inner wall surface of the polished rod segment.

[0016] When the axial coordinate of the measuring point is located between the axial start position and the axial end position of the thread segment, and the radial coordinate of the measuring point is located within the root neighborhood defined by the thread profile height, the measuring point is marked as belonging to the thread root region.

[0017] When the axial coordinate of the measuring point is within the axial range of the smooth rod section, and the radial coordinate of the measuring point corresponds to the radial position of the inner wall surface of the smooth rod section, the measuring point is marked as belonging to the smooth rod section area.

[0018] Furthermore, the cavity partition resistance characteristic map includes the regional equivalent filling resistance and partition resistance conflict amount of each region;

[0019] The equivalent filling resistance of the region is determined based on the upstream pressure of the region, the local cavity pressure of the region, the equivalent volumetric flow rate of the region, and the effective filling time of the region.

[0020] The partition resistance conflict amount is determined based on the robust center value of the equivalent filling resistance of the region, the center value of the resistance of the entire mold cavity, the regional spatial resistance gradient, and the center value of the stable spatial resistance gradient of the entire mold cavity.

[0021] Furthermore, the calculation of the equivalent filling resistance of the region and the conflict amount of the partition resistance is shown in the following formula:

[0022]

[0023]

[0024] in, Indicates the first In the first historical production cycle The equivalent filling resistance of each region; Indicates the first The effective fill time period for each area; This indicates upstream pressure in the region; Indicates the localized cavity pressure in the region; Indicates the equivalent volumetric flow rate of the region; Indicates the signal reliability weight; Indicates the first The amount of regional resistance conflict; Indicates the first The regional equivalent filling resistance robust center value of each region; This represents the center value of the total mold cavity resistance; Indicates the first The regional spatial resistance gradient of each region; This represents the center value of the drag gradient in the stable space of the entire mold cavity; and These represent the regional resistance stability scale and the spatial resistance gradient stability scale, respectively. and is a dimensionless fusion coefficient.

[0025] Furthermore, the pressure correction amount is determined based on the standard deviation of the particle size distribution of the current batch, the peak particle size of the current batch, the historical stable batch baseline, the powder detection reliability, and the pressure operating boundary, as shown in the following formula:

[0026]

[0027] in, Indicates the injection pressure correction amount; Indicates the baseline value of injection pressure; and Indicates the pressure operating boundary; Indicates the reliability of powder detection; and These represent the sensitivity coefficients of the deviation of the standard deviation of particle size distribution and the deviation of peak particle size to the correction of injection pressure, respectively. This indicates the standard deviation of the particle size distribution in the current batch; This represents the stable center value of the standard deviation of particle size distribution; This represents the stability scale of the standard deviation of particle size distribution; Indicates the peak particle size of the current batch; Indicates the peak particle size stability center value; This indicates the stable scale of the peak particle size.

[0028] Furthermore, the pressure holding time zoning compensation amount is determined based on the zoning resistance conflict amount, the reliability of the zoning map, the zoning pressure holding time reference value, and the zoning pressure holding time operating boundary, as shown in the following formula:

[0029]

[0030] in, Indicates the first Pressure holding time zone compensation amount for each area; Indicates the first Baseline values ​​for pressure holding time in each region; and Indicates the first Operating boundaries for pressure holding time in each region; Indicates the first Reliability of the map data for each region; Indicates the first The sensitivity coefficient of the resistance conflict amount in each region to the pressure holding time compensation. Indicates the first The amount of regional resistance conflict; Indicates the first Quality feedback residuals for each region; This represents the conversion factor from quality feedback residual to pressure holding time compensation.

[0031] According to another aspect of the present invention, a system for optimizing process parameters of metal powder injection molding is provided, comprising:

[0032] The powder flowability feature extraction module is used to perform segmented statistical processing on the particle size distribution curve of the current batch of powder according to the particle size range, and extract the peak particle size and the standard deviation of particle size distribution as powder flowability feature parameters.

[0033] The mold cavity partition resistance feature map generation module is used to divide the historical filling resistance data of the mold cavity into regions according to the spatial location of the thread root region and the smooth rod section region, calculate the equivalent filling resistance and partition resistance conflict of each region, and generate the mold cavity partition resistance feature map.

[0034] The parameter compensation calculation and instruction generation module is used to correlate powder flowability characteristic parameters with injection pressure reference value to obtain pressure correction amount, convert zone resistance conflict amount into holding time zone compensation amount, and output parameter adjustment instruction containing the pressure correction amount and holding time zone compensation amount;

[0035] The partition parameter execution and feedback module is used to configure the partition parameters of the injection molding equipment according to the parameter adjustment instructions, apply differentiated holding time compensation and injection pressure correction to the thread root region and the smooth rod section region respectively within the same injection cycle, and collect actual execution data and quality inspection results to form a closed-loop feedback.

[0036] Compared with existing technologies, the method for optimizing metal powder injection molding process parameters provided by this invention extracts the standard deviation and peak particle size of the particle size distribution curve of the current batch of powder as characteristic parameters of powder flowability by segmenting and statistically analyzing the particle size distribution curve. Furthermore, it spatially decouples the historical filling resistance data of the mold cavity by thread root region and smooth rod region to generate a mold cavity partition resistance characteristic map containing partition resistance conflict. This converts powder particle size distribution deviation into injection pressure correction and partition resistance conflict into holding time partition compensation, achieving differentiated parameter configuration for the thread root and smooth rod regions. This effectively solves the problem of uncontrolled uniformity of mold cavity density caused by the coupling of micro-characteristic fluctuations in powder batches and heterogeneity in mold cavity filling resistance. It helps to coordinate the conflicting requirements of thread root densification and shrinkage compensation and smooth rod anti-flash requirements, thereby significantly reducing the disorderly fluctuations in microporosity and flash within a single batch, improving batch consistency and product yield, and meeting the stringent quality stability requirements of medical-grade implants. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0038] Figure 1 This is a flowchart of a method for optimizing metal powder injection molding process parameters according to an embodiment of the present invention.

[0039] Figure 2This is a schematic diagram of the orthopedic implant screw cavity partitioning and sensor layout, illustrating the optimization method for metal powder injection molding process parameters according to an embodiment of the present invention.

[0040] Figure 3 This is a timing diagram of the coordinated control of injection pressure and zone holding time in a method for optimizing metal powder injection molding process parameters according to an embodiment of the present invention. Detailed Implementation

[0041] Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0042] Figure 1 This is a flowchart illustrating a method for optimizing metal powder injection molding process parameters according to an embodiment of the present invention. Figure 1 As shown, the optimization method for metal powder injection molding process parameters includes:

[0043] S1. First, obtain the particle size distribution detection data of the feed powder for the current batch of titanium alloy injection molding. Perform segmented statistical processing on the particle size distribution detection data according to the particle size interval. Then, extract the peak particle size and the standard deviation of particle size distribution from the segmented statistical results. Use the peak particle size and the standard deviation of particle size distribution as the flowability characteristic parameters of the current batch of powder and output them to the subsequent injection pressure correction process.

[0044] First, the particle size distribution curve of the current batch of powder is read from the particle size detection device or process detection database. This curve includes multiple particle size detection channels, the particle size range corresponding to each channel, and the proportion of particle volume distribution within that range. When the detection device outputs a continuous particle size distribution curve, it is divided into multiple continuous particle size intervals based on the device's calibration resolution and the historical stable particle size coverage of the current powder grade. Each interval includes the lower limit particle size, upper limit particle size, representative particle size, and proportion. Subsequently, the detection data from different sampling locations are processed to a unified standard. These locations can include the upper part of the silo, the middle part of the silo, the lower part of the silo, and the end of the feeding line. If multiple duplicate detection records exist for the same batch, each record is converted to the same particle size interval format, allowing data from different sampling locations and different detection counts to be statistically analyzed under the same particle size coordinate system.

[0045] Once the particle size interval format is prepared, an effective signal verification is performed for each interval. Specifically, the blank sample response, standard particle calibration record, scattering residual, or image overlap rate record of the detection device are read and matched with the current detection curve. For unstable responses caused by dispersant background, lens noise, or particle agglomeration, the corresponding particle size interval is not directly deleted. Instead, the reliability weight of the interval data is reduced based on repeatability and the calibration status of the detection device. Subsequently, using the total volume of effectively detected particles or the total amount of effectively detected signals in the current batch as a normalization benchmark, the proportion of each particle size interval is converted into a normalized proportion, so that the sum of the proportions of all particle size intervals is one. This normalization benchmark is derived from the actual detection data of the current batch and is not a fixed value given manually, thus adapting to changes in different sampling volumes and detection signal intensities.

[0046] Furthermore, after the proportions of each particle size interval are normalized, all particle size intervals are traversed, and the probability density of each interval is calculated. If the widths of all particle size intervals are the same, the proportion of each interval can be directly used to compare the peak positions of the particle size distribution. If the widths of each interval are different, the proportion of each interval is divided by the interval width to obtain the probability density per unit particle size width, and then the probability densities of each interval are compared. If the probability density of a certain interval is the highest among all intervals, and there are effective adjacent intervals on both sides of the interval, a local quadratic curve is fitted using the probability density data of the interval and its two adjacent intervals, and the particle size corresponding to the vertex of the fitted curve is taken as the peak particle size of the current batch of powder. If the interval is located at the edge of the detection range, or the data reliability of adjacent intervals is insufficient, the representative particle size of the interval is taken as the peak particle size of the current batch of powder, and the reliability information of the peak particle size is recorded in the output results.

[0047] Further, the standard deviation of the particle size distribution of the current batch of powder is calculated. Specifically, firstly, the average particle size of the current batch of powder is calculated based on the normalized proportion of each particle size interval and the representative particle size of each interval; then, the deviation between the representative particle size and the average particle size of each particle size interval is calculated, and the squared values ​​of the deviations are weighted and accumulated according to the normalized proportion of the particle size interval to obtain the discrete contribution between intervals; when the particle size interval widths are not the same, the discrete contribution caused by the particle size variation within each particle size interval is further added to the statistical results, and the discrete contribution within the interval is determined by the width of the particle size interval; when the detection device has standard particle calibration records, the contribution of the measurement uncertainty of the detection device to the particle size distribution width is also deducted according to the calibration records. The standard deviation of the particle size distribution obtained after the above processing is in micrometers and is used to characterize the degree of mixing of coarse and fine particles and the degree of fluctuation in powder flowability of the current batch of powder.

[0048] After the peak particle size and particle size distribution standard deviation are calculated, the historical stable batch baseline corresponding to the current powder grade, binder system, product specification, and testing device is read. The historical stable batch baseline includes the stable center value of the peak particle size, the stable center value of the particle size distribution standard deviation, the particle size detection repeatability scale, and process parameter records corresponding to qualified product batches. This historical stable batch baseline is formed by production batches that have continuously passed size detection, density detection, microporosity detection, and appearance detection, and is updated as new qualified batch data enters the database. Subsequently, the peak particle size, particle size distribution standard deviation, detection reliability, historical stable batch baseline version, and batch identifier of the current batch are encapsulated into a powder flowability characteristic data package, and this powder flowability characteristic data package is sent to the injection pressure correction calculation process in S3.

[0049] For example, a current batch of powder is divided into five consecutive particle size intervals, with representative particle sizes of 10 μm, 15 μm, 20 μm, 25 μm, and 30 μm, corresponding to effective volume distribution percentages of 0.05, 0.20, 0.45, 0.25, and 0.05, respectively. First, the sum of each percentage is confirmed to be one. Then, the average particle size is calculated to be approximately 20.25 μm. The standard deviation of the particle size distribution is obtained based on the dispersion of the representative particle size of each interval relative to the average particle size. Since the 20 μm interval corresponds to the highest particle size distribution percentage, the peak particle size is determined to be approximately 20 μm. Alternatively, a finer peak particle size can be obtained through local fitting when the data for adjacent intervals is reliable. The input to this example is the current batch's segmented particle size percentage and the representative particle size of each interval. The processing involves percentage normalization, peak interval identification, and standard deviation calculation. The output is the peak particle size and the standard deviation of the particle size distribution. This example is only used to illustrate the method of obtaining powder flowability characteristic parameters and does not limit the number of particle size intervals, the type of detection equipment, or specific particle size values.

[0050] S2. First, read the mold cavity geometric information and historical filling resistance data from the mold cavity three-dimensional geometric model database and the historical filling resistance database. Then, spatially assign the historical filling resistance data according to the thread root region and the smooth rod section region. Based on the regional pressure loss, equivalent flow rate and effective filling time period, construct the regional equivalent filling resistance and partition resistance conflict amount to generate the mold cavity partition resistance characteristic map.

[0051] First, the geometric structure information of the mold cavity corresponding to the titanium alloy orthopedic implant screw is read from the mold cavity 3D geometric model database. This geometric structure information includes the mold cavity coordinate system, the axial start position and end position of the thread segment, the radial coordinates of the deepest point of the thread profile, the thread profile height, the axial range of the polished rod segment, the radial coordinates of the inner wall of the polished rod segment, the gate location, and the topology of the runner system. Then, filling resistance data from multiple historical production cycles is read from the mold cavity historical filling resistance database. This historical filling resistance data includes nozzle pressure records, runner pressure records, local mold cavity pressure records, injection speed records, screw or plunger displacement records, valve needle opening and closing status records, and the 3D coordinates of each measuring point in the mold cavity coordinate system. Each measuring point is uniquely identified in the mold cavity coordinate system using its 3D coordinates and is associated with the corresponding pressure record via a sensor address.

[0052] Once the geometric structure information and measuring point coordinate information are prepared, all measuring points are traversed, and a region assignment determination is performed for each measuring point. If the axial coordinate of a measuring point is located between the axial start position and the axial end position of the thread segment, and the distance between the radial coordinate of the measuring point and the radial position of the deepest point of the thread profile is within the root neighborhood defined by the thread profile height, then the measuring point is marked as belonging to the thread root region. If the axial coordinate of a measuring point is within the axial range of the polished section, and the radial coordinate of the measuring point corresponds to the radial position of the inner wall surface of the polished section, then the measuring point is marked as belonging to the polished section region. For measuring points that are close to the functional section boundary, their main region assignment is determined according to the functional section boundary surface, the gate flow path, and the pressure response order of adjacent measuring points in the mold cavity CAD model. If the measuring point contributes to the pressure response of both regions, then the measuring point can be written as a boundary measuring point in the adjacency relationship field, and its weight is assigned according to the flow path contribution in the subsequent region resistance calculation.

[0053] Once the region affiliation of all measuring points is determined, the historical filling resistance data corresponding to each measuring point is tagged with a region label, forming a filling resistance dataset partitioned by spatial location. In this partitioned filling resistance dataset, the thread root region contains all measuring points marked as belonging to the thread root, along with their historical pressure records, flow records, and filling time period records; the polished rod section region contains all measuring points marked as belonging to the polished rod section, along with their historical pressure records, flow records, and filling time period records. For multi-cavity molds, a mold cavity number is also added to the region label, ensuring that each historical data entry is uniquely indexed in the format of "mold cavity number, region type, measuring point address, and production cycle number."

[0054] Furthermore, for each region, its corresponding historical pressure curve and equivalent flow curve are retrieved, and the effective filling period for that region in each historical production cycle is determined. Specifically, the continuous rising segment when the flow front reaches that region is identified from the local cavity pressure curve, the switching time from the filling stage to the holding stage is read from the equipment control events, and the time period between the arrival time of the flow front and the switching time of filling and holding is determined as the effective filling period for that region. If the local pressure sensor experiences a breakpoint, drift, or inconsistent response sequence with adjacent measuring points in a certain historical cycle, the reliability weight of the data for that cycle is reduced based on the sensor calibration record, curve continuity, and synchronization relationship with adjacent measuring points. If the equipment is equipped with a branch flow sensor, the flow record of the branch where that region is located is directly read. If the equipment is not equipped with a branch flow sensor, the equivalent volumetric flow rate of that region is calculated based on the screw or plunger speed, the equipment's calibrated effective cross-sectional area, and the feed compressibility calibration curve.

[0055] Furthermore, in order to decouple the local filling resistance of different regions of the mold cavity from the uniform injection pressure curve, the calculation formulas for the equivalent filling resistance of the region and the conflict amount of the zone resistance can be used as follows:

[0056]

[0057]

[0058] in, Indicates the first In the first historical production cycle The equivalent filling resistance of each region is expressed in megapascals per second per cubic centimeter. Indicates the first In the first historical production cycle The effective fill time period for each area is in seconds; Indicates the first In the first historical production cycle The upstream pressure of each region can be the nozzle pressure, the inlet pressure of the diversion channel, or the pressure of the adjacent upstream measuring point, and the unit is megapascals. Indicates the first Local cavity pressure in each region, in megapascals; Indicates entering the first The equivalent volumetric flow rate of each region is expressed in cubic centimeters per second. This indicates the pressure signal and flow signal at time... The dimensionless reliability weights are derived from sensor calibration status, time synchronization error, curve continuity, and flow calibration residuals.

[0059] When each region After all calculations are completed, robust statistics are performed on the equivalent filling resistance of multiple historical production cycles within the same region to obtain the resistance center value, resistance dispersion scale, filling arrival time center value, and spatial resistance gradient for that region. Specifically, for the thread root region, all measurement points marked as belonging to the thread root and historical cycle data are extracted. The equivalent filling resistance corresponding to each historical cycle is calculated according to the above formula, and the resistance center value of the thread root region is obtained based on the data reliability weight. Subsequently, the dispersion of the equivalent filling resistance of each historical cycle relative to the resistance center value is calculated to obtain the resistance dispersion scale of the thread root region. At the same time, the spatial resistance gradient of the thread root region is obtained based on the difference in resistance center values ​​between the thread root region and adjacent regions and the normalized coordinate distance of the mold cavity. For the smooth rod section region, the same processing procedure is performed to obtain the resistance center value, resistance dispersion scale, and spatial resistance gradient of the smooth rod section region. Subsequently, the resistance center value, spatial resistance gradient, and full mold cavity stable resistance baseline of the thread root region and the smooth rod section region are substituted into the above... The formulas are used to obtain the resistance conflict amount of the thread root region and the resistance conflict amount of the smooth section region, respectively.

[0060] Furthermore, a data structure for the cavity zone resistance characteristic map is created. This data structure can be organized as a two-dimensional table, key-value pairs, relational data table, or a structured object readable by the control system. Each region corresponds to a region entry in the map. After the data structure is created, the region identifier, cavity coordinate range, corresponding measuring point address, corresponding actuator address, region resistance center value, region resistance discrete scale, spatial resistance gradient, zone resistance conflict amount, and data reliability of the thread root region are written into the thread root region entry in the map. Subsequently, the region identifier, cavity coordinate range, corresponding measuring point address, corresponding actuator address, region resistance center value, region resistance discrete scale, spatial resistance gradient, zone resistance conflict amount, and data reliability of the polished rod section region are written into the polished rod section region entry in the map. For multi-cavity molds, each region entry in the map also includes a cavity number, enabling subsequent configuration of zone parameters according to specific cavities and functional regions.

[0061] After all region entries have been written, the integrity of the cavity partition resistance characteristic map is verified. Specifically, it is confirmed that each region entry includes a region identifier, coordinate range, resistance center value, resistance dispersion scale, spatial resistance gradient, partition resistance conflict amount, and actuator address. If a region lacks local pressure measurement points, an estimation entry is formed based on the pressure measurement points of adjacent regions, flow channel topology, and cavity geometry, and a lower data reliability is written into this entry. If a region lacks both direct measurement points and verifiable adjacent measurement point relationships, the map is marked as needing to supplement measurement points or calibration data. After the map passes the integrity verification, the cavity partition resistance characteristic map is stored in the data cache of the process parameter optimization system, expressing the filling resistance difference, resistance stability, and spatial heterogeneity of the thread root region and the smooth rod section region within the cavity in structured digital form.

[0062] For example, in a certain historical production cycle, the average pressure difference between the upstream pressure and the local cavity pressure in the thread root region during the effective filling period is 24 MPa, the equivalent volumetric flow rate is 3 cubic centimeters per second, and the effective filling time is 0.2 seconds. Therefore, the pressure difference time integral is approximately 4.8 MPa·s, and the filling volume is approximately 0.6 cubic centimeters. This yields an equivalent filling resistance of approximately 8.0 MPa·s / cubic centimeter for the thread root region in this cycle. In the same cycle, the average pressure difference in the polished section region is 15 MPa, the equivalent volumetric flow rate is 3 cubic centimeters per second, and the effective filling time is 0.2 seconds. Therefore, the equivalent filling resistance in the polished section region is approximately 5.0 MPa·s / cubic centimeter. After robust statistical analysis of multiple effective cycles, if both the center value of the resistance in the thread root region and the spatial resistance gradient deviate positively from the stable state of the entire cavity, the calculated conflict amount of the zoned resistance in the thread root region is positive. If the center value of the resistance in the polished section region is low and the spatial gradient is gentle, the calculated conflict amount of the zoned resistance in the polished section region is negative or close to the stability center. The inputs to this example are historical pressure curves, equivalent flow rates, effective filling time periods, and regional spatial relationships. The processing involves regional resistance integral calculation, robust statistics, and construction of zoned resistance conflict quantities. The output is a cavity zoned resistance characteristic map. This example is only used to illustrate how to use the core formula and does not limit the number of pressure measurement points, the number of cavities, or the granularity of the zone division.

[0063] S3. First, read the current batch powder flowability characteristic parameters, historical stable batch baseline and mold cavity partition resistance characteristic map. Then, convert the powder particle size distribution deviation into injection pressure correction amount, and convert the partition resistance conflict amount of each region into holding time partition compensation amount. Finally, generate parameter adjustment instructions containing pressure correction amount, holding time partition compensation amount and execution channel mapping relationship.

[0064] First, the peak particle size, particle size distribution standard deviation, powder detection reliability, and historical stable batch baseline version are read from the powder flowability characteristic data package output by S1. Then, the peak particle size stability center value, particle size distribution standard deviation stability center value, peak particle size stability scale, and particle size distribution standard deviation stability scale are read from the process database for the same powder grade, binder system, product specification, and mold number. The stability center value and stability scale are derived from historically stable batches that have continuously passed product quality testing, combined with the repeatability of the testing device and standard particle calibration records. Subsequently, the deviation of the current batch's peak particle size from the peak particle size stability center value and the deviation of the current batch's particle size distribution standard deviation from the particle size distribution standard deviation stability center value are calculated, and dimensionless processing is performed using the corresponding stability scale, allowing powder characteristics with different dimensions or different physical meanings to enter the same pressure correction calculation process.

[0065] Subsequently, the resistance conflict amount of the thread root region, the resistance conflict amount of the polished rod section region, the reliability of the region map, the address of the corresponding actuator, and the reference value of the region holding pressure time are read from the mold cavity partition resistance characteristic map output by S2. The region holding pressure time reference value is formed by production records that can simultaneously meet the requirements of thread root compactness, polished rod section flash amount, and product size consistency in historical stable production cycles; the reliability of the region map is jointly determined by the number of historical effective cycles in that region, sensor data integrity, resistance statistical dispersion, and map integrity verification results. When the resistance conflict amount of the thread root region is positive, it indicates that the region has higher local resistance or a stronger spatial resistance gradient compared to the stable filling state of the entire mold cavity; when the resistance conflict amount of the polished rod section region is negative or close to the stability center, it indicates that the region has lower filling resistance or weaker material replenishment requirements compared to the thread root region.

[0066] Furthermore, to uniformly convert powder batch fluctuations, regional resistance conflicts, and equipment operating capabilities into executable controllable quantities, the calculation formulas for the pressure correction and holding time zone compensation can be used as follows:

[0067]

[0068]

[0069] in, This indicates the injection pressure correction for the current batch, in megapascals (MPA). This represents the baseline value for injection pressure, in megapascals (MPA). and These represent the pressure operating boundaries determined by the rated pressure of the injection molding equipment, the mold clamping capacity, the runner sealing capacity, and the feed shear stability, respectively, in megapascals (MPA). This represents the reliability of powder testing and is a dimensionless quantity between 0 and 1. and These represent the dimensionless sensitivity coefficients for correcting injection pressure based on the deviation of the standard deviation of particle size distribution and the deviation of peak particle size, respectively. These sensitivity coefficients are obtained from controlled process validation tests and rolling calibration of historical qualified batches. This indicates the standard deviation of the particle size distribution of the current batch, in micrometers. This represents the center value of the standard deviation of particle size distribution, in micrometers. This represents the stability scale of the standard deviation of particle size distribution, in micrometers. This indicates the peak particle size of the current batch, in micrometers. This represents the peak particle size stability center value, in micrometers. This indicates the stable peak particle size scale, with units of micrometers.

[0070] In the formula for compensation amount of pressure holding time zone This represents the pressure holding time partition compensation amount for the \(r\)th region, in seconds; This represents the base value for the pressure holding time of the \(r\)th region, in seconds; and These represent the holding time operating boundaries determined by the gate freezing time, mold venting capacity, product size shrinkage allowable range, and actuator response capability of the region, respectively, in seconds; Let $r$ represent the reliability of the map in the \(r\)th region, which is a dimensionless quantity. The dimensionless sensitivity coefficient of the resistance conflict amount of the r-th region to the pressure holding time compensation is expressed as follows: Let represent the resistance conflict amount of the \(r\)th region generated by S2, which is a dimensionless quantity; This represents the dimensionless quantity of the quality residual of the most recent batch sampling inspection in the \(r\) region. For example, it is obtained by dividing the deviation of the micropore rate at the thread root or the deviation of the flash amount in the smooth section by the quality control scale formed during the process confirmation stage. This represents the conversion factor from quality feedback residual to holding time compensation, in seconds. When the current batch has not yet generated sampling quality feedback... Taking zero, the pressure holding time zoning compensation is dominated by the zoning resistance conflict. (Function) This means that the calculation results are limited to the operating range determined by the physical capabilities of the equipment, mold, and materials. Its purpose is to ensure that the output commands can be executed by the equipment and to avoid exceeding the allowable range of material shear stability, mold sealing capability, or zone gate freezing time.

[0071] After the pressure correction is calculated, the injection pressure baseline value is added to the pressure correction value to obtain the corrected injection pressure target value for the current batch. If the standard deviation of the particle size distribution in the current batch is significantly larger than that of historically stable batches, the deviation term of the particle size distribution standard deviation in the formula drives the injection pressure target value to correct for insufficient flowability through a nonlinear function; if the peak particle size of the current batch is close to that of historically stable batches, the contribution of the peak particle size term to the injection pressure correction is small. The nonlinear function is used to preserve the deviation direction and limit the excessive amplification of the pressure correction by abnormal detection data, avoiding sudden changes in the injection pressure command due to fluctuations in a single detection.

[0072] After calculating the compensation amount for the holding time in each region, the baseline value of the holding time in the thread root region is added to the compensation amount to obtain the compensated holding time in the thread root region. Simultaneously, the baseline value of the holding time in the polished section region is added to the compensation amount to obtain the compensated holding time in the polished section region. If the resistance conflict amount in the thread root region is positive and the region map reliability is high, the holding time in the thread root region is usually corrected to increase, extending the material replenishment effect in the high-resistance region before the gate freezes. If the resistance conflict amount in the polished section region is negative, or if the quality feedback residual indicates an over-replenishment tendency in this region, the holding time in the polished section region can be corrected to shorten, allowing the polished section region to isolate the holding time earlier after completing the necessary densification. The above correction direction is jointly determined by regional process calibration data and quality feedback data, and does not use uniform empirical judgments unrelated to the mold structure.

[0073] Furthermore, a consistency check is performed on the calculated pressure correction and the regional holding time compensation. Specifically, it is checked whether the corrected injection pressure target value is within the pressure operating boundary, whether the compensated holding time of each region is within the corresponding region's holding time operating boundary, and whether the holding time termination sequence of the thread root region and the smooth rod section region within the same mold cavity is consistent with the controllable sequence of the valve needle, the runner topology, and the gate freezing sequence. If the calculation result of a certain region is limited by the operating boundary, the difference before and after the limitation is recorded, and this difference is written into the subsequent quality feedback analysis field, without compensating other regions with uncalibrated additional pressure or additional holding time.

[0074] For example, in the historical stable batches of a product, the stable center value of the particle size distribution standard deviation is 4.0 micrometers, and the stable scale is 0.8 micrometers; the current batch's particle size distribution standard deviation is 4.8 micrometers, therefore the dimensionless deviation of the particle size distribution standard deviation is 1.0. The peak particle size of the current batch is close to the stable center value, therefore the peak particle size deviation is close to zero. If the injection pressure reference value is 95 MPa, the powder detection reliability is 0.95, and the particle size distribution standard deviation sensitivity coefficient is 0.06, then the pressure correction is approximately... That is, approximately 4.13 MPa; under the condition that the pressure operating boundary allows, the corrected injection pressure target value is approximately 99.13 MPa. Further, if the baseline value for the holding time in the thread root region is 2.2 seconds, the reliability of the region map is 0.92, the holding sensitivity coefficient is 0.25, the resistance conflict amount in the thread root region calculated by S2 is 1.0, and there is currently no quality feedback residual, then the compensation amount for the holding time in the thread root region is approximately... The actual holding time is approximately 0.39 seconds, and the compensated holding time is approximately 2.59 seconds. The inputs to this example are the current batch powder characteristics, historical stable batch baseline, zone resistance conflict amount, and process baseline value. The processing involves dimensionless deviation calculation, reliability adjustment, nonlinear limitation, and operating boundary constraints. The outputs are the pressure correction amount and the zone compensation amount for holding time in each region. This example is only used to illustrate the calculation method of the core formula and does not restrict the specific pressure, time, or sensitivity coefficient values.

[0075] After the pressure correction and the compensation for holding time in each region have passed consistency verification, a data structure for the parameter adjustment instruction is created. The parameter adjustment instruction includes at least the batch identifier, powder flowability characteristic parameter version, mold cavity zone resistance characteristic map version, injection pressure baseline value, pressure correction, corrected injection pressure target value, thread root region holding time baseline value, thread root region holding time zone compensation, thread root region post-compensation holding time, polished rod section region holding time baseline value, polished rod section region holding time zone compensation, polished rod section region post-compensation holding time, actuator address, operating boundary information, data reliability information, and the instruction's valid production cycle range. Subsequently, the parameter adjustment instruction is sent to the system, and the generation time, input data version, and calculation results of the instruction are stored in the process parameter optimization system for subsequent association with actual production execution records and quality inspection results.

[0076] S4. First, read the parameter adjustment command and map the injection pressure correction amount and the regional holding time compensation amount to the pressure control channel, regional holding pressure control channel and corresponding valve needle or flow channel actuator of the injection molding equipment. Then, in the same injection cycle, perform differentiated holding time configuration and injection pressure correction configuration on the thread root region and the smooth rod section region respectively to complete the optimization configuration of the process parameters for the current batch.

[0077] First, the parameter adjustment command generated by S3 is read from the data cache of the process parameter optimization system, and the batch identifier, mold number, area identifier, corrected injection pressure target value, holding time after compensation for the thread root area, holding time after compensation for the smooth rod section area, actuator address, and operating boundary information are parsed. Then, the actual clamping mold number, current hopper batch number, pressure control channel status, valve needle status, runner control status, and mold temperature control status of the current injection molding equipment are read, and the consistency between the actual status and the batch identifier and mold number in the parameter adjustment command is confirmed. If the current equipment status does not match the parameter adjustment command, the command issuance is stopped, and the mismatched fields are returned for regeneration or update; if the current equipment status matches the parameter adjustment command, the partition parameter configuration process begins.

[0078] Further, the mapping relationship of the actuators in the cavity partition resistance characteristic map is read. For dies equipped with independent pressure holding channels, controllable valve needles, hot runner branch valves, or equivalent partition isolation actuators, the thread root region and the polished rod section region correspond to independent or sequentially isolated control channels, respectively; each control channel has a unique address in the equipment control system and corresponds to the region entry in S2. The pressure holding time after compensation for the thread root region is written into the control parameter field of the actuator corresponding to the thread root region, and the pressure holding time after compensation for the polished rod section region is written into the control parameter field of the actuator corresponding to the polished rod section region. For multi-cavity dies, the control parameters are written one by one according to the mapping relationship of "cavity number, region type, actuator address" so that the thread root region and polished rod section region of different cavities in the same batch can all perform pressure holding time that matches their partition resistance conflict amount.

[0079] Subsequently, the corrected injection pressure target value is converted into a servo pressure command executable by the injection equipment. If the equipment primarily uses nozzle pressure closed-loop control, the corrected injection pressure target value is written into the pressure target field of the filling stage, and pressure closed-loop control is executed in conjunction with the injection speed curve. If the equipment has runner pressure adjustment capability, the pressure transmission calibration curve formed during the mold acceptance stage is further read, and the corrected injection pressure target value is converted into each runner inlet pressure command based on the correspondence between nozzle pressure, runner inlet pressure, and local cavity pressure. The pressure transmission calibration curve is used to compensate for pressure transmission deviations caused by differences in runner length, local corners, gate structure, and branch flow rates, ensuring that the pressure correction amount generated by the powder flowability characteristics can act on the actual cavity filling process in a way that is executable by the equipment.

[0080] Once the pressure control parameters and zone holding pressure control parameters are written, the injection cycle execution process begins. At the start of the injection cycle, the filling stage control is executed according to the corrected injection pressure target value, and nozzle pressure, runner pressure, local mold cavity pressure, screw displacement, injection speed, and valve needle position are collected simultaneously. As the flow front advances in the mold cavity, the holding pressure timing start point for the smooth section region and the thread root region is determined based on the local mold cavity pressure curve entering a continuous rising phase, the valve needle position changing, or the screw position reaching the region filling position determined by the runner volume calibration curve. This holding pressure timing start point is triggered by the actual flow front event, rather than by a uniform fixed time, thus adapting to changes in powder batch flowability and different region filling delays.

[0081] When the holding pressure timing start point of the polished section is triggered, the holding pressure timing of the corresponding actuator in the polished section is started, and the holding pressure effect in this section is maintained according to the compensated holding pressure time determined in S3. When the holding pressure time in the polished section reaches the compensated holding pressure time, the valve needle, flow divider valve, or holding pressure channel corresponding to the polished section is closed or isolated, so that the subsequent feeding pressure is no longer continuously transmitted to the polished section. Subsequently, when the holding pressure timing start point of the thread root section is triggered, the holding pressure timing of the corresponding actuator in the thread root section is started, and the holding pressure effect in this section is maintained according to the compensated holding pressure time determined in S3. When the holding pressure time in the thread root section reaches the compensated holding pressure time, the actuator corresponding to the thread root section is closed or isolated. Through the above control, the polished section section can end the holding pressure effect earlier after completing the necessary filling and densification, and the thread root section can obtain more sufficient feeding time according to its high resistance state, forming a differentiated holding pressure configuration between the two in the same injection cycle.

[0082] In one implementation, for equipment with only a single injection unit but possessing sequential valve needles or multi-segment pressure control capabilities, a zonal configuration can be implemented by combining sequential pressure curves with valve needle opening and closing timing. Specifically, when the flow front corresponds to the smooth rod segment region, a holding pressure duration matching the smooth rod segment region is executed, and pressure transmission is closed or reduced after the holding pressure in the smooth rod segment region ends; when the flow front corresponds to the thread root region, a holding pressure duration matching the thread root region is executed. If the mold structure does not have a zonal isolation actuator, only the modified overall injection pressure and the timing-segmented pressure curve can be executed, and the zonal holding pressure capability under this mold structure can be marked as limited in the production execution record; preferably, the mold is equipped with a zonal holding pressure actuator capable of independently isolating the thread root region and the smooth rod segment region, so that the holding time zonal compensation amount generated in S3 can be fully executed.

[0083] Furthermore, during the injection cycle, the actual execution curve is continuously acquired, and the actual nozzle pressure, actual runner pressure, actual local mold cavity pressure, actual valve needle opening and closing time, and actual regional pressure holding duration are aligned with the target data in the parameter adjustment command. This alignment process uses equipment response calibration error, pressure sensor calibration uncertainty, and normal fluctuation distribution in historical stable cycles as references. When the actual execution curve deviates continuously from the target curve, the region, time period, and actuator address where the deviation occurs are recorded, and the reliability of the region's graph is fed back for reduction. If the deviation is related to actuator response lag, runner blockage, pressure sensor drift, or incomplete valve needle closure, the product in the corresponding production cycle is marked as a sample requiring enhanced subsequent non-destructive testing or dimensional inspection, and the relevant execution records are written into the process traceability database.

[0084] Furthermore, while performing pressure correction and zoned pressure holding, baseline consistency control is maintained for mold temperature and cooling rate. Specifically, the mold temperature controller status and cooling circuit status are read, and the current mold temperature, cooling water flow rate, or cooling circuit open / closed status is matched with the auxiliary operating condition fields at the time the S2 chart was formed. If the mold temperature or cooling status deviates from historical stable operating conditions, the deviation is recorded and used as input for re-evaluating the applicability of the cavity zoned resistance characteristic chart in the next cycle. This process is used to avoid misinterpreting changes in feed viscosity caused by temperature or cooling status variations as powder batch fluctuations or zoned resistance changes.

[0085] For example, if the parameter adjustment command includes a corrected injection pressure target of 99.13 MPa, a holding time of 2.59 seconds after compensation for the thread root region, and a holding time of 1.53 seconds after compensation for the smooth rod region, and the mold cavity partition resistance characteristic map records the actuator address corresponding to the thread root region as R01 and the actuator address corresponding to the smooth rod region as S01, then first confirm that R01 and S01 of the current mold are both in a controllable state, and convert 99.13 MPa into a nozzle pressure or runner inlet pressure command according to the pressure transmission calibration curve. After the injection cycle starts, filling is performed according to the corrected pressure target; when the local pressure curve of the smooth rod region enters the continuous rising segment, the holding timer corresponding to S01 is started, and S01 is closed or isolated after 1.53 seconds; when the local pressure curve of the thread root region enters the continuous rising segment, the holding timer corresponding to R01 is started, and maintained until 2.59 seconds before R01 is closed or isolated. After execution, the output includes the actual pressure curve of this cycle, valve needle opening and closing time, zone pressure holding duration, and deviation record from the command. The inputs to this example are parameter adjustment commands, actuator mapping relationships, and equipment calibration curves. The processing steps include channel mapping, pressure command conversion, flow front event triggering, zone pressure holding timing, and valve needle isolation. The outputs are the configured and executed zone process parameters and traceable execution records. This example is only for illustrating zone parameter configuration and execution methods and does not limit the actuator model, valve needle quantity, or specific pressure and time values.

[0086] After the injection cycle is completed, the parameter adjustment command version, actual pressure curve, actual holding time, valve needle opening and closing status, mold temperature record, and product number of this cycle are associated and stored. Subsequently, the production line can use industrial CT, microscopic cross-section, local density detection, or equivalent non-destructive testing to generate microporosity or density feedback for the thread root area, and machine vision, contour scanning, or dimensional detection to generate flash amount feedback for the smooth rod section area. The feedback data is dimensionally consistent to form the quality feedback residual in S3, which is used to update the holding time zoning compensation amount in subsequent production cycles. Thus, in the initial stage of the current batch, parameter adjustment commands are generated based on the powder particle size distribution and mold cavity zoning resistance characteristic spectrum. During the production process, subsequent cycles are corrected based on the actual execution curve and sampling quality feedback, so that the process parameter optimization configuration has a continuous and traceable closed-loop update capability.

[0087] In summary, the optimization method for metal powder injection molding process parameters based on the embodiments of the present invention is clarified. It involves segmenting and statistically analyzing the particle size distribution curve of the current batch of powder to extract the standard deviation and peak particle size as characteristic parameters of powder flowability. Furthermore, it spatially decouples the historical filling resistance data of the mold cavity by thread root region and smooth rod region to generate a mold cavity partition resistance characteristic map containing partition resistance conflict amounts. This allows for the conversion of powder particle size distribution deviation into injection pressure correction amounts and partition resistance conflict amounts into holding time partition compensation amounts, achieving differentiated parameter configuration for the thread root and smooth rod regions. This effectively solves the problem of uncontrolled uniformity of mold cavity density caused by the coupling of microscopic characteristic fluctuations in powder batches and heterogeneous mold cavity filling resistance. It helps to coordinate the conflicting requirements of thread root densification and shrinkage compensation with anti-flash requirements in the smooth rod region, thereby significantly reducing the disorderly fluctuations in microporosity and flash amount within a single batch, improving batch consistency and product yield, and meeting the stringent quality stability requirements of medical-grade implants.

[0088] Here, those skilled in the art will understand that the specific operations of each step in the above-mentioned optimization system for metal powder injection molding process parameters have been referenced above. Figures 1 to 3 The optimization method for metal powder injection molding process parameters has been described in detail in the previous section, and therefore, its repeated description will be omitted.

[0089] In summary, the optimization system for metal powder injection molding process parameters based on the embodiments of the present invention is explained. It extracts the standard deviation and peak particle size of the particle size distribution as characteristic parameters of powder flowability by segmenting and statistically analyzing the particle size distribution curve of the current batch of powder. Furthermore, it spatially decouples the historical filling resistance data of the mold cavity according to the thread root region and the smooth rod region to generate a mold cavity partition resistance characteristic map containing partition resistance conflict amounts. Then, it converts the powder particle size distribution deviation into injection pressure correction amounts and the partition resistance conflict amounts into holding time partition compensation amounts, achieving differentiated parameter configuration for the thread root and smooth rod regions. This effectively solves the problem of uncontrolled uniformity of the overall mold cavity density caused by the coupling of micro-characteristic fluctuations in powder batches and heterogeneity in mold cavity filling resistance. It helps to coordinate the conflicting requirements of thread root densification and shrinkage compensation and smooth rod anti-flash requirements, thereby significantly reducing the disorderly fluctuations in microporosity and flash amount within a single batch, improving batch consistency and product yield, and meeting the stringent quality stability requirements of medical-grade implants.

Claims

1. A method for optimizing process parameters of metal powder injection molding, characterized in that, include: The particle size distribution curve of the current batch of powder is segmented and statistically processed according to the particle size interval. The peak particle size and the standard deviation of the particle size distribution curve are extracted as the flowability characteristic parameters of the current batch of powder. The historical filling resistance data of the mold cavity is divided into regions according to the spatial location of the thread root region and the smooth rod section region. The statistical characteristic of the filling resistance of each region is calculated to generate a mold cavity partition resistance characteristic map. The pressure correction amount is calculated based on the powder flowability characteristic parameters and the injection pressure reference value, and the holding time partition compensation amount is calculated based on the difference in filling resistance of each region in the mold cavity partition resistance characteristic map. The parameter adjustment command containing the pressure correction amount and the holding time partition compensation amount is output. The parameters of the injection molding equipment are configured according to the parameter adjustment instructions. Differentiated holding time compensation is applied to the thread root region and the smooth rod section region respectively. The injection pressure is adjusted according to the pressure correction amount to complete the process parameter optimization configuration for the current batch.

2. The method for optimizing metal powder injection molding process parameters according to claim 1, characterized in that, The current batch of powder flowability characteristics includes peak particle size, standard deviation of particle size distribution, and historical stable batch baseline corresponding to the current batch. The peak particle size is determined based on the proportion of particle volume distribution or particle size probability density in each particle size range. The standard deviation of particle size distribution is determined based on the representative particle size of each particle size interval, the proportion of normalized particle volume distribution, and the width of the particle size interval. The historical stable batch baseline includes the peak particle size stability center value, the particle size distribution standard deviation stability center value, the peak particle size stability scale, and the particle size distribution standard deviation stability scale.

3. The method for optimizing metal powder injection molding process parameters according to claim 2, characterized in that, The historical filling resistance data of the mold cavity has a region attribution label, which is determined based on the three-dimensional geometric structure information of the mold cavity and the spatial coordinates of the filling resistance measurement points. The three-dimensional geometric structure information of the mold cavity includes the axial starting position of the thread segment, the axial ending position of the thread segment, the radial position of the deepest point of the thread profile, the thread profile height, the axial range of the polished rod segment, and the radial position of the inner wall surface of the polished rod segment. When the axial coordinate of the measuring point is located between the axial start position and the axial end position of the thread segment, and the radial coordinate of the measuring point is located within the root neighborhood defined by the thread profile height, the measuring point is marked as belonging to the thread root region. When the axial coordinate of the measuring point is within the axial range of the smooth rod section, and the radial coordinate of the measuring point corresponds to the radial position of the inner wall surface of the smooth rod section, the measuring point is marked as belonging to the smooth rod section area.

4. The method for optimizing metal powder injection molding process parameters according to claim 3, characterized in that, The cavity partition resistance characteristic map includes the equivalent filling resistance of each region and the partition resistance conflict amount; The equivalent filling resistance of the region is determined based on the upstream pressure of the region, the local cavity pressure of the region, the equivalent volumetric flow rate of the region, and the effective filling time of the region. The partition resistance conflict amount is determined based on the robust center value of the equivalent filling resistance of the region, the center value of the resistance of the entire mold cavity, the regional spatial resistance gradient, and the center value of the stable spatial resistance gradient of the entire mold cavity.

5. The method for optimizing metal powder injection molding process parameters according to claim 4, characterized in that, The calculation of the equivalent filling resistance of the region and the conflict amount of the partition resistance is shown in the following formula: ； ； in, Indicates the first In the first historical production cycle The equivalent filling resistance of each region; Indicates the first The effective fill time period for each region; This indicates upstream pressure in the region; Indicates the localized cavity pressure in the region; Represents the equivalent volumetric flow rate of the region; Indicates the signal reliability weight; Indicates the first The amount of regional resistance conflict; Indicates the first The regional equivalent fill resistance robust center value of each region; This represents the center value of the total mold cavity resistance; Indicates the first The regional spatial resistance gradient of each region; This represents the center value of the drag gradient in the stable space of the entire mold cavity; and These represent the regional resistance stability scale and the spatial resistance gradient stability scale, respectively. and is a dimensionless fusion coefficient.

6. The method for optimizing metal powder injection molding process parameters according to claim 5, characterized in that, The pressure correction amount is determined based on the standard deviation of the particle size distribution of the current batch, the peak particle size of the current batch, the historical stable batch baseline, the reliability of powder detection, and the pressure operating boundary, as shown in the following formula: ； in, Indicates the injection pressure correction amount; Indicates the baseline value of injection pressure; and Indicates the pressure operating boundary; Indicates the reliability of powder detection; and These represent the sensitivity coefficients of the deviation of the standard deviation of particle size distribution and the deviation of peak particle size to the correction of injection pressure, respectively. This indicates the standard deviation of the particle size distribution in the current batch; This represents the stable center value of the standard deviation of particle size distribution; This represents the stability scale of the standard deviation of particle size distribution; Indicates the peak particle size of the current batch; Indicates the peak particle size stability center value; This indicates the stable scale of the peak particle size.

7. The method for optimizing metal powder injection molding process parameters according to claim 6, characterized in that, The pressure holding time zoning compensation amount is determined based on the zoning resistance conflict amount, the reliability of the zoning map, the zoning pressure holding time reference value, and the zoning pressure holding time operating boundary, as shown in the following formula: ； in, Indicates the first Pressure holding time compensation amount for each region; Indicates the first Baseline values ​​for pressure holding time in each region; and Indicates the first Operating boundaries for pressure holding time in each region; Indicates the first Reliability of the map data for each region; Indicates the first The sensitivity coefficient of the resistance conflict amount in each region to the pressure holding time compensation. Indicates the first The amount of regional resistance conflict; Indicates the first Quality feedback residuals for each region; This represents the conversion factor from quality feedback residual to pressure holding time compensation.

8. A system for optimizing process parameters of metal powder injection molding, characterized in that, include: The powder flowability feature extraction module is used to perform segmented statistical processing on the particle size distribution curve of the current batch of powder according to the particle size range, and extract the peak particle size and the standard deviation of particle size distribution as powder flowability feature parameters. The mold cavity partition resistance feature map generation module is used to divide the historical filling resistance data of the mold cavity into regions according to the spatial location of the thread root region and the smooth rod section region, calculate the equivalent filling resistance and partition resistance conflict of each region, and generate the mold cavity partition resistance feature map. The parameter compensation calculation and instruction generation module is used to correlate powder flowability characteristic parameters with injection pressure reference value to obtain pressure correction amount, convert zone resistance conflict amount into holding time zone compensation amount, and output parameter adjustment instruction containing the pressure correction amount and holding time zone compensation amount; The partition parameter execution and feedback module is used to configure the partition parameters of the injection molding equipment according to the parameter adjustment instructions, apply differentiated holding time compensation and injection pressure correction to the thread root region and the smooth rod section region respectively within the same injection cycle, and collect actual execution data and quality inspection results to form a closed-loop feedback.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for optimizing the metal powder injection molding process parameters according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for optimizing the process parameters of metal powder injection molding as described in any one of claims 1 to 7.

Images