An intelligent optimization method for powder metallurgy process parameters of automobile steering pulleys

Abstract

The present invention discloses an intelligent optimization method for powder metallurgy process parameters of an automobile steering bevel pulley, which relates to the field of industrial intelligent optimization technology. The method comprises an intelligent optimization terminal analyzing a standardized bevel pulley process characteristic data set to obtain a pressing density coefficient, a sintering shrinkage coordination coefficient, and a surface integrity coefficient, and inputting the three coefficients into a process robustness evaluation model to output a process optimization index; a parameter execution terminal dynamically selects an adjustment strategy for pressing pressure, holding time, or demolding rate according to the threshold interval of the process optimization index. The present invention calculates the pressing density coefficient, the sintering shrinkage coordination coefficient, and the surface integrity coefficient through an intelligent optimization terminal, and inputs the inputs into a process robustness evaluation model to output an optimization index, thereby quantifying the impact of each process link on product quality and achieving a scientific evaluation of process parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial intelligent optimization, and in particular to an intelligent optimization method for powder metallurgy process parameters of an automobile steering bevel pulley. Background Art

[0002] With the rapid development of the automotive industry toward electrification and intelligent driving, the performance requirements for automotive components are increasing. This is particularly true for the diagonal pulleys in steering systems, which must exhibit high precision, high strength, and excellent wear resistance to ensure steering system stability and safety. Powder metallurgy processes are widely used in automotive component manufacturing due to their advantages such as high material utilization, near-net-shape formation, and high production efficiency. However, the powder metallurgy forming process for diagonal pulleys involves the coordinated control of process parameters in multiple stages, including pressing and sintering. Traditional process parameter optimization methods are unable to meet the high-precision and high-reliability production requirements of modern automotive components.

[0003] Currently, the optimized design of powder metallurgy process parameters for automotive steering pulleys still relies heavily on engineering experience or traditional trial-and-error optimization methods. These methods lack systematic analysis of the coupled effects of multiple parameters, such as pressing force, demolding rate, sintering temperature, and holding time. Furthermore, the mechanisms for real-time data collection and dynamic adjustment during the process are imperfect, making it difficult to precisely control process parameters. Furthermore, existing technologies lack a robust quantitative assessment system for key quality indicators, such as workpiece density, sintering shrinkage coordination, and surface integrity. This results in a lack of scientific quantitative basis for process parameter optimization, making it difficult to adapt to the complex and ever-changing production scenarios of powder metallurgy processes.

[0004] Existing technologies face many technical bottlenecks in optimizing the powder metallurgy process parameters for automotive steering pulleys: on the one hand, the process parameters in the pressing and sintering stages have poor dynamic matching, which can easily lead to quality problems such as excessive internal porosity in the workpiece, an unreasonable proportion of pearlite phase, and surface profile deviation; on the other hand, there is a lack of real-time collection, intelligent analysis, and closed-loop optimization mechanisms for process data, making it impossible to dynamically adjust the process parameters based on quality feedback during the production process, resulting in low production efficiency and insufficient product quality stability.

[0005] Therefore, it is necessary to invent an intelligent optimization method for the powder metallurgy process parameters of an automobile steering bevel pulley to solve the above problems. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for intelligently optimizing the powder metallurgy process parameters of an automobile steering bevel pulley to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligently optimizing process parameters of powder metallurgy for an automobile steering pulley, comprising a process sensing terminal, a process data processing terminal, an intelligent optimization terminal, a parameter execution terminal, and a quality verification terminal, specifically comprising the following steps:

[0008] S1. The process sensing terminal collects real-time process data during the pressing and sintering stages through embedded pressure sensors and thermocouples to form a process feature data set of the inclined pulley; the process feature data set of the inclined pulley includes a pressing feature data set, a sintering feature data set, and a geometric feature data set;

[0009] S2. The process data processing terminal performs outlier elimination and normalization preprocessing on the inclined pulley process feature data set to obtain a standardized inclined pulley process feature data set;

[0010] S3. The intelligent optimization terminal analyzes the standardized inclined pulley process feature data set to obtain the pressing density coefficient, sintering shrinkage coordination coefficient, and surface integrity coefficient. These three coefficients are then input into the process robustness evaluation model to output the process optimization index.

[0011] The compression density coefficient is specifically: ,

[0012] Among them, η is the material compression factor, α is the powder deformation sensitivity, β is the demoulding damage factor, P max is the peak pressing force per unit area, K is the powder hardening modulus, v e is the actual demoulding rate, v e0 is the critical demoulding rate, n is the strain hardening exponent, and e is the natural constant.

[0013] The sintering shrinkage coordination coefficient is specifically: ,

[0014] Where γ is the diffusion kinetics correction factor, λ is the time decay constant, and k h is the actual heating rate, k h0 is the standard heating rate, E a is the apparent activation energy, R is the gas constant, T s is the sintering temperature, t s is the actual holding time, t s0 is the benchmark holding time, e is a natural constant, and ln is the logarithm with e as the base.

[0015] The surface integrity coefficient is specifically: ,

[0016] Among them, δ c is the actual profile deviation, δ c0is the contour tolerance threshold, k1 is the contour weight index, k2 is the end face runout weight index, J r is the actual end face runout, J r0 is the reference value of end face runout, △w is the actual groove width, △w0 is the nominal value of the groove width, and e is a natural constant.

[0017] The process robustness assessment model is specifically: ,

[0018] Wherein, D is the pressing density coefficient, S is the sintering shrinkage coordination coefficient, I is the surface integrity coefficient, D0 is the pressing density coefficient benchmark threshold, S0 is the sintering shrinkage coordination coefficient benchmark threshold, I0 is the surface integrity coefficient benchmark threshold, e is a natural constant, ω1, ω2 and ω3 are weight coefficients, ω1+ω2+ω3=1 and ω1, ω2 and ω3∈[0,1].

[0019] S4. The parameter execution terminal dynamically selects an adjustment strategy for pressing pressure, holding time, or demolding rate based on the threshold range of the process optimization index;

[0020] S5. The quality verification terminal detects the porosity and pearlite ratio of the adjusted workpiece through an X-ray diffractometer, and obtains surface profile data through the morphology detection terminal, and feeds it back to the intelligent optimization terminal.

[0021] Preferably, the pressing feature data set includes the peak pressing force per unit area and the actual demolding rate; the sintering feature data set includes the cooling rate, the atmospheric hydrogen concentration, the actual heating rate, the sintering temperature and the actual holding time; the geometric feature data set includes the actual profile deviation, the actual end face runout, the actual groove width, the internal porosity of the workpiece and the proportion of pearlite phase.

[0022] Preferably, the adjustment strategy is implemented as follows:

[0023] When the process optimization index Q is less than the threshold Q1, press Corrected pressing pressure, where △P max is the peak pressing force per unit area, K p is the pressure regulation gain, P max is the maximum pressing pressure, Q target Target value for process optimization index;

[0024] When the process optimization index Q is less than the threshold Q2 and greater than or equal to the threshold Q1, press Corrected holding time, where △t s is the holding time adjustment, t s is the actual holding time;

[0025] When the process optimization index Q is greater than or equal to the threshold Q2, press Fine-tune the demoulding rate, where △v e is the demoulding rate adjustment, v e0 is the critical demoulding rate.

[0026] Preferably, the feedback mechanism of the quality verification terminal includes:

[0027] Porosity threshold control inside the workpiece: When X-ray detection porosity V p Exceeds the preset threshold V p0 When , the sintering feature data set is triggered to be optimized again;

[0028] Pearlite content control: When the pearlite phase accounts for F p Below the preset threshold F p0 When increasing the atmospheric hydrogen concentration and reducing the cooling rate;

[0029] Surface profile tolerance control: When the actual profile deviation δ c Exceeding the contour tolerance threshold δ c0 , or actual end face runout J r Exceeding the end face runout reference value J r0 , or when the absolute value of the difference between the actual groove width △w and the nominal groove width △w0 exceeds 0.5×△w0, the pressing and demolding parameters are triggered to be re-optimized.

[0030] Technical effects and advantages of the present invention:

[0031] The present invention uses an intelligent optimization terminal to calculate the compaction density coefficient, sintering shrinkage coordination coefficient, and surface integrity coefficient, and inputs them into a process robustness assessment model to output an optimization index. This quantifies the impact of each process link on product quality and enables a scientific evaluation of process parameters.

[0032] The present invention dynamically adjusts the pressing pressure, holding time or demoulding rate according to the threshold range of the optimization index through the parameter execution terminal, thereby achieving adaptive optimization of process parameters and improving the stability of the production process and product consistency.

[0033] The present invention obtains porosity, pearlite phase ratio and surface profile data through the X-ray diffractometer of the quality verification terminal and the morphology detection terminal and feeds back to the intelligent optimization terminal, forming a closed-loop control, timely discovering process defects and triggering re-optimization, effectively improving product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a schematic diagram of the device connection of the present invention.

[0035] Figure 2 The figure is a flow chart of the method steps of the present invention. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] The present invention provides Figure 1 The equipment connection diagram shown includes a process sensing terminal, a process data processing terminal, an intelligent optimization terminal, a parameter execution terminal, and a quality verification terminal;

[0038] The present invention provides Figure 2 The method for intelligently optimizing the powder metallurgy process parameters of an automobile steering pulley shown in the figure specifically comprises the following steps:

[0039] S1. The process sensing terminal collects real-time process data during the pressing and sintering stages through embedded pressure sensors and thermocouples to form a bevel pulley process data set; the bevel pulley process data set includes a pressing feature data set, a sintering feature data set, and a geometric feature data set;

[0040] Furthermore, in the above technical solution, the pressing feature data set includes the peak pressing force per unit area and the actual demolding rate; the sintering feature data set includes the cooling rate, the atmospheric hydrogen concentration, the actual heating rate, the sintering temperature and the actual holding time; the geometric feature data set includes the actual contour deviation, the actual end face runout, the actual groove width, the internal porosity of the workpiece and the proportion of pearlite phase.

[0041] It should be noted that the peak pressing force per unit area is collected in real time by an embedded pressure sensor, which is integrated into the press mold or hydraulic system. The sensor directly monitors the dynamic pressure changes during the pressing process, records the peak value of the pressure curve and divides it by the mold contact area, and finally outputs the peak pressing force data in MPa.

[0042] The actual demoulding rate is collected by a displacement sensor and a high-precision timer: the displacement sensor tracks the stroke changes of the demoulding process in real time, and the timer records the corresponding time difference. The demoulding rate is dynamically calculated by the ratio of the displacement increment to the time increment to ensure the controllability of the demoulding action;

[0043] The cooling rate, actual heating rate, and sintering temperature are collected online via distributed thermocouples. Thermocouples are embedded in key locations on the workpiece surface and in the atmosphere flow channel within the sintering furnace to continuously record the temperature-time curve. The actual heating rate and cooling rate are calculated in real time using the slope of the curve. The sintering temperature is the stable value from the holding stage.

[0044] The atmospheric hydrogen concentration is collected by a gas concentration sensor installed in the atmosphere circulation system of the sintering furnace, such as a hydrogen concentration sensor, which detects the hydrogen concentration parameter in the furnace in real time;

[0045] The actual holding time is automatically recorded by the PLC control system: when the thermocouple detects that the furnace temperature reaches the preset sintering temperature, the timing module is triggered and the timing stops until the temperature drops, and the actual holding time in minutes is directly output;

[0046] The actual profile deviation and actual end face runout are obtained by scanning the workpiece surface with an optical profiler or a 3D scanner at a shape detection terminal, comparing the workpiece surface with the design model to calculate the profile deviation, and then rotating the workpiece and measuring the end face runout with a laser sensor.

[0047] The actual belt groove width is collected by an optical image measuring instrument: a multi-point optical scan is performed on the characteristic area of ​​the V-groove of the inclined pulley, and the actual groove width is output through an edge recognition algorithm and compared with the nominal value of the belt groove width;

[0048] The porosity inside the workpiece is determined by transmitting X-rays through the workpiece through an X-ray diffractometer. The instrument receives diffraction signals after the workpiece passes through the workpiece and performs analytical calculations on the diffraction signals based on theoretical density parameters of the material, thereby obtaining a specific value of the porosity inside the workpiece.

[0049] The pearlite phase ratio is quantitatively analyzed by comparing the characteristic diffraction peak intensity of the pearlite phase in the X-ray diffraction pattern with the standard pattern.

[0050] S2. The process data processing terminal performs outlier elimination and normalization preprocessing on the inclined pulley process feature data set to obtain a standardized inclined pulley process feature data set;

[0051] It should be noted that the execution process of the process data processing terminal is to independently perform an outlier removal operation on each subset of the pressing feature data set, the sintering feature data set, and the geometric feature data set, and use the 3σ criterion to calculate the mean and standard deviation of each parameter, and remove data points that meet the single acquisition value deviation from the mean by more than 3 times the standard deviation. At the same time, based on the process logic to verify the physical range, directly remove data whose sintering temperature exceeds the material melting point range, whose demolding rate exceeds the equipment limit, or whose porosity exceeds the theoretical threshold. After completing the outlier removal, each parameter is linearly mapped to the [0, 1] interval by the maximum-minimum normalization method, where the parameter range is taken from the reasonable boundary of the historical data, the peak range of the unit area pressing force is set to 300MPa to 600MPa, the sintering temperature range is set to 1050°C to 1150°C, and the profile deviation range is set to 0mm to 0.1mm. Finally, the three normalized subsets are reorganized into a standardized inclined pulley process feature data set according to the original structure. Its data format is unified into a matrix form, where each row represents a production batch and each column corresponds to a normalized parameter.

[0052] S3. The intelligent optimization terminal analyzes the standardized inclined pulley process feature data set to obtain the pressing density coefficient, sintering shrinkage coordination coefficient, and surface integrity coefficient. These three coefficients are then input into the process robustness evaluation model to output the process optimization index.

[0053] Furthermore, in the above technical solution, the compression density coefficient is specifically: , where η is the material compression factor, α is the powder deformation sensitivity, β is the demoulding damage factor, and P max is the peak pressing force per unit area, K is the powder hardening modulus, v e is the actual demoulding rate, v e0 is the critical demoulding rate, n is the strain hardening exponent, and e is the natural constant.

[0054] It should be noted that the core design goal of the compaction density coefficient formula is to establish a coupling effect model of compaction force and demoulding rate on workpiece density, and to quantify the process efficiency of the powder metallurgy compaction stage through mathematical forms. The specific structure adopts a modular design strategy, coupling the positive compaction force contribution term and the negative demoulding damage term in the form of a product, where the first half The exponential saturation model is used to characterize the mechanism of compression force: when the peak compression force per unit area P max When the pressure is low, the density increases slowly; as the pressure increases, the powder particles undergo plastic deformation, causing the density to increase rapidly; when the pressure exceeds the critical value, the density increase slows down due to the powder hardening effect and eventually approaches the theoretical limit of the material. This process is regulated by the powder deformation sensitivity α to control the pressure sensitivity, the powder hardening modulus K reflects the material's ability to resist deformation, and the strain hardening exponent n describes the compression nonlinear characteristics; the latter half For the demoulding damage mechanism: when the actual demoulding rate v e Exceeding the critical demoulding rate v e0 When the demoulding kinetic energy is released, it will cause micro-shard cracking or elastic rebound, resulting in density decay, and the degree of damage is amplified quadratically with the rate deviation. This effect is quantified by the demoulding damage factor β. This design is rigorous in mapping the physical mechanism. The latter term corresponds to the hardening resistance that needs to be overcome when the powder is compressed. The positive correlation between demoulding kinetic energy and internal defect generation is shown. The boundary behavior of the formula further verifies its rationality: under extreme high pressure, the former converges to η, which conforms to the densification saturation law, and the latter approaches zero at high-speed demoulding, reflecting a serious damage state. In engineering applications, the density coefficient D output by this formula is directly used as the input of the process robustness evaluation model, which is the pressing pressure P max and demoulding rate v e Provide a quantitative basis for dynamic optimization, ultimately achieving the core goal of suppressing demoulding defects and improving product consistency;

[0055] The material compression factor η reflects the maximum theoretical compression ratio of the powder material under ideal conditions, and its value is set to 0.93; the powder deformation sensitivity α describes the difficulty of plastic deformation of the powder particles, and its value is set as follows: if it is cemented carbide powder, α is set to 0.5; if it is iron-based powder, α is set to between 0.6 and 0.8; the demolding damage factor β quantifies the degree of microscopic damage caused by too fast a demolding rate, and its value is set to between 0.05 and 0.15; the powder hardening modulus K characterizes the ability of the powder to resist deformation under high pressure, and its value is set to between 800 MPa and 1200 MPa; the critical demolding rate v e0 The maximum allowable demolding rate without delamination or rebound defects is set between 5 mm / s and 10 mm / s. The strain hardening exponent n describes the nonlinear hardening behavior during powder compression and is set between 0.2 and 0.4.

[0056] Furthermore, in the above technical solution, the sintering shrinkage coordination coefficient is specifically: , where γ is the diffusion kinetics correction factor, λ is the time decay constant, and k h is the actual heating rate, k h0 is the standard heating rate, E a is the apparent activation energy, R is the gas constant, T s is the sintering temperature, t s is the actual holding time, t s0 is the benchmark holding time, e is a natural constant, and ln is the logarithm with e as the base.

[0057] It is important to note that the core design goal of the sintering shrinkage coordination coefficient formula is to quantify the synergistic effect of heating, heat preservation and diffusion mechanisms on sintering uniformity, and to solve the deformation and stress concentration problems caused by shrinkage incoordination through mathematical models. The formula uses a three-level coupling structure to achieve physical mechanism mapping: the diffusion kinetics term By heating rate ratio and the Arrhenius temperature term The product of the initial contact state and the atomic diffusion capacity represents the actual heating rate k h Too high will lead to insufficient activation of the particle surface, and the sintering temperature T s The diffusion rate is explicitly controlled by exponential form, and the square root exponent of one-half strictly conforms to the nonlinear theory of initial sintering shrinkage rate and diffusion dynamics; the insulation aging term Describes the marginal effect of holding time. When the actual holding time t s Below the benchmark value t s0 When the shrinkage is incomplete, the grains will grow abnormally when the shrinkage exceeds the limit. The time decay constant λ accurately adjusts the aging sensitivity to match the recrystallization characteristics of the material. The boundary behavior of the formula verifies the logical rigor: when T s As the diffusion resistance approaches infinity, t s =t s0 When the insulation term does not decay, and k h ≪k h0 or ts≫t s0 This triggers an attenuation response with uncontrolled contraction;

[0058] The diffusion dynamics correction factor γ is set to 0.95 to correct the deviation between the actual diffusion rate and the theoretical value; the time decay constant λ is set to 0.08; the reference holding time t s0 Set to 40min; the apparent activation energy E a The typical values ​​in the Powder Metallurgy Handbook can be directly quoted; the standard heating rate k h0 According to the type of powder material used for the inclined pulley, such as iron-based, copper-based, etc., combined with the technical manual provided by the material supplier or industry standards, such as the recommended heating rate range in the MPIF standard.

[0059] Furthermore, in the above technical solution, the surface integrity coefficient is specifically: , where δ c is the actual profile deviation, δ c0 is the contour tolerance threshold, k1 is the contour weight index, k2 is the end face runout weight index, J r is the actual end face runout, J r0 is the reference value of end face runout, △w is the actual groove width, △w0 is the nominal value of the groove width, and e is a natural constant.

[0060] It is important to note that the design logic of the surface integrity coefficient formula revolves around the comprehensive impact of surface geometric accuracy, form and position tolerances, and dimensional deviations on integrity, and is quantitatively characterized through dimensional modeling, function characteristic matching, and engineering weight adjustment. The formula uses three multiplication terms to decompose the core influences: the contour deviation term Using the reciprocal polynomial, when the actual profile deviation δ c Over-contour tolerance threshold δ c When the denominator increases rapidly due to the power k1, the nonlinear law of damage acceleration after the actual profile deviation exceeds the profile tolerance threshold is adapted; the end face runout term Using the exponential decay function, when the actual end face runout J r When it increases, the exponential term continues to decay, matching the law of cumulative damage to surface flatness caused by runout. k2 can adjust the sensitivity of runout. For example, if the brittle material needs to be strengthened, increase k2; the groove width deviation term The deviation ratio and integrity loss are related by linear subtraction. The actual belt groove width △w and the nominal belt groove width △w0 directly reflect the direct impact of dimensional deviation on assembly clearance and stress distribution.

[0061] The value of the profile weight index k1 is set to 1.5; the value of the end face runout weight index k2 is set to 2; the nominal value of the belt groove width △w0 is selected according to the ISO 5296 standard; the profile tolerance threshold δ c0 The value is set to between 0.05mm and 0.1mm, depending on the size and material; the end face runout reference value J r0 The value is set between 0.08mm and 0.15mm.

[0062] Furthermore, in the above technical solution, the process robustness assessment model is specifically: , where D is the pressing density coefficient, S is the sintering shrinkage coordination coefficient, I is the surface integrity coefficient, D0 is the pressing density coefficient benchmark threshold, S0 is the sintering shrinkage coordination coefficient benchmark threshold, I0 is the surface integrity coefficient benchmark threshold, e is a natural constant, ω1, ω2 and ω3 are weight coefficients, ω1+ω2+ω3=1 and ω1, ω2 and ω3∈[0,1].

[0063] It should be noted that the value of the pressing density coefficient reference threshold D0 is the statistical lower limit of the historical qualified batch D, ensuring that the workpiece strength meets the standard and there is no risk of cracking; the value of the sintering shrinkage coordination coefficient reference threshold S0 is the minimum value of the qualified dimensional tolerance after sintering; the value of the surface integrity coefficient reference threshold I0 is the historical actual profile deviation δ c , Actual end face runout J rThe lowest comprehensive score when both the actual belt groove width △w meet the standards;

[0064] The initial values ​​of the weight coefficients ω1, ω2 and ω3 are set to ω1=0.5, ω2=0.3, ω3=0.2, and are adjusted through a dynamic adjustment mechanism, specifically:

[0065] First define the defect rate of each process link: , , ;

[0066] Then use the weighted moving average method to balance historical data and current status: , , ,

[0067] Among them, Ψ is the historical weight inheritance factor, Ψ∈[0.6, 0.8];

[0068] Final Order ;

[0069] The dynamic adjustment mechanism is implemented under the constraint protection mechanism, wherein the constraint protection mechanism includes weight fluctuation limit, dormant period lock and manual intervention; the weight fluctuation limit is to limit the single adjustment range. , to prevent drastic weight jumps caused by fluctuations in a single batch of data, where i=1, 2, 3; the dormant period lock is when 10 consecutive batches are defect-free, freezing the weight adjustment to avoid over-optimization; the manual intervention is that engineers can manually override the weight, such as when the material changes, at which time the dynamic adjustment is suspended.

[0070] S4. The parameter execution terminal dynamically selects an adjustment strategy for pressing pressure, holding time, or demolding rate based on the threshold range of the process optimization index;

[0071] Furthermore, in the above technical solution, the adjustment strategy is implemented as follows:

[0072] When the process optimization index Q is less than the threshold Q1, press Corrected pressing pressure, where △P max is the peak pressing force per unit area, K p is the pressure regulation gain, P max is the maximum pressing pressure, Q target Target value for process optimization index;

[0073] When the process optimization index Q is less than the threshold Q2 and greater than or equal to the threshold Q1, press Corrected holding time, where △t s is the holding time adjustment, t s is the actual holding time;

[0074] When the process optimization index Q is greater than or equal to the threshold Q2, adjust the demolding rate slightly, where △v e is the demolding rate adjustment amount, and v e0 is the critical demolding rate.

[0075] It should be noted that the initial value of the target value Q of the process optimization index target is taken as the statistical mean of the process optimization index of historical qualified batches, usually set between 0.85 and 0.95. If the porosity or morphology exceeds the standard in the quality verification feedback, then according to Q target，new =Q target，old × (1 - 0.05 × defect rate) to lower the target value and gradually tighten the standard;

[0076] The values of the process optimization index thresholds Q1 and Q2 are adjusted according to the dynamic adjustment rules:

[0077] For the adjustment of Q1, when R 压制 > 0.3, that is, when the proportion of pressing defects exceeds 30% and D < D0 for three consecutive times, the adjusted value of Q1 is obtained through the formula and Q1 ∈ [0.65, 0.75];

[0078] For the adjustment of Q2, when δ > δ c0 or J r > J r0 or |△w - △w0| > 0.5 × △w0, the adjusted value of Q2 is obtained through the formula where N fail is the number of parts with surface profile out-of-tolerance, and N total is the total output of the current batch; when the porosity V p > V p0 for two consecutive batches, the adjusted value of Q2 is obtained through the formula The above adjustment of Q2 needs to satisfy Q2 ∈ [0.8, 0.85];

[0079] If there are no defects in 10 consecutive batches, lock the adjustment of Q1 and Q2.

[0080] S5. The quality verification terminal detects the porosity and pearlite phase proportion of the adjusted workpiece through an X-ray diffractometer, and obtains the surface profile data through the morphology detection terminal, and feeds it back to the intelligent optimization terminal.

[0081] Furthermore, in the above technical solution, the feedback mechanism of the quality verification terminal includes:

[0082] Porosity threshold control: When the porosity V detected by X-ray p exceeds the preset threshold V p0 , trigger the re-optimization of the sintering feature data set;

[0083] Pearlite content control: When the pearlite phase accounts for F p Below the preset threshold F p0 When increasing the atmospheric hydrogen concentration and reducing the cooling rate;

[0084] Surface profile tolerance control: When the actual profile deviation δ c Exceeding the contour tolerance threshold δ c0 , or actual end face runout J r Exceeding the end face runout reference value J r0 , or when the absolute value of the difference between the actual groove width △w and the nominal groove width △w0 exceeds 0.5×△w0, the pressing and demolding parameters are triggered to be re-optimized.

[0085] It should be noted that the sintering feature data set re-optimization includes:

[0086] Sintering temperature correction: press T s，new =T s +△T increases the temperature and enhances the atomic diffusion ability, where △T is the sintering temperature adjustment value, △T∈[10℃, 20℃];

[0087] Extending the holding time: executing the adjustment strategy of step S4;

[0088] Heating rate control: the actual heating rate k h Reduced to standard value k h0 80% to 90% of the original material to avoid thermal stress cracks;

[0089] Increase the atmospheric hydrogen concentration: Increase the hydrogen concentration by 5% to 10% to suppress the formation of oxides;

[0090] The pressing and demoulding parameters are then optimized to perform the adjustment strategy of step S4.

[0091] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent optimization method for powder metallurgy process parameters of an automobile steering pulley, characterized in that: It includes process sensing terminal, process data processing terminal, intelligent optimization terminal, parameter execution terminal and quality verification terminal, and specifically includes the following steps: S1. The process sensing terminal collects real-time process data during the pressing and sintering stages through embedded pressure sensors and thermocouples to form a process feature data set of the inclined pulley; the process feature data set of the inclined pulley includes a pressing feature data set, a sintering feature data set, and a geometric feature data set; S2. The process data processing terminal performs outlier elimination and normalization preprocessing on the inclined pulley process feature data set to obtain a standardized inclined pulley process feature data set; S3. The intelligent optimization terminal analyzes the standardized inclined pulley process feature data set to obtain the pressing density coefficient, sintering shrinkage coordination coefficient, and surface integrity coefficient. These three coefficients are then input into the process robustness evaluation model to output the process optimization index. The compression density coefficient is specifically: , Among them, η is the material compression factor, α is the powder deformation sensitivity, β is the demoulding damage factor, P max is the peak pressing force per unit area, K is the powder hardening modulus, v e is the actual demoulding rate, v e0 is the critical demoulding rate, n is the strain hardening exponent, and e is the natural constant; The sintering shrinkage coordination coefficient is specifically: , Where γ is the diffusion kinetics correction factor, λ is the time decay constant, and k h is the actual heating rate, k h0 is the standard heating rate, E a is the apparent activation energy, R is the gas constant, T s is the sintering temperature, t s is the actual holding time, t s0 is the reference holding time, e is a natural constant, and ln is the logarithm with e as the base; The surface integrity coefficient is specifically: , Among them, δ c is the actual profile deviation, δ c0 is the contour tolerance threshold, k1 is the contour weight index, k2 is the end face runout weight index, J r is the actual end face runout, J r0 is the end face runout reference value, △w is the actual belt groove width, △w0 is the nominal belt groove width, and e is a natural constant; The process robustness assessment model is specifically: , Where D is the pressing density coefficient, S is the sintering shrinkage coordination coefficient, I is the surface integrity coefficient, D0 is the pressing density coefficient benchmark threshold, S0 is the sintering shrinkage coordination coefficient benchmark threshold, I0 is the surface integrity coefficient benchmark threshold, e is a natural constant, ω1, ω2 and ω3 are weight coefficients, ω1+ω2+ω3=1 and ω1, ω2 and ω3∈[0,1]; S4. The parameter execution terminal dynamically selects an adjustment strategy for pressing pressure, holding time, or demolding rate based on the threshold range of the process optimization index; S5. The quality verification terminal detects the porosity and pearlite ratio of the adjusted workpiece through an X-ray diffractometer, and obtains surface profile data through the morphology detection terminal, and feeds it back to the intelligent optimization terminal.

2. The method for intelligent optimization of powder metallurgy process parameters of an automobile steering pulley according to claim 1 is characterized in that: The pressing feature data set includes the peak pressing force per unit area and the actual demolding rate; the sintering feature data set includes the cooling rate, atmospheric hydrogen concentration, actual heating rate, sintering temperature and actual holding time; the geometric feature data set includes the actual profile deviation, actual end face runout, actual groove width, internal porosity of the workpiece and the proportion of pearlite phase.

3. The method for intelligent optimization of powder metallurgy process parameters of an automobile steering pulley according to claim 1 is characterized in that: The adjustment strategy is implemented as follows: When the process optimization index Q is less than the threshold Q1, press Corrected pressing pressure, where △P max is the peak pressing force per unit area, K p is the pressure regulation gain, P max is the maximum pressing pressure, Q target Target value for process optimization index; When the process optimization index Q is less than the threshold Q2 and greater than or equal to the threshold Q1, press Corrected holding time, where △t s is the holding time adjustment, t s is the actual holding time; When the process optimization index Q is greater than or equal to the threshold Q2, press Fine-tune the demoulding rate, where △v e is the demoulding rate adjustment, v e0 is the critical demoulding rate.

4. The method for intelligent optimization of powder metallurgy process parameters of an automobile steering pulley according to claim 1 is characterized in that: The feedback mechanism of the quality verification terminal includes: Porosity threshold control inside the workpiece: When X-ray detection porosity V p Exceeds the preset threshold V p0 When , the sintering feature data set is triggered to be optimized again; Pearlite content control: When the pearlite phase accounts for F p Below the preset threshold F p0 When increasing the atmospheric hydrogen concentration and reducing the cooling rate; Surface profile tolerance control: When the actual profile deviation δ c Exceeding the contour tolerance threshold δ c0 , or actual end face runout J r Exceeding the end face runout reference value J r0 , or when the absolute value of the difference between the actual groove width △w and the nominal groove width △w0 exceeds 0.5×△w0, the pressing and demolding parameters are triggered to be re-optimized.

Images