An exhaust parameter optimization method for an electric clip blade MIM powder metallurgy process mold

Abstract

The application discloses an exhaust parameter optimization method for an electric clipper blade MIM powder metallurgy process mold, and comprises the following steps: collecting mold core cavity characteristic parameters and electric clipper blade tooth part model data; acquiring temperature field and pressure field data of the mold core working state through a thermal imaging detection device and a pressure sensor, and generating a thermal force diagram and a pressure nephogram; determining a gas trapping point based on an image fusion and gradient analysis algorithm; generating a plurality of exhaust layout schemes including adding, deleting, position adjusting or aperture optimizing of capillary exhaust holes; calculating the cavity pressure field, temperature field change and capillary hole distribution density under each scheme through numerical simulation, and sorting by using a weighted comprehensive evaluation algorithm; and finally selecting an optimal scheme according to user requirements and outputting to a production terminal. The application realizes accurate prediction of the gas trapping point and intelligent optimization of the exhaust scheme, and effectively improves the MIM forming quality and production efficiency of the electric clipper blade.

Description

Technical Field

[0001] This invention relates to the field of molds, and more particularly to a method for optimizing the venting parameters of an electric push-shear blade MIM powder metallurgy process mold. Background Technology

[0002] Metal Injection Molding (MIM) is a near-net-shape forming process widely used in the manufacture of complex and precision metal parts, such as the electric pusher blades described in this application. These blades have a blade body and a toothed portion with several teeth at one end. Metal powder is mixed with a binder and injected into a mold cavity, followed by debinding and sintering to obtain the final product. However, during the injection process, residual gas in the mold cavity is often trapped due to the rapid filling of the metal feed, creating trapped air. This trapped air is particularly prone to form in the narrow toothed portion of the electric pusher blade mold. Trapped air can lead to defects such as pores, surface depressions, or insufficient filling inside the part, severely affecting the product's mechanical properties and appearance quality.

[0003] In existing technologies, methods to solve the problem of trapped air are typically employed, such as optimizing the mold's venting structure, by setting venting grooves or capillary venting holes in the mold core. For example, prior art document CN223056725U discloses a device for solving trapped air during metal powder injection. This device uses a stirring ribbon structure in the lower and upper molds to rotate and agitate the metal powder during injection, thereby reducing voids and air bubbles between powder particles, improving material flowability, and reducing trapped air. Although this device improves the trapped air problem to some extent through mechanical agitation, it still has the following limitations:

[0004] Passive venting and lack of optimization: This device relies on physical agitation to promote venting, which is a passive solution. It cannot predict or precisely control the location and degree of trapped air formation. In actual production, trapped air points are often closely related to the part structure, such as the teeth of the electric pusher blades. However, this device does not involve the analysis of mold core structural features and part geometry data, so it is difficult to optimize venting for complex structures. Although existing injection mold cores integrate pressure sensors, temperature sensors and other devices, these sensors are only used to monitor the system. They cannot obtain the temperature and pressure field distribution inside the mold core in real time during injection, let alone dynamically identify trapped air points based on real-time data. They cannot quantitatively evaluate and optimize the venting effect in a closed loop.

[0005] Therefore, current MIM mold venting technology still generally suffers from problems such as reliance on experience-based design, lack of data-driven approaches, and inability to provide real-time prediction and adaptive optimization. There is an urgent need for an intelligent venting optimization method that can deeply integrate multi-source data and achieve accurate prediction and dynamic optimization control of trapped air, in order to improve the molding quality and production efficiency of complex MIM parts such as electric push-shear blades. Summary of the Invention

[0006] This invention addresses the problems of existing MIM mold venting technologies, which often rely on experience-based design, lack data-driven approaches, and are unable to provide real-time prediction and adaptive optimization. It provides a method for optimizing venting parameters of electrically driven shear blade MIM powder metallurgy molds.

[0007] To solve the above-mentioned technical problems, the present invention provides a method for optimizing the venting parameters of an electric push-shear blade MIM powder metallurgy process mold, comprising the following steps:

[0008] The mold core data acquisition step involves acquiring the characteristic parameters of the mold core cavity and the tooth model data of the electric push shear blade. The characteristic parameters of the mold core cavity include the layout information of the main exhaust channel and capillary exhaust holes of the mold core, and the tooth model data includes the inter-tooth gap depth, inter-tooth gap width, and number of teeth.

[0009] The working data acquisition step involves using a thermal imaging detection device that performs thermal imaging of the mold core cavity and a pressure sensor distributed within the mold core to acquire temperature and pressure field distribution data of the mold core under working conditions, based on the characteristic parameters of the mold core cavity and the monitoring area determined by the tooth model data. This process generates corresponding thermal maps and pressure cloud maps.

[0010] The step of determining air trapping points involves using the aforementioned thermal map and pressure cloud map, combined with tooth model data, to determine the overlapping areas in the fused image that exceed the preset temperature and the preset pressure gradient as air trapping points in the mold cavity through image fusion and gradient analysis algorithms.

[0011] The optimization scheme generation step generates multiple exhaust layout optimization schemes, including adding or deleting capillary exhaust holes, adjusting their positions, or optimizing their diameters, based on the location information of the trapped air points and the tooth model data. Each of the multiple exhaust layout optimization schemes is configured with a corresponding capillary exhaust hole distribution density.

[0012] The evaluation steps involve calculating the pressure field distribution, temperature field changes, and capillary exhaust hole distribution density of each exhaust layout optimization scheme using numerical simulation methods. The exhaust efficiency index of each exhaust layout optimization scheme is evaluated, and the exhaust layout optimization schemes are ranked based on a weighted comprehensive evaluation algorithm.

[0013] The final optimization output step is set up to receive user requirements, select the exhaust layout scheme that best meets the user's needs from multiple optimization schemes based on the comprehensive results of the scheme evaluation step and the specific requirements input by the user, and output the determined exhaust layout optimization scheme to the production terminal.

[0014] By adopting the above technical solution, a complete data-driven mold venting optimization closed-loop system was constructed. By systematically collecting static characteristic parameters of the mold core cavity and precise geometric data of the electric shear blade teeth, a precise data foundation was laid for subsequent analysis. Based on the key monitoring areas identified by this data, a thermal imaging and distributed pressure sensing network was deployed to form intuitive heat maps and pressure cloud maps, providing dynamic data support for subsequent analysis. Subsequently, advanced image fusion and gradient analysis algorithms were used to fuse multi-source data, accurately identifying entrapment points where temperature and pressure gradients overlapped, achieving a fundamental shift from experience-based judgment to scientific diagnosis. Based on the precisely located entrapment point information, various detailed layout schemes were automatically generated, including the addition, deletion, position adjustment, and aperture optimization of capillary venting holes. Each scheme... Each design includes clearly defined capillary vent distribution density parameters, making the design specific and feasible. Numerical simulation technology is then used to virtually simulate each candidate design, quantifying and predicting key performance indicators such as pressure field distribution, temperature field changes, and venting efficiency index. A weighted comprehensive evaluation algorithm is then used to rank the designs, overcoming the limitations of traditional single-indicator evaluation. Finally, the user's specific preferences and needs are received through a human-computer interaction interface. The system intelligently selects the venting layout design that best meets the user's actual production goals from the ranked optimized designs and directly outputs it to the production terminal to guide manufacturing. This entire process greatly improves the scientific rigor, accuracy, and automation of mold venting design, significantly shortens the mold development cycle, reduces the number of trial moldings, and fundamentally improves the molding quality and consistency of electric push-shear blade MIM products.

[0015] The present invention is further configured to include:

[0016] Subsequent process verification steps: A high-precision optical scanning measuring instrument is used to detect defects in the electric clipper blades produced at the air trapping point, obtaining defect data caused by the actual air trapping point.

[0017] Based on the defect data, the impact of air entrapment points on product quality is evaluated by image registration and difference comparison algorithms, the air entrapment defect severity index is calculated, and the air entrapment defect severity index is fed back to the air entrapment point determination step and the scheme evaluation step.

[0018] Based on the severity index of gas entrapment defects, the weighted comprehensive evaluation algorithm in the scheme evaluation step is adjusted.

[0019] By adopting the above technical solution, a crucial post-verification feedback mechanism is introduced on the basis of closed-loop optimization. A high-precision optical scanning measuring instrument is used to precisely inspect the actual produced electric pusher blades, accurately acquiring real defect data caused by trapped air, such as pore distribution and dimensional deviations. This allows for the correlation between simulated predictions and actual results. Subsequently, image registration and difference comparison algorithms are used to quantitatively evaluate the impact of trapped air points on the final product quality and calculate a quantifiable trapped air defect severity index. This index serves as a key feedback signal, sent back to the trapped air point judgment step and the scheme evaluation step. This allows the threshold and sensitivity of the trapped air point judgment algorithm to be adaptively calibrated based on actual product quality feedback, reducing false positives and false negatives. Simultaneously, in the scheme evaluation step, this index is incorporated as an important negative evaluation indicator into the weighted comprehensive evaluation system, prompting the optimization algorithm to be more inclined to avoid exhaust layouts that have been proven to cause serious defects in actual production when generating new schemes. This achieves closed-loop optimization from virtual simulation to physical world verification, greatly improving the reliability and practicality of the optimization system. It ensures that the optimized scheme not only performs well in simulation but also withstands the test of actual production, ultimately leading to the continuous production of high-quality products.

[0020] The present invention is further configured such that: the working data acquisition step generates corresponding thermograms and pressure cloud maps by using the temperature field and pressure field distribution data of the mold core in the working state;

[0021] Exhaust parameter optimization methods also include:

[0022] The mold core status monitoring and blockage early warning steps involve real-time acquisition of thermal maps and pressure cloud maps of the mold core under working conditions. The thermal maps are compared with historical thermal maps under normal conditions. When an abnormal high temperature gradient distribution is detected in the area corresponding to the capillary vent hole in the current thermal map, and the distributed pressure sensor detects an abnormal increase in the mold core cavity pressure, and subsequent process verification steps confirm that the product has defects caused by trapped air, it is determined that the capillary vent hole of the mold core is blocked, and a blockage alarm signal is generated.

[0023] When the blockage alarm signal is received, the continuous working time from the time the mold is put into use to the time the blockage alarm signal is first generated is recorded, and this time is used as a durability reference index and introduced into the weighted comprehensive evaluation algorithm of the scheme evaluation step to adjust the weight allocation of subsequent exhaust layout optimization schemes.

[0024] By adopting the above technical solution, the entire optimization system is endowed with proactive health management capabilities. Through real-time monitoring of the mold core's thermal and pressure cloud data during production and continuous comparison with historical normal baselines, it can accurately detect two key fault characteristics: abnormal high-temperature gradients in the capillary vent area and abnormal increases in cavity pressure. Combined with subsequent process verification steps to confirm product defects, this achieves accurate diagnosis and alarm for capillary vent blockage faults, avoiding false alarms. More importantly, the system records the continuous working time from when the mold core is put into use until the first blockage alarm occurs, and uses this time as a baseline. An objective durability reference index is introduced into the weighted comprehensive evaluation algorithm of the scheme evaluation step. This means that when the optimization algorithm designs exhaust schemes for new products, it will not only consider the exhaust efficiency of the scheme, but also take the anti-clogging durability of the scheme as an important evaluation dimension. Historically, schemes that are prone to early clogging will have their weight reduced in the evaluation, thereby guiding the system to generate a durable optimization scheme that is both efficient and robust. Ultimately, it has achieved a leap from simple performance design to reliability design, effectively extending the service life of the mold and reducing unplanned downtime and maintenance costs caused by sudden mold clogging.

[0025] The present invention is further configured such that, after the mold core status monitoring and blockage early warning step, it also includes:

[0026] The adaptive weight optimization step records the frequency of each blockage alarm signal and the location distribution of the trapped air points under the same exhaust layout scheme;

[0027] Based on the aforementioned clogging frequency and distribution, a correlation model is established between the clogging risk of capillary vents and the layout parameters.

[0028] Based on this model, the weighted comprehensive evaluation algorithm in the scheme evaluation step is adjusted to dynamically increase the proportion of relevant exhaust structure data with blockage frequency exceeding the preset value in the weight allocation.

[0029] By adopting the above technical solution, the system's self-learning and self-evolution functions are further realized on the basis of status monitoring. By systematically recording the frequency and corresponding spatial distribution of multiple blockage alarm signals under the same exhaust layout scheme, a failure database is accumulated. Based on this data, a quantitative correlation model between capillary exhaust hole blockage risk and layout scheme parameters is established. For example, it is found that exhaust holes with specific apertures, specific distribution densities, or specific positional orientations have a higher probability of blockage. This model enables the system to deeply understand the relationship between design decisions and potential risks. Subsequently, the weighted comprehensive evaluation algorithm in the scheme evaluation step is adjusted in a targeted manner using this model, dynamically increasing the proportion of relevant exhaust structure parameters that have been identified by the model as having high blockage frequency characteristics in the weight allocation. This means that the algorithm will be more vigilant and penalize designs with high-risk characteristics when evaluating schemes, thereby actively avoiding known reliability pitfalls and promoting the optimization results to continuously evolve towards a safer and more reliable direction. This mechanism of algorithm self-iteration based on historical failure data enables the entire system to continuously accumulate knowledge, become more intelligent with use, and finally output an optimized scheme that has been verified by reliability.

[0030] The present invention is further configured such that the weighted comprehensive evaluation algorithm in the scheme evaluation step is performed in the following manner:

[0031] The evaluation indicators are determined, including the exhaust efficiency index and the capillary exhaust hole distribution density.

[0032] Assign initial weights to each indicator of the evaluation criteria;

[0033] The initial weights of the aforementioned indicators are dynamically adjusted based on the severity index of trapped air defects from the subsequent process verification step, the durability reference index from the mold core condition monitoring and blockage early warning step, and the blockage frequency from the adaptive weight optimization step.

[0034] Based on the adjusted weights, calculate the comprehensive score of each exhaust layout optimization scheme and rank them accordingly.

[0035] By adopting the above technical solution, the dynamic adjustment mechanism of the weighted comprehensive evaluation algorithm is concretized, making it a multi-objective, multi-constraint intelligent decision-making core. The algorithm first establishes core evaluation indicators such as the exhaust efficiency index and capillary exhaust hole distribution density, assigns initial weights to them, and constructs a basic evaluation framework. Its key feature is its ability to receive and integrate multi-source feedback signals from different functional modules of the system, including the entrapment defect severity index from subsequent process verification steps (reflecting the actual quality output effect of the solution), the durability reference index from the mold core status monitoring and blockage early warning steps (reflecting the historical reliability performance of the solution), and other related indicators. The congestion frequency data in the adaptive weight optimization step reveals the risk level of specific design features. Based on these real-time feedback signals, the algorithm dynamically adjusts the initial weights of various indicators. For example, if a solution is highly efficient but repeatedly causes serious defects or early congestion, the weights of its quality and reliability indicators will be dynamically increased, thereby reducing its overall score. This dynamic weighting mechanism that integrates multiple factors ensures that the optimal solution in the final ranking is an optimal balance point among multiple objectives such as exhaust efficiency, product quality, and mold life, rather than a winner of a single performance indicator, thus meeting the core requirement of modern intelligent manufacturing to maximize comprehensive benefits.

[0036] The present invention is further configured such that: the scheme evaluation step is also based on a pre-established correlation between capillary exhaust hole distribution density and blockage duration, and a correlation between capillary exhaust hole distribution density and exhaust efficiency.

[0037] The final optimization output step selects the corresponding capillary venting density optimization scheme according to user needs, wherein:

[0038] If the user selects the exhaust efficiency priority strategy, an optimization scheme that increases the density of capillary exhaust holes will be generated.

[0039] If the user selects the lifespan priority strategy, an optimized solution that reduces the density of capillary venting holes will be generated.

[0040] By adopting the above technical solution, two clear optimization strategies tailored to different production goals are provided, giving users the flexibility to customize the final solution according to their actual needs. The underlying basis is a pre-established quantitative correlation between capillary vent density, clogging duration, and venting efficiency. This reveals that increasing capillary vent density usually improves venting efficiency because it provides more venting paths, but it may also reduce the strength of the mold core in that area due to the increased number of vents, making the mold core more prone to deformation. This could lead to risks associated with air trapping points caused by capillary vent blockage or shorten the average maintenance cycle. Conversely, reducing density has the opposite effect. Based on this physical principle... The system follows a pattern: if the user selects the exhaust efficiency priority strategy, the system will automatically generate an optimization scheme that increases the density of capillary vents, aiming to maximize the exhaust capacity per unit time to meet the needs of high-efficiency production. This is particularly suitable for scenarios with extremely high production cycle requirements. If the user selects the service life priority strategy, the system will automatically generate an optimization scheme that reduces the density of capillary vents, extending the continuous working time of the mold by reducing potential failure points. This is particularly suitable for unattended operation or scenarios where maximizing mold life is desired. This clear strategy division makes the optimization system no longer a black box, and its output scheme is interpretable, directly meeting the diverse production management goals of users.

[0041] The present invention is further configured to include:

[0042] Predictive maintenance steps, based on the continuous working time, frequency of occurrence and location distribution of blockage alarm signals obtained in the mold core status monitoring and blockage early warning steps, establish a blockage prediction model for the capillary vent holes of the mold core;

[0043] Based on the output of the congestion prediction model, the remaining service life estimate of the mold core is calculated. When the remaining service life estimate is lower than a preset threshold, a predictive maintenance plan containing maintenance time and spare parts information is generated and output to the production terminal.

[0044] By adopting the above technical solution, the concept of predictive maintenance is deeply integrated into the mold management system. Based on the dynamic data such as continuous working time, frequency of blockage, and location distribution accumulated in the status monitoring and blockage early warning steps, a blockage prediction model for the capillary venting holes of the mold core is established using this data. This model can learn and predict the performance degradation law of the venting holes under specific working conditions and designs. Then, based on the model output, the remaining service life estimate of the mold core is accurately calculated, realizing a precise insight into the health status of the mold. When the remaining service life estimate is lower than the preset safety threshold, the system can proactively generate a predictive maintenance plan containing specific maintenance time windows and required spare parts information, and output this plan to the production terminal in advance. This function upgrades mold maintenance from traditional post-fault repair or periodic preventive maintenance to on-demand predictive maintenance, allowing production planners to arrange maintenance activities in advance without interrupting the production process, thereby minimizing unexpected downtime, optimizing spare parts inventory management, improving the overall utilization rate of equipment, and ultimately achieving cost reduction and efficiency improvement.

[0045] The present invention is further configured such that the characteristic parameters of the mold core cavity also include the mold core hardness. In the optimization scheme generation step, the shortest distance between capillary vent holes is set according to the mold core hardness. When the mold core hardness is higher, the shortest distance between capillary vent holes is smaller.

[0046] By adopting the above technical solution, and incorporating the key material property of mold core hardness into the generation logic of the venting layout, the minimum spacing between capillary vents is set according to the specific value of the mold core hardness during the optimization scheme generation step. This is because mold core hardness is directly related to its wear resistance, toughness, and sensitivity to stress concentration. When the mold core hardness is high, more and denser capillary vents can be opened. In this case, reducing the minimum spacing between capillary vents allows for a denser layout of vents, thereby improving venting efficiency. However, since high-hardness materials have strong crack propagation resistance, they can withstand the potential stress risk brought about by the reduction in vent spacing. Conversely, for low-hardness, tough materials, a larger minimum vent spacing needs to be set to prevent cracks from forming under repeated injection pressure. This mechanism, which dynamically links material mechanical properties with design rules, ensures that the generated optimized scheme is not only functionally excellent but also structurally safe and reliable. This fundamentally avoids early mold cracking failure caused by improper design, improving the overall lifespan of the mold and production safety.

[0047] The present invention is further configured to include:

[0048] The knowledge base construction step is used to establish a historical database for storage, which is used to store historical model design parameters, exhaust layout schemes, corresponding thermal data, corresponding pressure data, and entrapment point determination results.

[0049] In the scheme evaluation step, historical success cases and failure data from the historical database are called up, and data are provided for the assessment of exhaust efficiency and risk coefficient of the current optimization scheme through similarity matching.

[0050] By adopting the above technical solutions, a knowledge hub supporting the continuous evolution of the entire optimization system is constructed. A historical database is established to systematically store the complete data chain of past projects, including historical model design parameters, the exhaust layout schemes adopted, corresponding thermal and pressure data, and the results of trapped air point determination. This forms a valuable domain knowledge base. When evaluating schemes for new projects, the system can call upon historical successful cases and failure data from this knowledge base for similarity matching. This provides solid data support and prior verification for the exhaust efficiency and potential risk coefficient of the currently evaluated exhaust layout scheme. For example, if the current scheme is highly similar to a scheme in the historical database that caused severe blockage, the system will issue an early warning of its high risk. Conversely, if it is similar to a highly efficient and stable scheme, its confidence level will be enhanced. This mechanism of decision support based on historical big data greatly improves the accuracy and reliability of scheme evaluation, avoids repeating mistakes, and allows for the digital accumulation and reuse of expert design experience. This enables the entire system to continuously learn and inherit knowledge, becoming increasingly intelligent with use.

[0051] The present invention is further configured such that the working data acquisition step includes:

[0052] Based on the variation pattern of the temperature field data acquired by the thermal imaging detection device, the interval detection cycle for the mold core is set.

[0053] If, within the specified interval detection period, the distributed pressure sensor detects that the pressure value of the mold cavity exceeds the preset pressure value, then a supplementary detection of the mold temperature field data is triggered.

[0054] By adopting the above technical solution, the resource allocation and triggering logic of the data acquisition process are optimized, achieving an intelligent balance between energy consumption and efficiency. By analyzing the changing patterns of temperature field data acquired by the thermal imaging detection device, the interval detection cycle for the mold core is set. For example, during the stable production stage, the detection interval is automatically extended to reduce data processing volume and system energy consumption. Within the interval detection cycle, the distributed pressure sensor acts as a sensitive trigger defense. If the pressure value of the mold core cavity exceeds the preset pressure threshold, a key signal indicating that an abnormality may occur in the filling process, supplementary detection of the mold core temperature field data is immediately triggered. This intelligent collaborative acquisition strategy, based on low-frequency temperature monitoring and triggered by abnormal pressure events, ensures that data under most steady-state conditions can be captured, while also ensuring that high-frequency temperature field data can be acquired immediately for detailed analysis when transient abnormal events occur. Thus, while ensuring that the system monitoring effect is not compromised, the average computational load and data storage pressure of the system are significantly reduced, improving the engineering practicality and economy of the entire method.

[0055] This invention, by employing the above-mentioned technical solutions, possesses significant technical effects: The technical solutions claimed in this patent application constitute an unprecedented intelligent design, production, and operation and maintenance lifecycle management system for venting of MIM powder metallurgy molds with electric push-button blades. Its effects far exceed the simple summation of the effects of each claim; rather, it generates a significant synergistic multiplier effect. This system begins with precise data acquisition of the mold core and product tooth profile. Through the integration of thermal imaging and a distributed pressure sensing network, it achieves real-time, full-domain perception of the mold's working status. Furthermore, it utilizes multi-source information fusion and gradient analysis algorithms to scientifically diagnose venting points, completely overturning the traditional trial-and-error mode that relies on experience. This allows for the automatic generation of multiple detailed venting layout optimization schemes, which are then virtually simulated and ranked using numerical simulation and multi-objective weighted evaluation algorithms. Finally, it outputs the optimal solution that balances efficiency and reliability. Its innovation is further reflected in the introduction of a closed-loop feedback mechanism for subsequent process verification, enabling the optimization system to optimize based on actual product quality data. The system performs self-calibration and possesses advanced status monitoring and blockage early warning capabilities. It can detect faults early and record durability data for feedback on design optimization. It can even build a blockage prediction model based on historical data to achieve predictive maintenance and plan maintenance schedules in advance to avoid unplanned downtime. It is particularly noteworthy that the system incorporates material properties such as mold core hardness into the design rules to ensure structural safety and builds a historical knowledge base to support decision-making using big data similarity matching. Its data acquisition strategy also achieves intelligent and energy-saving operation. In summary, this solution successfully transforms mold venting design from an isolated craft into a comprehensive intelligent engineering project integrating sensing technology, data analysis, machine learning, numerical simulation, and decision science. The resulting technical effects are a comprehensive improvement in the quality consistency and yield of electric shear blade MIM products, a significant reduction in mold development and debugging cycles, a significant extension of mold lifespan, and the realization of on-demand predictive maintenance to maximize equipment utilization, ultimately achieving a leapfrog improvement in production efficiency and economic benefits. Attached Figure Description

[0056] Figure 1 This is a flowchart of the steps for optimizing the exhaust parameters of MIM powder metallurgy molds using electric push shear blades;

[0057] Figure 2 It's a heat map;

[0058] Figure 3 This is a diagram showing the distribution of capillary exhaust holes in the exhaust layout optimization scheme. Detailed Implementation

[0059] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0060] Example:

[0061] In existing technologies, metal powder injection molding is widely used in the manufacture of complex and precision parts, such as electric push cutter blades. The toothed structure of the electric push cutter blades is complex, and the mold cavity is prone to gas entrapment during injection due to rapid metal feeding, resulting in trapped gas. Existing technologies typically employ passive venting using mechanical agitation or fixed venting structures, which cannot be dynamically adjusted based on part geometry and real-time process data. This leads to low venting efficiency and difficulty in accurately locating trapped gas areas. To address this problem, a dynamic optimization method combining mold structure parameters and real-time process data is needed. Analysis reveals a strong correlation between trapped gas and mold cavity structure, as well as temperature and pressure distribution during injection. However, existing technologies lack the ability to fuse and analyze multi-source data. Therefore, the following technical approach is proposed: A monitoring area is constructed by collecting mold structure parameters and part geometry data; real-time temperature and pressure field distributions are acquired using distributed sensors; trapped gas areas are identified using image processing algorithms; multiple venting optimization schemes are generated based on the identification results and quantitatively evaluated; finally, the optimal scheme is selected according to actual needs.

[0062] This application proposes an exhaust parameter optimization method including a mold core data acquisition step, a working data acquisition step, a trapped air point determination step, an optimization scheme generation step, a scheme evaluation step, and a final optimization output step. The mold core data acquisition step acquires the mold core cavity characteristic parameters and the electric pusher blade tooth model data. The working data acquisition step acquires temperature and pressure field distribution data through a thermal imaging detection device and a pressure sensor. The trapped air point determination step uses image fusion and gradient analysis algorithms to identify abnormal temperature and pressure areas. The optimization scheme generation step generates a capillary exhaust hole adjustment scheme based on the trapped air point location. The scheme evaluation step evaluates the exhaust efficiency index through numerical simulation and performs comprehensive ranking. The final optimization output step selects the optimal scheme based on user requirements and outputs it to the production terminal.

[0063] The characteristic parameters of the mold core cavity include the layout information of the main venting channel and capillary venting holes, which can be obtained using a 3D scanner or mold design drawing analysis tools to establish a spatial relationship model between the mold structure and the venting channels. Tooth model data includes the inter-tooth clearance depth, inter-tooth clearance width, and number of teeth, which can be obtained through CAD model analysis or optical measurement equipment to determine the monitoring area and sensitive parameters. Thermal imaging detection devices can use infrared thermal imagers, such as non-contact temperature measurement devices with a wavelength range of 8-14μm, or distributed matrix temperature sensors to capture temperature distribution changes during mold operation. Pressure transmission... The sensor can be a miniature piezoresistive sensor, such as a distributed array sensor with a range of 0-50MPa, used to monitor pressure fluctuations in the mold cavity in real time. The image fusion and gradient analysis algorithm can use multi-scale image registration technology combined with edge detection operators, such as a combination of Sobel and Canny operators, to identify overlapping areas of temperature and pressure anomalies. The diameter of the capillary vent is designed to be smaller than the particle size of the MIM feed powder; its specific diameter range is 0.005mm to 0.008mm. This range of diameters belongs to ultra-microscale processing. In molds, it is generally processed by laser micromachining technology, but micro-electrical discharge machining technology can also be used.

[0064] The mold core data acquisition step establishes a 3D model including the exhaust channel layout and tooth structure by analyzing the mold design parameters and part geometric features. The working data acquisition step simultaneously acquires temperature and pressure field distribution data during mold operation. The generated thermal map and pressure cloud map achieve spatial matching through coordinate alignment. The air trapping point determination step performs gradient calculation on the fused image. When the temperature exceeds the set threshold and the pressure gradient abrupt change area overlaps, it is determined to be an air trapping point. The optimization scheme generation step adjusts the capillary exhaust hole layout according to the location of the air trapping point, such as increasing the exhaust hole density or adjusting the hole diameter in the air trapping area. The scheme evaluation step calculates the exhaust efficiency index under different schemes through finite element simulation and sorts the schemes by combining a weighted algorithm. The final optimization output step selects the optimal scheme from the sorting results and outputs processing instructions according to the user-defined priorities, such as production efficiency or mold life requirements.

[0065] Existing technologies rely on fixed exhaust structures or passive exhaust via mechanical agitation, which cannot be dynamically adjusted based on actual process data. This solution achieves precise positioning of the trapped air point by integrating mold structure parameters, real-time temperature and pressure data, and part geometric features. Through multi-scheme simulation evaluation and matching with user needs, a closed-loop optimization mechanism is formed. Existing technologies lack quantitative evaluation indicators. This solution introduces an exhaust efficiency index and a weighted comprehensive evaluation algorithm to provide a quantifiable basis for scheme selection.

[0066] This application can effectively identify air trapping points in the toothed area of ​​the electric shear blade mold and dynamically generate an optimization scheme based on real-time process data. Through numerical simulation and comprehensive evaluation, it ensures that the selected exhaust layout scheme takes into account both exhaust efficiency and mold life requirements. This method solves the problem that traditional passive exhaust structures cannot adapt to the geometric features of complex parts, reduces the defect rate of parts caused by air trapping, and improves the molding quality of MIM process.

[0067] The method also includes a subsequent process verification step, which is equipped with a high-precision optical scanning measuring instrument to perform defect detection on the electric clipper blades produced through the air trapping point, obtain the defect data caused by the actual air trapping point, evaluate the degree of impact of the air trapping point on product quality based on the defect data, calculate the air trapping defect severity index, and feed the air trapping defect severity index back to the air trapping point judgment step and the scheme evaluation step; and adjust the weighted comprehensive evaluation algorithm in the scheme evaluation step according to the air trapping defect severity index.

[0068] A high-precision optical scanning measuring instrument is a device that uses non-contact optical imaging technology to measure the three-dimensional shape of a part surface. Specifically, it can be implemented using laser triangulation or structured light scanning technology to capture the geometric features of surface depressions or porosity defects caused by trapped air. The image registration and difference comparison algorithm is an algorithm that analyzes the difference areas after spatially aligning the actual product defect image with the theoretical model. Specifically, it can be implemented by combining feature point matching with grayscale difference calculation to quantify the size and distribution density of trapped air defects. The trapped air defect severity index is a quantitative indicator that comprehensively considers the number, size, and location of defects on product performance. Specifically, it can be calculated by weighted summation of the defect area and the overlap of the key area to reflect the actual impact of trapped air on the product pass rate.

[0069] After the electric shear blades are demolded, a high-precision optical scanning measuring instrument performs a full-surface scan of the tooth area, generating a three-dimensional defect distribution map including depressions and pores. The defect map is aligned with the original tooth model through image registration to identify the defect features corresponding to the trapped air points. The difference comparison algorithm calculates the overlap ratio between the defect area and the key area of ​​the tooth, and generates a trapped air defect severity index based on the defect depth data. This index is fed back to the trapped air point judgment step in real time to correct the temperature and pressure gradient thresholds. At the same time, it is input into the scheme evaluation step to dynamically adjust the weight allocation ratio of exhaust efficiency and capillary density in the weighted comprehensive evaluation algorithm.

[0070] Existing solutions rely solely on passive venting optimization using sensor data inside the mold, which cannot verify the correlation between actual product defects and trapped air determination. This solution introduces optical scanning and defect feedback mechanisms to establish a closed-loop verification link between trapped air determination results and actual product quality, thus solving the problem of disconnect between trapped air prediction and defect occurrence in existing technologies.

[0071] This application achieves dynamic calibration of the accuracy of trapped air determination, avoiding over-optimization or under-optimization caused by model errors. After the defect data is fed back to the comprehensive evaluation algorithm, the matching degree between the exhaust layout scheme and the product qualification rate can be improved, and the repetitive defects caused by trapped air can be reduced.

[0072] The method also includes mold core status monitoring and blockage early warning steps. In the real-time data acquisition step, thermal maps and pressure cloud maps of the mold core under working conditions are collected and compared with historical thermal maps under normal conditions. When an abnormal high-temperature gradient distribution is detected in the current thermal map corresponding to the capillary vent hole area, and the distributed pressure sensor detects an abnormal increase in mold core cavity pressure, and subsequent process verification steps confirm that the product has defects caused by trapped air, the capillary vent hole of the mold core is determined to be blocked, and a blockage alarm signal is generated. When a blockage alarm signal is received, the continuous working time from when the mold core is put into use until the first blockage alarm signal is generated is recorded, and this time is used as a durability reference index and introduced into the weighted comprehensive evaluation algorithm of the scheme evaluation step to adjust the weight allocation of subsequent venting layout optimization schemes.

[0073] Abnormal high temperature gradient distribution refers to the temperature change rate in the area surrounding the capillary vent holes exceeding the temperature change threshold under historical normal conditions. Specifically, thermal imaging detection devices can be used to capture temperature field data, and gradient calculation algorithms can be used to identify areas of temperature abrupt change. Abnormal pressure increase refers to the pressure value inside the mold cavity exceeding the preset pressure fluctuation range during injection. Specifically, distributed pressure sensors can be used to collect pressure data in real time and compare it with historical normal pressure curves. Blockage alarm signal refers to early warning information generated based on the correlation between temperature and pressure anomalies. Specifically, a logic judgment module can be used to synchronously analyze abnormal temperature gradients and abnormal pressures, triggering an alarm when both occur simultaneously. Durability reference index refers to the length of time from when the mold is put into use to when the first blockage occurs. Specifically, a timing module can be used to record the working cycle of the mold, and this data can be used as a basis for evaluating the stability of the venting layout scheme.

[0074] During the mold core's operation, a thermal imaging detection device continuously collects temperature field distribution data on the mold core surface. Simultaneously, a distributed pressure sensor monitors pressure changes inside the cavity. When the thermal map shows an abnormal high-temperature gradient in the capillary vent area, and the pressure sensor detects a pressure value exceeding the normal fluctuation range, the system automatically correlates this with the defect detection results in subsequent process verification steps. If the product exhibits defects caused by trapped air at this time, it is determined that the capillary vent is blocked, triggering a blockage alarm signal. The system records the continuous operating time of the mold core from the start of use to the first alarm trigger and uses this time as a durability reference indicator. In subsequent solution evaluation steps, this indicator is introduced into a weighted comprehensive evaluation algorithm to adjust the weight allocation of the vent layout optimization scheme, for example, prioritizing the recommendation of schemes with higher durability.

[0075] Existing mold venting optimization methods rely solely on static parameter adjustments, failing to monitor the mold core's working status in real time and lacking an early warning mechanism for capillary vent blockage. This solution, however, integrates dynamic monitoring data from temperature and pressure fields, combined with historical status comparisons and defect verification, to proactively identify blockage risks. Furthermore, by incorporating durability indicators into the evaluation system, the venting layout scheme not only focuses on short-term venting efficiency but also considers long-term stability. This application can identify capillary vent blockage risks in real time during production, triggering timely maintenance warnings and preventing continuous defective products due to venting failure. By introducing durability reference indicators, the evaluation system of the optimization scheme can dynamically balance venting efficiency and mold lifespan, providing an venting layout optimization strategy for the electric push-shear blade MIM process that balances short-term quality and long-term stability.

[0076] This method adds an adaptive weight optimization step after the core status monitoring and blockage early warning steps. It records the occurrence frequency of each blockage alarm signal and the location distribution of trapped air points under the same exhaust layout scheme. Based on the blockage frequency and distribution, it establishes a correlation model between the blockage risk of capillary exhaust holes and the layout scheme parameters. According to this model, the weighted comprehensive evaluation algorithm in the scheme evaluation step is adjusted to dynamically increase the proportion of relevant exhaust structure data with blockage frequency exceeding the preset value in the weight allocation.

[0077] The adaptive weight optimization step refers to adjusting the weight allocation of evaluation indicators by analyzing historical blockage event data. Specifically, this can be achieved by using data mining algorithms to analyze the correlation between blockage frequency and exhaust structure parameters. This step can identify easily blocked exhaust port layout patterns and enhance their weights. The frequency of blockage alarm signals refers to the number of times the capillary exhaust ports in the mold core trigger a blockage alarm per unit time. This can be achieved by counting alarm logs and dividing them into time windows using a statistics module. This parameter reflects the anti-blockage capability of a specific exhaust layout scheme. The location distribution of trapped air points refers to the set of coordinates of defects caused by trapped air in the mold cavity. This can be achieved by acquiring defect location data using an optical scanner and mapping it to the mold coordinate system. This data is used to locate weak areas in the exhaust structure. The correlation model refers to a predictive model describing the mathematical relationship between exhaust port layout parameters and blockage risk. This can be achieved by using regression analysis or machine learning algorithms to train on historical data. This model provides a quantitative basis for weight adjustment. Dynamic weight allocation adjustment refers to changing the importance coefficient of the evaluation indicators based on real-time updated blockage risk data. This can be achieved by iteratively correcting the initial weights using a weighting factor calculation module. This mechanism makes the evaluation algorithm adaptive.

[0078] During the operation of the model, each time a blockage alarm is triggered, the corresponding exhaust layout parameters and trapped air location data are recorded. After multiple data accumulations, the causal relationship between different exhaust hole distribution densities, diameters, or arrangements and blockage frequency is analyzed through a correlation model. For example, when blockages frequently occur in dense areas of certain capillary exhaust holes, the model marks such layout parameters as high-risk factors. In subsequent scheme evaluations, the weight of exhaust structure data involving high-risk factors in the comprehensive score is dynamically increased, prompting the optimized scheme to prioritize avoiding layout patterns prone to blockage. Thus, the evaluation algorithm can continuously optimize the weight allocation strategy based on actual production data, forming a closed-loop feedback mechanism.

[0079] Existing technologies, such as the prior art CN223056725U, rely solely on mechanical agitation to passively improve exhaust performance. This approach cannot predict the location of trapped air points, nor can it optimize exhaust structure design based on historical blockage data. This solution proactively identifies high-risk layout parameters by establishing a correlation model and dynamically adjusts evaluation weights based on real-time data. This transforms exhaust scheme design from experience-driven to data-driven, effectively reducing the risk of repeated blockages. It can continuously optimize the evaluation criteria of the exhaust layout scheme based on blockage event data in actual production, ensuring that the final selected scheme achieves a balance between exhaust efficiency and anti-blockage performance. This extends the service life of the mold and reduces part defects caused by exhaust structure failure.

[0080] This method also includes a weighted comprehensive evaluation algorithm in the scheme evaluation step, which is performed as follows: Evaluation indicators are determined, including the exhaust efficiency index and capillary exhaust hole distribution density; initial weights are assigned to each indicator; the initial weights of each indicator are dynamically adjusted based on the entrapment defect severity index from the subsequent process verification step, the durability reference index from the mold core status monitoring and blockage early warning step, and the blockage frequency from the adaptive weight optimization step; and the comprehensive score of each exhaust layout optimization scheme is calculated and ranked according to the adjusted weights.

[0081] The exhaust efficiency index is a quantitative indicator of the exhaust structure's ability to expel gas from the mold core, calculated through numerical simulation. Specifically, it can be achieved using a fluid dynamics simulation model combined with gas diffusion rate parameters, reflecting the actual exhaust effect of different exhaust layout schemes. The initial weight refers to the fixed weight allocation ratio of the evaluation indicators before dynamic adjustment is introduced. This can be set using expert experience or the analytic hierarchy process (AHP) to establish a preliminary evaluation benchmark. Dynamic adjustment refers to real-time correction of the weight parameters based on defect data, mold core durability indicators, and clogging frequency generated during actual production. This can be achieved using a linear regression model or fuzzy logic algorithm, adapting the evaluation system to optimization needs under different operating conditions. In the scheme evaluation process, basic evaluation indicators are first constructed based on the exhaust efficiency index and capillary exhaust hole distribution density, and initial weights are assigned to form a preliminary scoring model. Subsequently, the weight of the exhaust efficiency index is corrected by introducing a gas trapping defect severity index, and the weight ratio of the capillary exhaust hole distribution density is adjusted using durability reference indicators. Simultaneously, the two indicators are correlated and compensated using clogging frequency data. Finally, a comprehensive score for each scheme is obtained through weighted calculation, achieving dynamic adaptation between the evaluation results and actual production data.

[0082] Existing evaluation methods only rank schemes based on fixed weights, which cannot dynamically optimize according to actual production defects, mold core losses, and exhaust structure blockage risks. This application improves the accuracy and adaptability of exhaust layout scheme selection by integrating multi-source data to correct weight parameters in real time, enabling the evaluation system to reflect priority changes under different operating conditions.

[0083] This application can dynamically adjust the weight of evaluation indicators based on actual production data, so that the evaluation results of the exhaust layout optimization scheme are more in line with actual process requirements, effectively balance the contradiction between exhaust efficiency and mold core service life, and reduce maintenance costs caused by exhaust structure blockage.

[0084] This method also includes a scheme evaluation step. Based on the pre-established correlation between capillary vent density and blockage duration, as well as the correlation between capillary vent density and exhaust efficiency, the final optimization output step selects the corresponding capillary vent density optimization scheme according to user needs. If the user selects the exhaust efficiency priority strategy, an optimization scheme that increases capillary vent density is generated. If the user selects the service life priority strategy, an optimization scheme that decreases capillary vent density is generated.

[0085] The correlation between capillary venting density and clogging duration refers to the negative correlation between the number of venting holes and the continuous working time of the mold core, established through historical data statistics. Specifically, this can be achieved by using regression analysis to fit and model the clogging time of different density schemes in multiple production cycles. This relationship is used to predict the durability of the mold core under different density schemes. The correlation between capillary venting density and venting efficiency refers to the positive correlation between the number of venting holes and the gas discharge rate. Specifically, this can be achieved by using fluid dynamics simulation to calculate the change in mold cavity pressure gradient under different densities. This relationship is used to evaluate the potential for improving venting efficiency. The user-demand-priority strategy refers to the trade-off between venting efficiency and mold life based on production goals. Specifically, this can be achieved by receiving user input of quality standards and cost constraint parameters through a human-machine interface.

[0086] In the scheme evaluation step, a quantitative model is established between capillary vent density, blockage risk, and venting efficiency through numerical simulation and historical data analysis. When the user selects venting efficiency as the priority, the system generates a layout scheme that increases the number of vents based on the density-efficiency model, thereby reducing the probability of trapped air defects by increasing the venting speed. When the user selects service life as the priority, the system generates a layout scheme that reduces the number of vents based on the density-blockage model, thereby extending the mold core maintenance cycle by reducing the risk of vent blockage. Both strategies automatically generate corresponding capillary vent distribution parameters through the correlation model and integrate them into the optimization scheme for execution at the production terminal.

[0087] Traditional mold venting designs employ fixed venting structures, making it impossible to dynamically adjust venting density according to production needs. For example, in the prior art, mechanical agitation is used to passively improve venting performance, which cannot optimize venting hole distribution for different part structures or achieve an intelligent balance between efficiency and lifespan. This solution establishes a quantitative correlation model, enabling the venting layout to automatically generate the optimal density configuration based on user-defined priorities. This solves the problems of over-design or under-design caused by experience-based design in traditional methods. This application can achieve an intelligent balance between venting efficiency and mold lifespan based on actual production needs. When product precision requirements are high, venting efficiency is prioritized to effectively reduce surface defects caused by trapped air. When production cost control is strict, extending mold lifespan is prioritized to reduce maintenance frequency and spare parts consumption. This configurable optimization strategy significantly improves the adaptability and economy of the mold venting system.

[0088] This method also includes a predictive maintenance step. Based on the continuous working time, frequency of occurrence and location distribution data of blockage alarm signals obtained in the mold core status monitoring and blockage early warning steps, a blockage prediction model for the capillary vent holes of the mold core is established. According to the output of the blockage prediction model, the remaining service life estimate of the mold core is calculated. When the remaining service life estimate is lower than a preset threshold, a predictive maintenance plan containing maintenance time and spare parts information is generated and the predictive maintenance plan is output to the production terminal.

[0089] A blockage prediction model is a mathematical model built based on historical data of the mold core's working state. Specifically, it can use machine learning algorithms to train data on continuous working duration, blockage frequency, and location distribution to predict future trends in capillary vent blockage. The remaining service life estimate is the estimated time the mold core can continue to be used before severe blockage occurs, calculated by the blockage prediction model. This can be achieved using regression analysis or time series forecasting methods. A predictive maintenance plan is a maintenance scheme that includes maintenance time points, the type and quantity of spare parts to be replaced, and can be automatically generated by linking the spare parts inventory database with the production scheduling system. During mold core operation, real-time data on continuous working duration, the frequency of blockage alarm signals, and their location distribution are collected and input into the blockage prediction model for analysis. This model predicts the probability and timing of future blockage of the mold core's capillary vents by identifying the time-series patterns and spatial distribution characteristics of blockage events. When the remaining service life estimate falls below a preset threshold, the system automatically triggers the maintenance plan generation process, combining spare parts inventory data and production scheduling to generate a maintenance plan. This solution sends maintenance instructions, including maintenance time and spare parts list, to the production terminal. This allows maintenance operations to be performed before the mold core completely fails, preventing production interruptions due to sudden blockages. Compared to existing technologies like CN223056725U, which only passively improves venting through mechanical agitation and doesn't address mold core blockage prediction or maintenance strategies, this solution establishes a data-driven blockage prediction model that directly links mold core status monitoring data with maintenance decisions. This proactively predicts blockage risks and plans maintenance timing. Existing technologies rely on manual experience for maintenance operations, while this solution provides an objective basis for maintenance planning by quantifying remaining service life estimates. This significantly reduces production losses due to sudden mold core failures and enables proactive early warning of blockage risks in the mold core capillary vents, preventing batch defects in parts caused by undetected vent blockages. Through the linkage between predictive maintenance plans and the production terminal, maintenance operations can be precisely matched to the production rhythm, reducing unplanned downtime. Furthermore, dynamically adjusting maintenance strategies based on remaining service life estimates can extend mold core lifespan and reduce costs associated with frequent mold core replacements.

[0090] In the optimization scheme generation step, the characteristic parameters of the mold cavity also include mold core hardness. The minimum spacing between capillary vent holes is set according to the mold core hardness. When the mold core hardness is higher, the minimum spacing between capillary vent holes is smaller.

[0091] Mold core hardness refers to the mold core material's resistance to deformation. It can be measured using a Rockwell hardness tester or a Vickers hardness tester. The higher the mold core hardness, the lower the risk of plastic deformation of the material under injection pressure. The shortest distance between capillary vent holes refers to the minimum allowable distance between the centers of adjacent vent holes. It can be dynamically calculated through numerical simulation combined with the mechanical property parameters of the mold material. The correlation between mold core hardness and capillary vent hole spacing can be obtained through a pre-established mold material database.

[0092] When generating the venting layout optimization scheme, the hardness of the mold core is input as a key parameter into the capillary venting hole distribution rules. When the hardness of the mold core is in a high range, due to the strong resistance to deformation of the material, the spacing of the capillary venting holes can be set to a smaller value, thereby arranging more venting holes per unit area to improve venting efficiency. When the hardness of the mold core is low, in order to avoid the densely distributed venting holes weakening the strength of the mold structure, the minimum spacing needs to be increased to prevent the mold core from cracking under injection pressure. This spacing control is achieved by calling the pre-stored mold material parameter reference table. At the same time, the stress distribution state of the mold core structure under different hardnesses is verified by combining finite element analysis, and finally the minimum spacing of venting holes that meets the strength requirements is determined.

[0093] Traditional mold venting designs typically only consider the geometric features of the parts while neglecting the properties of the mold material. This leads to structural risks in the venting hole layout. For example, the agitator ribbon structure disclosed in the prior art does not address the correlation between mold hardness and venting hole arrangement. In practical applications, this may result in both mold cracking in densely vented areas and air trapping defects in areas with insufficient venting. This solution establishes a quantitative relationship between mold core hardness and venting hole spacing, achieving synergistic optimization of venting efficiency and mold structural reliability. This application effectively solves the matching problem between mold material properties and venting structure design. It maximizes venting efficiency while ensuring the integrity of the mold core structure. The minimum venting hole spacing is dynamically adjusted according to the mold core hardness, avoiding the waste of venting capacity caused by overly conservative design in high-hardness molds and preventing structural failure caused by excessively dense venting holes in low-hardness molds. This achieves precise adaptation between the venting layout scheme and mold material properties.

[0094] This method also includes a knowledge base construction step, which is used to establish a historical database for storage. The historical database is used to store historical model design parameters, exhaust layout schemes, corresponding thermal data, corresponding pressure data, and entrapment point determination results. In the scheme evaluation step, historical success cases and failure data in the historical database are called up, and data are provided for the exhaust efficiency and risk coefficient assessment of the current optimization scheme through similarity matching.

[0095] The historical database refers to a collection that stores different mold core design schemes and their corresponding process parameters. Specifically, it can be implemented using a relational database or a time-series database. It is used to record historical data such as mold core structural parameters, exhaust layout schemes, temperature field distribution, pressure field distribution, and entrapment point determination results. Among them, similarity matching refers to calculating the feature similarity between the current mold core design parameters and historical cases. Specifically, it can be implemented using Euclidean distance or cosine similarity algorithms to select reference cases from the historical database that are closest to the current optimization requirements.

[0096] During the evaluation process, successful cases in the historical database were used to identify effective exhaust layout patterns, while failure data were used to avoid potential design risks. By calculating the similarity between the current mold core's tooth clearance depth, width, and number of teeth and historical data, a set of cases with similar structural features was selected. Subsequently, based on the exhaust efficiency index and trapped air defect records of these cases, the performance of the current optimization scheme in actual production was predicted, and data support was provided for adjusting the exhaust hole distribution density.

[0097] In some specific implementations, the historical database can be integrated with machine learning models to extract features from historical heat maps and pressure cloud maps, establish association rules between core structure parameters and the distribution of trapped air points, and when performing similarity matching, quickly locate the historical exhaust layout scheme with the highest matching degree with the current tooth model data by searching the association rule base, thereby shortening the evaluation cycle of the optimization scheme.

[0098] Existing mold exhaust optimization methods rely on single process data and lack systematic utilization of historical experience, resulting in randomness in scheme evaluation. This scheme, however, by constructing a historical database and introducing a similarity matching mechanism, can transform past successful experiences and lessons learned into quantitative reference indicators, effectively avoiding the defects of repeated design, and improving the accuracy and reliability of exhaust layout optimization.

[0099] This application solves the problem that traditional MIM mold venting design relies on manual experience and lacks data support. By using structured storage and intelligent retrieval of historical data, it provides a traceable reference for venting layout optimization, significantly reducing the risk of part defects caused by trapped air, while improving the design efficiency of mold venting structure.

[0100] This method also includes a working data acquisition step, which includes setting an interval detection cycle for the mold core based on the changing pattern of the temperature field data obtained by the thermal imaging detection device; within the interval detection cycle, if the distributed pressure sensor detects that the pressure value of the mold core cavity exceeds the preset pressure threshold, it triggers supplementary detection of the mold core temperature field data.

[0101] The interval detection period refers to the periodic detection frequency determined based on the dynamic change trend of temperature field data. Specifically, it can be achieved by analyzing the correlation between temperature fluctuation amplitude and time series. For example, when the temperature change rate is lower than a set threshold, the detection interval is extended, and when the temperature change rate is higher than the set threshold, the detection interval is shortened. This feature is used to balance detection resource consumption and data acquisition accuracy, and avoid resource waste or data loss caused by fixed detection frequency.

[0102] Supplementary detection refers to the temporary acquisition of temperature field data triggered when pressure is abnormal. Specifically, it involves real-time monitoring of the mold cavity pressure through distributed pressure sensors. When the pressure value exceeds a preset threshold, a thermal imaging detection device is immediately activated for supplementary scanning. This feature is used to capture the correlation between pressure and temperature anomalies, ensuring the timeliness and accuracy of air entrapment point detection.

[0103] During the mold core's operation, the temperature field detection frequency is dynamically adjusted based on the changing patterns of thermal imaging data. For example, the detection interval is extended when the temperature field tends to stabilize, and the detection frequency is increased when the temperature fluctuates drastically. Simultaneously, the mold core cavity pressure is continuously monitored by distributed pressure sensors. When the detected pressure value exceeds the preset threshold, even if the current set detection cycle has not been reached, the thermal imaging detection device is immediately triggered to perform a supplementary scan of the mold core temperature field. This dual triggering mechanism ensures detection efficiency under normal conditions and can promptly acquire temperature field data when there is a sudden pressure anomaly, providing complete data support for subsequent air trapping point determination.

[0104] Compared with existing technologies, which typically employ fixed-frequency detection methods (e.g., the prior art only periodically detects the mold core state), the detection frequency cannot be dynamically adjusted according to actual operating conditions, easily leading to wasted detection resources or omission of key data. This solution, however, dynamically sets the detection cycle based on temperature field changes and combines it with pressure anomaly triggering supplementary detection, forming an adaptive detection mechanism. This significantly reduces the operating load of the detection equipment while ensuring data integrity. This application effectively solves the problem of mismatch between detection frequency and operating condition changes in existing technologies. It can intelligently adjust the detection strategy according to the actual working state of the mold core, avoiding resource waste caused by over-detection and timely capturing temperature field data when pressure is abnormal, ensuring the accuracy of trapped gas point determination. Simultaneously, the dual-trigger mechanism prevents the omission of sudden abnormal data caused by fixed detection cycles, providing a more reliable data foundation for exhaust gas optimization.

Claims

1. A method for optimizing the exhaust parameters of an electric shaver blade MIM powder metallurgy process mold, characterized in that, The method comprises the following steps: a die core data acquisition step of acquiring characteristic parameters of a die core cavity and tooth part model data of an electric clipper blade; a working data acquisition step of acquiring temperature field and pressure field distribution data of the die core in a working state based on a monitoring area determined according to the characteristic parameters of the die core cavity and the tooth part model data, and through a thermal imaging detection device for thermal imaging of the die core and a distributed pressure sensor arranged in the die core; a trapped gas point determination step of determining a trapped gas point in the die core cavity based on the temperature field and pressure field distribution data and in combination with the tooth part model data, through image fusion and gradient analysis algorithms; an optimization scheme generation step of generating a plurality of exhaust layout optimization schemes according to position information of the trapped gas point and the tooth part model data; a scheme evaluation step of evaluating exhaust efficiency indexes in the plurality of exhaust layout optimization schemes by respectively calculating pressure field distribution, temperature field change and capillary exhaust hole distribution parameters of the die core cavity under each exhaust layout optimization scheme through a numerical simulation method, and sorting the plurality of exhaust layout optimization schemes based on a weighted comprehensive evaluation algorithm; wherein the exhaust efficiency index is an index for quantifying the exhaust capacity of the exhaust structure in the die core obtained through numerical simulation; the weighted comprehensive evaluation algorithm comprises: determining evaluation indexes, the evaluation indexes comprising the exhaust efficiency index and capillary exhaust hole distribution density; assigning initial weights to the evaluation indexes; dynamically adjusting the initial weights based on external feedback signals; and calculating comprehensive scores of the plurality of exhaust layout optimization schemes according to the adjusted weights and sorting the plurality of exhaust layout optimization schemes according to the comprehensive scores; a final optimization output step of selecting an exhaust layout scheme most suitable for user requirements from the plurality of optimization schemes according to a comprehensive result of the scheme evaluation step and specific requirements input by the user, and outputting the determined exhaust layout optimization scheme to a production terminal.

2. The method of optimization of the venting parameters of the MIM powder metallurgy process die for the electric clipper blade of claim 1, characterized in that, Further comprising: a subsequent process verification step of configuring a high-precision optical scanning measuring instrument to detect defects of the electric clipper blade produced by the trapped gas point, and acquire defect data caused by the trapped gas point; based on the defect data, evaluating the influence degree of the trapped gas point on product quality through an image registration and difference comparison algorithm, calculating a trapped gas defect severity index, and feeding back the trapped gas defect severity index to the trapped gas point determination step and the scheme evaluation step; adjusting the weighted comprehensive evaluation algorithm in the scheme evaluation step according to the trapped gas defect severity index.

3. The method of optimization of the venting parameters of the MIM powder metallurgy process die for the electric clipper blade of claim 2, characterized in that, The working data acquisition step generates corresponding thermal maps and pressure cloud maps through the temperature field and pressure field distribution data of the die core in the working state; The exhaust parameter optimization method further comprises: a die core state monitoring and blockage early warning step of acquiring thermal maps and pressure cloud maps of the die core in the working state in the working data acquisition step, comparing the thermal maps with historical normal state thermal maps, determining that the capillary exhaust hole of the die core is blocked when an abnormally high temperature gradient distribution is detected in the capillary exhaust hole corresponding area of the current thermal map, the pressure in the die core cavity abnormally rises, and the subsequent process verification step confirms that the product has defects caused by trapped gas, and generating a blockage alarm signal. When the blockage alarm signal is received, the continuous working time from the die insert being put into use to the first generation of the blockage alarm signal is recorded, and the time is taken as a durability reference index, which is introduced into the weighted comprehensive evaluation algorithm of the scheme evaluation step, and is used to adjust the weight distribution of the subsequent exhaust layout optimization scheme.

4. The exhaust parameter optimization method of the MIM powder metallurgy process die for electric hair clippers according to claim 3, characterized in that: After the die insert state monitoring and blockage early warning step, further comprising: An adaptive weight optimization step, which records the occurrence frequency of each blockage alarm signal and the position distribution of the gas trapping points under the same exhaust layout scheme; Based on the occurrence frequency of the blockage alarm signal and the position distribution of the gas trapping points, an association model of the blockage risk of the capillary exhaust hole and the layout scheme parameters is established; According to the model, the weighted comprehensive evaluation algorithm in the scheme evaluation step is adjusted to dynamically increase the proportion of the related exhaust structure data in the weight distribution when the blockage frequency exceeds the preset value.

5. The method of optimization of the venting parameters of the MIM powder metallurgy process die for the electric clipper blade of claim 4, characterized in that, The weighted comprehensive evaluation algorithm in the scheme evaluation step is performed by the following way: Determine the evaluation index, which includes the exhaust efficiency index and the capillary exhaust hole distribution density; Assign initial weights to each index of the evaluation index; Based on the gas trapping defect severity index of the subsequent process verification step, the durability reference index of the die insert state monitoring and blockage early warning step, and the blockage frequency of the adaptive weight optimization step, the initial weights of the evaluation index are dynamically adjusted; According to the adjusted weights, the comprehensive scores of each exhaust layout optimization scheme are calculated and sorted accordingly.

6. The method of optimization of the venting parameters of the MIM powder metallurgy process die for the electric clipper blade of claim 5, characterized in that, The scheme evaluation step is also based on the pre-established association between the capillary exhaust hole distribution density and the blockage time, and the association between the capillary exhaust hole distribution density and the exhaust efficiency, The final optimization output step selects the corresponding capillary exhaust hole density optimization scheme according to user demand, wherein: If the user selects the exhaust efficiency priority strategy, an optimization scheme that increases the capillary exhaust hole density is generated; If the user selects the service life priority strategy, an optimization scheme that reduces the capillary exhaust hole density is generated.

7. The method of optimization of the venting parameters of the MIM powder metallurgy process die for the electric clipper blade of claim 3, characterized in that, Further comprising: A predictive maintenance step, based on the continuous working time, the occurrence frequency and the position distribution data of the blockage alarm signal obtained in the die insert state monitoring and blockage early warning step, a blockage prediction model of the die insert capillary exhaust hole is established; According to the output of the blockage prediction model, the remaining service life estimate of the die insert is calculated, and when the remaining service life estimate is lower than the preset threshold, a predictive maintenance plan containing maintenance time and spare parts information is generated, and the predictive maintenance plan is output to the production terminal.

8. The method of optimization of venting parameters of MIM powder metallurgy process die for electric clipper blades as claimed in claim 3 wherein, The feature parameters of the die insert cavity include the layout information of the die insert exhaust main flow channel and the capillary exhaust hole, and the die insert hardness. In the optimization scheme generation step, the shortest distance between the capillary exhaust holes is set according to the die insert hardness. The higher the die insert hardness, the smaller the shortest distance between the capillary exhaust holes.

9. The method of optimization of the venting parameters of the MIM powder metallurgy process die for the electric clipper blade of claim 1, characterized in that, Further comprising: A knowledge base construction step for establishing a historical database for storage, which is used to store historical die insert design parameters, exhaust layout schemes, corresponding thermal data, corresponding pressure data, and gas trapping point determination results; In the scheme evaluation step, historical successful cases and failure data in the history database are called to provide data for the exhaust efficiency and risk coefficient evaluation of the current optimization scheme through similarity matching.

10. The method of optimization of venting parameters of MIM powder metallurgy process die for electric clipper blades as claimed in claim 1 wherein, The working data collection step comprises: Based on the change rule of the temperature field data obtained by the thermal imaging detection device, the interval detection period of the mold core is set; In the interval detection period, if the distributed pressure sensor detects that the pressure value of the mold core cavity exceeds the preset pressure value, the supplementary detection of the mold core temperature field data is triggered.

Images