A monitoring system for the production of a metal composite ceramic powder material

Abstract

The application discloses a kind of metal composite ceramic powder material preparation's supervision system, it is related to material preparation supervision technical field, including: potential source set generation module, for according to the actual inhalation nanodose of worker, superfine powder volume fraction and actual sintering density measured value generation potential source set;Potential intensity generation module is used for respectively to inhalation nanodose and sintering density measured value carry out power law mapping, obtain exposure layer potential intensity and density layer potential intensity;Potential field generation module is used for respectively to exposure layer potential intensity and density layer potential intensity carry out weighting aggregation operation, obtain the exposure layer potential field and density layer potential field under the proportion of measured batching;Predicted value generation module is used for respectively according to exposure layer potential field and density layer potential field generation at the proportion of measured batching predicted inhalation nanodose and predicted sintering density.The application solves the shortcoming that density improvement and exposure limit value conflict need artificial trade-off.

Description

Technical Field

[0001] This invention relates to the field of material preparation monitoring technology, and in particular to a monitoring system for the preparation of metal composite ceramic powder materials. Background Technology

[0002] With the rapid development of additive manufacturing and high-density powder metallurgy, metal-ceramic powder materials are widely used in high-end applications such as hot-end components for aero-engines, heat dissipation base plates for semiconductor packaging, and high-wear-resistant molds. The final performance of these materials is sensitive to sintering density. The industry generally adopts a graded batching strategy that incorporates a large number of ultrafine ceramic particles smaller than 100nm to utilize the interstitial effect to reduce porosity and increase the relative density of the sintered body. Meanwhile, production workshops require prolonged processes such as spray granulation, fluidized bed drying, and plasma spheroidization, which easily generate high concentrations of nanoparticle dust. Occupational health regulations in various countries set extremely low exposure limits for dust in this particle size range, measured in micrograms per cubic meter, and require companies to establish real-time monitoring, risk assessment, and dynamic ventilation systems. This creates a dual rigid demand in the production environment: on the one hand, the proportion of ultrafine powder must be increased to meet customer quality standards for high-density products; on the other hand, the instantaneous and cumulative nanoparticle dust dose in the breathing zone of operators must be strictly controlled to meet regulatory and occupational health requirements.

[0003] However, existing regulatory methods for powder preparation mainly focus on a single dimension: formulation engineers adjust the proportion of ultrafine powder based on experience or linear regression models, while safety engineers manage dust exposure using fixed airflow and shift schedules. These two approaches are independent of each other and lack a unified data-driven decision-making framework. When the proportion of ultrafine powder needs to be significantly increased, traditional methods can only rely on manual trial and error to repeatedly compromise between increasing density and occupational exposure limits. This leads to delayed ingredient adjustments, redundant ventilation, or frequent shift changes, easily resulting in a mismatch where the density has not yet met the standard but the dust exposure limit has already been reached. Alternatively, the proportion of ultrafine powder may be excessively reduced to meet exposure limits, resulting in insufficient sintering performance and affecting the delivery of high-end customer orders. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies where there is a conflict between density enhancement and exposure limits, requiring manual balancing, and to propose a monitoring system for the preparation of metal composite ceramic powder materials.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution:

[0006] A monitoring system for the preparation of metal composite ceramic powder materials, comprising:

[0007] The potential source set generation module is used to generate a potential source set based on the worker's actual inhaled nano-dose, ultrafine powder volume fraction, and actual measured sintering density.

[0008] The potential intensity generation module is used to perform power-law mapping on the inhaled nano-dose and the measured values ​​of sintering density to obtain the potential intensity of the exposed layer and the potential intensity of the dense layer.

[0009] The potential field generation module is used to perform weighted aggregation calculations on the potential strength of the exposed layer and the potential strength of the compacted layer, respectively, to obtain the potential field of the exposed layer and the potential field of the compacted layer under the tested ingredient ratio.

[0010] The prediction value generation module is used to generate the predicted inhaled nanodose and predicted sintering density at the ratio of the ingredients to be tested, based on the potential field of the exposed layer and the potential field of the density layer, respectively.

[0011] The objective function construction module is used to construct an objective function based on the predicted inhaled nano-dose and the predicted sintering density, and to establish the constraints of the objective function.

[0012] The parameter optimization module is used to optimize the parameters of the objective function under constraints to obtain the optimal parameters of the objective function.

[0013] The management plan development module is used to develop management plans for powder material preparation based on optimal parameters.

[0014] Preferably, a potential source set is generated based on the worker's actual inhaled nano-dose, the volume fraction of the ultrafine powder, and the measured values ​​of the actual sintering density, including:

[0015] For each batch The instantaneous inhalation volume of the worker is obtained by multiplying the concentration of particles smaller than 100 nm in the powder material preparation workshop with the worker's timely ventilation volume. The instantaneous inhalation volume is then multiplied with the sealing factor of the protective equipment to obtain the actual inhaled nano-dose.

[0016] For each batch Real-time collection of the ultrafine powder volume fraction and actual sintering density measured values ​​from the workers;

[0017] A potential source set is generated based on the actual inhaled nano-dose, the volume fraction of ultrafine powder, and the actual measured values ​​of sintering density.

[0018] Preferably, a power-law mapping is performed on the inhaled nano-dose and the measured sintering density to obtain the potential intensity of the exposed layer and the potential intensity of the dense layer, including:

[0019] The exposure layer potential of the inhaled nano-dose is calculated based on a preset power-law mapping relationship;

[0020] The density layer potential strength is calculated based on the pre-set power-law mapping relationship to determine the measured sintering density.

[0021] Preferably, a weighted aggregation operation is performed on the potential intensity of the exposed layer and the potential intensity of the compacted layer, respectively, to obtain the potential field of the exposed layer and the potential field of the compacted layer under the tested ingredient ratio, including:

[0022] The current volume fraction of ultrafine powder is compared with the first... using a preset kernel function. The distance between the volume fractions of ultrafine powder in different batches is converted into a weighting factor;

[0023] The potential field of the exposed layer is obtained by weighting and summing the weighting factor and the potential field of the exposed layer under the test ingredient ratio;

[0024] The potential field of the dense layer under the test ingredient ratio is obtained by weighting and summing the weighting factor and the density layer potential.

[0025] Preferably, the predicted inhalation nanodose and predicted sintering density at the tested ingredient ratio are generated based on the potential field of the exposed layer and the potential field of the density layer, respectively, including:

[0026] Calculate the first-order local gradients of the potential fields of the exposed layer and the compacted layer at the ratio of the ingredients to be measured, respectively.

[0027] The first-order local gradient is suppressed using a pre-defined set of gradient penalty factors to obtain the suppression result;

[0028] The suppression result and the potential field of the exposed layer are coupled and mapped to obtain the predicted inhaled nanodose.

[0029] The predicted sintering density is obtained by coupling the suppression result and the density layer potential field.

[0030] Preferably, the objective function is constructed based on the predicted inhaled nano-dose and the predicted sintering density, including:

[0031] The combined term predicting the inhaled nano-dose is used as the cost term of the objective function, and the combined term predicting the sintering density is used as the benefit term of the objective function.

[0032] Construct an objective function based on cost and benefit terms.

[0033] Preferably, the constraints on the objective function include:

[0034] Constraints for the objective function are established based on the upper limit of occupational exposure safety for predicted inhaled nano-dose and the minimum process requirements for predicted sintering density.

[0035] Preferably, a management plan for powder material preparation is formulated based on optimal parameters, including:

[0036] The optimal personnel scheduling scheme and optimal air supply and exhaust volume are determined based on the optimal ultrafine powder volume fraction in the optimal parameters.

[0037] A management plan is developed based on the optimal ultrafine powder volume fraction, the optimal personnel scheduling plan, and the optimal air supply and exhaust volume. The management plan includes: batching powder materials according to the optimal ultrafine powder volume fraction, arranging workers to perform operations according to the optimal personnel scheduling plan, and supplying and exhausting air according to the optimal air supply and exhaust volume.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] 1. In this invention, a three-level mapping from potential source to potential strength to potential field is first used to normalize the core parameters that originally belonged to different positions and different dimensions: ultrafine powder volume fraction, actual inhaled nano-dose, and sintering density into a continuous potential field on the same coordinate axis. The system collects the measured values ​​of ultrafine powder volume fraction and density for each batch in real time, generating a potential source set containing three types of elements. Then, power-law mapping is performed on the measured values ​​of actual inhaled nano-dose and actual sintering density to obtain the potential strength of the exposed layer and the potential strength of the density layer. Then, the kernel function is used to aggregate the potential fields of the exposed layer and the density layer according to the material distance to form the potential field of the exposed layer and the potential field of the density layer, ensuring that the influence of historical batches on the current material is both continuous and differentiable. Two originally isolated evaluation chains are compressed into the same numerical domain, and the formulation engineer and safety engineer share the same real-time digital twin interface.

[0040] 2. In this invention, based on a unified potential field, the first-order gradients of the potential fields of the exposure layer and the density layer are calculated simultaneously. After gradient penalty, they are mapped to the measured inhaled nano-dose and the predicted sintering density. The two predictions are assembled into a dual objective function that minimizes health risk and maximizes density. The gradient descent algorithm is used to quickly converge within the feasible region to obtain the optimal ultrafine powder volume fraction, thus avoiding the mismatch risk of reaching the exposure limit before the density is met or the density is insufficient to ensure safety due to powder reduction. Attached Figure Description

[0041] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0042] Figure 1 This is a functional block diagram of a monitoring system for the preparation of metal composite ceramic powder materials according to an embodiment of the present invention. Detailed Implementation

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0044] Example: This example provides a monitoring system for the preparation of metal composite ceramic powder materials. See [link to example]. Figure 1,include:

[0045] The potential source set generation module is used to generate a potential source set based on the worker's actual inhaled nano-dose, ultrafine powder volume fraction, and actual measured sintering density.

[0046] In embodiments of the present invention, a potential source set is generated based on the worker's actual inhaled nano-dose, the volume fraction of ultrafine powder, and the measured value of actual sintering density, including:

[0047] For each batch i, the concentration of particles smaller than 100 nm in the powder material preparation workshop is multiplied by the worker's timely ventilation to obtain the worker's instantaneous inhalation volume. The instantaneous inhalation volume is then multiplied by the sealing factor of the protective equipment to obtain the actual inhaled nano-dose.

[0048] In this embodiment, for each production batch i, a laser particle size analyzer with a particle size resolution of up to ten nanometers is first placed at the breathing zone height of the operating area in the powder material preparation workshop. The mass concentration of particles with a diameter less than one hundred nanometers is continuously monitored at sampling intervals of no more than one second, with the obtained concentration data expressed in micrograms per cubic meter. Next, each worker wears a wearable device with an integrated respiratory flow sensor on their chest to measure their instantaneous ventilation volume in cubic meters per minute. The system in the background cross-references the particle concentration data at the same moment with the corresponding worker's instantaneous ventilation volume and performs a product calculation to obtain the worker's instantaneous inhalation volume, in micrograms per minute. Subsequently, the sealing integrity of the protective equipment is evaluated in real time using an RFID tag and a micro-differential pressure sensor installed on the protective equipment, and a sealing factor is calculated, with a value ranging from 0 to 1, where a value closer to zero indicates better protection. The system then multiplies the instantaneous inhalation volume by the sealing factor to obtain the worker's actual inhaled nano-dose at that moment, also in micrograms per minute. Finally, the system continuously records the actual inhaled dose data at all times during a complete shift and captures the peak values ​​as input parameters for exposure assessment and subsequent optimization scheduling for this batch.

[0049] For each batch i, the volume fraction of ultrafine powder and the actual measured value of sintering density are collected in real time by the workers.

[0050] In this embodiment, to obtain batch-level process parameters in real time, an online laser diffraction particle size analyzer is first installed in series between the granulation spray outlet and the powder storage silo, with its optical path adjusted to cover the full range from 10 nanometers to 300 micrometers. The system automatically extracts a representative powder stream as the powder is conveyed through the pipeline, dilutes it with gas, and then enters the detection chamber, continuously outputting the volume fraction of ultrafine powder. Specifically, the volume fraction of ultrafine powder refers to the real-time percentage of the particle size distribution (smaller than 100 nanometers) in the entire particle size distribution. The analyzer completes a measurement every three seconds and writes the results into the process data table of the manufacturing execution system via industrial Ethernet. To avoid powder bed blockage and measurement drift, compressed air backflushing and reference standard calibration are automatically triggered every two hours of operation. Secondly, a dual-station automatic sampling robot is installed at the discharge end of the sintering cooling section. After each sintering cycle, two representative samples are grabbed after cooling to room temperature and sequentially fed into the automatic density measurement module according to a preset trajectory. This module employs a combined process of true density helium specific gravity and volume density Archimedes' method. First, helium is injected into a vacuum chamber to determine the volume of closed pores between particles. Then, the sample is immersed in temperature-controlled deionized water to determine the volume due to gravitational displacement. The system automatically calculates and outputs the actual measured density of the sintered body based on true density. The measured data is written back to the same batch data line in real time via the OPC-UA interface, along with the sampling timestamp and furnace number. This ensures a one-to-one correspondence between the ultrafine powder volume fraction and the corresponding actual measured sintered density in the database, providing traceable real-time raw data for subsequent potential source mapping and process optimization.

[0051] A potential source set is generated based on the actual inhaled nano-dose, the volume fraction of ultrafine powder, and the actual measured sintering density. The expression for the potential source set is as follows:

[0052]

[0053] In the formula, It is a collection of potential sources. This is the actual inhaled nano-dose in the i-th batch. It is the first Actual sintered density measured value for the batch. This is the volume fraction of ultrafine powder in the i-th batch. This is the total number of batches.

[0054] In this embodiment, the potential source set is used to organize the key raw data of historical batches into basic units that can be used for subsequent mathematical mapping. Each element contains three interrelated physical quantities: the volume fraction of ultrafine powder records the percentage of volume distribution of particles smaller than 100 nanometers in the i-th batch of feed, which is equivalent to the horizontal axis in the set and is used to define the position of each potential source on the response surface; the actual inhaled nanoparticle dose reflects the peak nanoparticle dust dose exposed to workers during the production process of this batch, which is mapped to the potential strength of the exposure layer in the subsequent model and determines the contribution weight of this batch to the occupational safety risk field; the actual sintering density measured value characterizes the density performance of the finished product of this batch, which is subsequently converted into the potential strength of the density layer and affects the level of the performance response field.

[0055] Overall, by simultaneously storing ingredient ratios, peak exposure values, and measured performance values ​​in the same data structure, the potential source set achieves coupled storage of three-dimensional information. This allows subsequent kernel function weighting to synchronously generate a safety potential field and a performance potential field on the same common coordinate axis, providing interpretable, traceable, and physically meaningful data support for ingredient optimization.

[0056] The potential intensity generation module is used to perform power-law mapping on the inhaled nano-dose and the measured values ​​of sintering density to obtain the potential intensity of the exposed layer and the potential intensity of the dense layer.

[0057] In embodiments of the present invention, power-law mapping is performed on the inhaled nano-dose and the measured values ​​of sintering density to obtain the potential intensity of the exposed layer and the potential intensity of the dense layer, including:

[0058] The exposure layer potential for inhaled nano-dose is calculated based on a pre-defined power-law mapping relationship. The formula for calculating the exposure layer potential is as follows:

[0059]

[0060] In the formula, The exposed layer potential is stronger in the i-th batch. It is the exposure normalization coefficient. It is the decay index. This is the actual inhaled nanodose in the i-th batch;

[0061] In detail, the power-law mapping relationship is a function that converts the original physical quantity into a dimensionless weight according to the normalization and exponential amplification rules: first, the data is scaled to a uniform dimension or interval using preset coefficients, and then the exponential operation is performed on it to make high values ​​stand out and low values ​​compress, finally obtaining a weight value that changes exponentially with the original quantity; this weight not only preserves the relative order of the original data, but also amplifies the sensitivity to extreme or critical intervals.

[0062] In this embodiment, to facilitate the conversion of inhaled nano-dose data from historical batches into weighted values ​​that can participate in field strength superposition, the system first presets a set of power-law mapping constants in the parameter library, including an exposure normalization coefficient and an attenuation exponent. When the database reads inhaled nano-dose data from the i-th batch, the mapping function module is called to perform dimensional consistency correction and interval clipping on the inhaled nano-dose data to ensure that it falls within the effective input range corresponding to the normalization coefficient. Then, a power-law operation is performed, and the normalized dose value is exponentially mapped according to the attenuation exponent to obtain the exposure layer potential strength of that batch. In respiratory toxicology, when the inhaled dose exceeds a certain level, the increase in harm to the body tends to slow down or saturate. Therefore, it is not necessary to further weight extremely high doses during the weighting stage. The dose is magnified multiple times, otherwise the subsequent weighted summation will be overwhelmed by a few extreme batches. Using a negative exponent can compress the excessively high dose into a limited range, but still maintain the relative order.

[0063] The density layer potential strength of the sintered density is calculated based on the preset power-law mapping relationship. The formula for calculating the density layer potential strength is as follows:

[0064]

[0065] In the formula, The density layer potential is stronger in the i-th batch. It is the density normalization coefficient. It is an amplified index. It is the actual measured value of sintering density in the i-th batch.

[0066] In this embodiment, after dimensional correction of the measured sintering density of each batch, power-law mapping is directly performed to amplify the correction value according to a preset exponent to obtain the density layer potential.

[0067] The potential field generation module is used to perform weighted aggregation calculations on the potential strength of the exposed layer and the potential strength of the compacted layer, respectively, to obtain the potential field of the exposed layer and the potential field of the compacted layer under the tested ingredient ratio.

[0068] In embodiments of the present invention, a weighted aggregation operation is performed on the potential intensity of the exposed layer and the potential intensity of the compacted layer to obtain the potential field of the exposed layer and the potential field of the compacted layer under the tested ingredient ratio, including:

[0069] The distance between the current ultrafine powder volume fraction and the ultrafine powder volume fraction of the i-th batch is converted into a weighting factor using a preset kernel function, which is as follows:

[0070]

[0071] In the formula, It is a weighting factor. It is the attenuation coefficient. It is the absolute value of the difference between the volume fraction of ultrafine powder under the tested ingredient ratio and the volume fraction of ultrafine powder under the i-th batch. It is the volume fraction of ultrafine powder under the tested ingredient ratio. It is the volume fraction of ultrafine powder in the i-th batch;

[0072] In detail, the preset kernel function maps the batching distance to a continuous weight of 0-1, so that when the potential field is superimposed, it can both highlight the contribution of nearby batches and smoothly weaken the interference of distant batches: the constant term ensures that the weight is always one when the distance is zero, ensuring that the self-weight of the current batch is not reduced; the first and second polynomial terms provide linear slope and curvature adjustment for the nearest neighbor interval, respectively, so that the weight curve has sufficient resolution in the small distance segment and can sensitively distinguish slight batching differences; the exponential decay term uses the square root distance as the exponential independent variable to form a long-tailed decay curve that is smoother than linear or square distance, which can suppress distant noise and avoid the weight from dropping to zero too early; the whole is multiplied by the same scaling factor so that the curve decay rate can be adjusted according to process experience.

[0073] In this step, the volume fraction of the ultrafine powder in the current batch to be tested is compared with the data in the database. The volume fractions of ultrafine powder recorded in each batch are subtracted and the absolute value is taken to obtain the distance quantification result between the two batches in terms of ingredients. Then, this distance is used as the only input to a preset kernel function. The preset kernel function takes the distance as the independent variable and can automatically output a weighting factor between zero and one according to the distance: the smaller the distance, the closer the weighting factor is to one; the larger the distance, the weighting factor decreases smoothly according to the decay law defined by the kernel function.

[0074] The potential field of the exposed layer is obtained by weighting and summing the weighting factor and the potential field of the exposed layer under the tested ingredient ratio. The formula for calculating the potential field of the exposed layer is as follows:

[0075]

[0076] In the formula, It is the potential field of the exposed layer. It is the volume fraction of ultrafine powder under the tested ingredient ratio. The exposed layer potential is stronger in the i-th batch. It is a weighting factor. It is the absolute value of the difference between the current volume fraction of ultrafine powder and the volume fraction of ultrafine powder in the i-th batch. This is the total number of batches;

[0077] In this step, each historical batch has only one point-like occupational exposure data point, dispersed across different ultrafine powder volume fractions. The exposure layer potential strength obtained for each historical batch is multiplied by its corresponding weighting factor and then summed, continuously mapping the discrete occupational exposure data to a single value under the current tested ingredient ratio, forming an exposure layer potential field. The potential strength itself represents the risk intensity of inhaling nano-dose in that batch. By weighted summing of similarity and risk intensity, the system can achieve physically interpretable integration of all historical data without relying on empirical thresholds or complex black-box models, obtaining a smooth and numerically differentiable exposure layer potential field value. This provides a reliable and safe foundational input for subsequent gradient correction and ingredient optimization.

[0078] The density layer potential field is obtained by weighting and summing the weighting factor and the density layer potential. The formula for calculating the density layer potential is as follows:

[0079]

[0080] In the formula, It is the potential field of the compacted layer. It is the volume fraction of ultrafine powder under the tested ingredient ratio. The density layer potential is stronger in the i-th batch. It is a weighting factor. It is the absolute value of the difference between the current volume fraction of ultrafine powder and the volume fraction of ultrafine powder in the i-th batch. This is the total number of batches.

[0081] The prediction value generation module is used to generate the predicted inhaled nanodose and predicted sintering density at the ratio of the ingredients to be tested, based on the potential field of the exposed layer and the potential field of the density layer, respectively.

[0082] In embodiments of the present invention, the predicted inhalation nanodose and predicted sintering density at the tested ingredient ratio are generated based on the potential field of the exposed layer and the potential field of the density layer, respectively, including:

[0083] Calculate the first-order local gradients of the potential fields of the exposed layer and the compacted layer at the ratio of the ingredients to be measured, respectively.

[0084] The first-order local gradient is suppressed using a pre-defined set of gradient penalty factors to obtain the suppression result;

[0085] The suppression result and the potential field of the exposed layer are coupled and mapped to obtain the predicted inhaled nanodose. The calculation formula for the predicted inhaled nanodose is as follows:

[0086]

[0087] In the formula, This is the predicted inhaled nanodose under the tested ingredient ratio. It is the volume fraction of ultrafine powder under the tested ingredient ratio. It is the potential field of the exposed layer. It is the first-order local gradient of the potential field of the exposed layer at the ratio of the ingredients to be measured. It is the first penalty factor in the preset set of gradient penalty factors;

[0088] The suppression result and the density layer potential field are coupled and mapped to obtain the predicted sintering density. The calculation formula for the predicted sintering density is as follows:

[0089]

[0090] In the formula, It is the predicted sintering density under the tested ingredient ratio. It is the volume fraction of ultrafine powder under the tested ingredient ratio. It is the potential field of the compacted layer. It is the first-order local gradient of the density layer potential field at the measured ingredient ratio. It is the second penalty factor in the preset set of gradient penalty factors.

[0091] In detail, coupling mapping combines the potential field value and the gradient suppression result, two interrelated quantities, in the same step of calculation. The result is transformed into a new output value through a mapping function. This retains the historical data weights reflected by the potential field, while simultaneously mapping local fluctuations into the result through gradient penalty, forming a prediction that is both accurate and stable.

[0092] In this step, the first-order local gradients of the potential fields of the exposed layer and the density layer are obtained by performing differential operations at the ratio of the ingredients to be tested, so as to characterize the rate of change of the potential field at this ratio of ingredients. The absolute value of the first-order local gradient is weighted and suppressed by the first penalty factor / second penalty factor in the preset gradient penalty factor set, so as to smooth the gradient with drastic changes and obtain the suppression result. The coupling mapping is realized by floating-point division, that is, the original potential field value is divided by the gradient suppression result containing the penalty term to directly obtain the predicted inhaled nanodose and the predicted sintering density.

[0093] In general, the predicted inhaled nano-dose and predicted sintering density at the tested ingredient ratio are generated based on the potential fields of the exposed layer and the density layer, respectively. The purpose is to calculate the first-order local gradient of the potential fields of the exposed layer and the density layer, suppress the gradient by combining it with a preset set of gradient penalty factors, and then couple and map it with the potential field. Thus, considering the smoothness of the potential field change rate, the potential field mapped from historical batch data is transformed into a quantifiable predictive value at a specific ingredient ratio. This provides a physically interpretable and numerically reliable input for the subsequent construction of the objective function and optimization of the ingredient parameters, so as to achieve accurate prediction of occupational exposure risk and sintering performance during the preparation of metal composite ceramic powder materials, and thus provide a scientific basis for optimizing the preparation process and formulating management plans.

[0094] The objective function construction module is used to construct an objective function based on the predicted inhaled nano-dose and the predicted sintering density, and to establish the constraints of the objective function.

[0095] In embodiments of the present invention, an objective function is constructed based on the predicted inhaled nanodose and the predicted sintering density, including:

[0096] The combined term predicting the inhaled nano-dose is used as the cost term of the objective function, and the combined term predicting the sintering density is used as the benefit term of the objective function.

[0097] Construct an objective function based on the cost and benefit terms, as follows:

[0098]

[0099] In the formula, It is the objective function. It is the volume fraction of ultrafine powder under the tested ingredient ratio. This is the predicted inhaled nanodose under the tested ingredient ratio. It is the predicted sintering density under the tested ingredient ratio. It is the exposure loss weight. It is the density gain weight.

[0100] Based on actual production and safety needs, the predicted inhaled nano-dose is directly related to the occupational exposure risk of personnel during the production process. The higher the dose, the greater the health hazard, which needs to be suppressed as a cost. The predicted sintering density is related to product performance and quality. The higher the density, the stronger the product value and practicality, which needs to be enhanced as a benefit. Based on this logic, the technical solution first processes the two types of indicators by combining the first and second terms of the predicted inhaled nano-dose, and weighting them by exposure loss weights to construct a cost term of the objective function, used to quantify the negative impact of health risks. Similarly, the predicted first and second terms of the predicted sintering density are combined, and weighted by density benefit weights to construct a benefit term, used to quantify the positive value brought by product performance.

[0101] In embodiments of the present invention, the constraints on the objective function are established, including:

[0102] The objective function is constrained based on the upper limit of occupational exposure safety for predicted inhaled nano-dose and the minimum process requirement for predicted sintering density. The constraints are as follows:

[0103]

[0104] In the formula, This is the predicted inhaled nanodose under the tested ingredient ratio. It is the predicted sintering density under the tested ingredient ratio. It is the volume fraction of ultrafine powder under the tested ingredient ratio. This is the upper limit of occupational exposure safety. This is the minimum process requirement.

[0105] In detail, if the predicted inhaled nanoparticle dose exceeds the safety threshold, it will pose an occupational health risk. Therefore, it is necessary to follow the occupational exposure safety limits stipulated by industry / regulations. For predicting inhaled nanodose On the one hand, constraints need to be imposed; on the other hand, if the predicted sintering density is lower than the basic process requirements, the product will not meet the usage requirements, so it is necessary to base the minimum process requirements on the production standards. This constrains the prediction of sintering density.

[0106] The parameter optimization module is used to optimize the parameters of the objective function under constraints using the gradient descent algorithm to obtain the optimal parameters of the objective function.

[0107] In this step, the gradient of the objective function with respect to each optimization parameter is calculated, and the gradient is then applied using the differentiation rule. The partial derivative is calculated to obtain the gradient vector, which reflects the trend of the objective function value in the parameter space. Then, based on the gradient descent direction, the parameter values ​​are updated according to the preset learning rate, i.e., new parameter value = original parameter value - learning rate × gradient component. During the update process, it is checked in real time whether the predicted inhaled nano-dose and predicted sintering density corresponding to the new parameters meet the constraints. If not, the parameters are corrected through constraint projection or other methods to bring them back to the constrained feasible region. Then, the gradient calculation, parameter update and constraint verification steps are repeated and iterated until the iteration termination condition is met, such as the change in the objective function value being less than a set threshold or reaching the maximum number of iterations. Finally, when the iteration terminates, the current parameters are the optimal parameters of the objective function obtained by the gradient descent algorithm under the constraints.

[0108] The management plan development module is used to develop management plans for powder material preparation based on optimal parameters.

[0109] In embodiments of the present invention, a management scheme for powder material preparation is formulated based on optimal parameters, including:

[0110] The optimal personnel scheduling scheme and optimal air supply and exhaust volume are determined based on the optimal ultrafine powder volume fraction in the optimal parameters.

[0111] In this step, firstly, based on the optimal parameters of the objective function obtained through gradient descent algorithm optimization in the previous stage, the optimal ultrafine powder volume fraction is extracted. This parameter corresponds to the key ingredient ratio information in the preparation of metal composite ceramic powder. Secondly, based on the preset adaptation model of ultrafine powder volume fraction with personnel work intensity and working hours, combined with the daily production capacity demand of the production workshop, the required personnel workload at different times under the optimal ultrafine powder volume fraction is analyzed. Considering the occupational exposure risk of personnel, through the established personnel scheduling constraints, integer programming method is used to plan the shift allocation, working hours and job rotation arrangements of personnel at different skill levels, forming an initial personnel scheduling plan. At the same time, based on the correlation model between ultrafine powder volume fraction and dust diffusion and accumulation in the workshop, combined with the constraint requirements of predicting inhaled nano-dosage, fluid dynamics simulation is used. The initial values ​​of supply and exhaust air volume required to control the dust concentration in the workshop within the occupational exposure safety limit under the optimal ultrafine powder volume fraction were calculated. Then, the initial personnel shift plan and the initial values ​​of supply and exhaust air volume were substituted into the production simulation system to simulate the actual production scenario. The system was used to verify whether the predicted inhaled nano-dose met the constraints under the combined effect of personnel working hours, frequency, and supply and exhaust air volume. If not, the supply and exhaust air volume and personnel shift were adjusted, and the simulation was repeated. The system was iterated and adjusted until the simulation results met the constraints and personnel working efficiency and energy consumption were balanced. The personnel shift arrangement, working time period, and corresponding supply and exhaust air volume parameters obtained at this point are the optimal personnel shift plan and optimal supply and exhaust air volume for powder material preparation based on the optimal ultrafine powder volume fraction, realizing the synergistic optimization of occupational health and safety and production efficiency in the production process.

[0112] A management plan is developed based on the optimal ultrafine powder volume fraction, the optimal personnel scheduling plan, and the optimal air supply and exhaust volume. The management plan includes: batching powder materials according to the optimal ultrafine powder volume fraction, arranging workers to perform operations according to the optimal personnel scheduling plan, and supplying and exhausting air according to the optimal air supply and exhaust volume.

[0113] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A monitoring system for the preparation of metal composite ceramic powder materials, characterized in that, include: The potential source set generation module is used to generate a potential source set based on the worker's actual inhaled nano-dose, ultrafine powder volume fraction, and actual measured sintering density. The potential intensity generation module is used to perform power-law mapping on the inhaled nano-dose and the measured values ​​of sintering density to obtain the potential intensity of the exposed layer and the potential intensity of the dense layer. The potential field generation module is used to perform weighted aggregation calculations on the potential strength of the exposed layer and the potential strength of the compacted layer, respectively, to obtain the potential field of the exposed layer and the potential field of the compacted layer under the tested ingredient ratio. The distance between the current volume fraction of ultrafine powder and the volume fraction of ultrafine powder in the i-th batch is converted into a weighting factor using a preset kernel function. The preset kernel function is as follows: In the formula, It is a weighting factor. It is the attenuation coefficient. It is the absolute value of the difference between the volume fraction of ultrafine powder under the tested ingredient ratio and the volume fraction of ultrafine powder under the i-th batch. It is the volume fraction of ultrafine powder under the tested ingredient ratio. It is the volume fraction of ultrafine powder in the i-th batch; The potential field of the exposed layer is obtained by weighting and summing the weighting factor and the potential field of the exposed layer under the test ingredient ratio; The potential field of the compacted layer under the tested ingredient ratio is obtained by weighting and summing the weighting factor and the compacted layer potential. The prediction value generation module is used to generate the predicted inhaled nanodose and predicted sintering density at the ratio of the ingredients to be tested, based on the potential field of the exposed layer and the potential field of the density layer, respectively. The specific steps for generating the predicted inhaled nanodose and predicted sintering density at the tested ingredient ratio are as follows: Calculate the first-order local gradients of the potential fields of the exposed layer and the compacted layer at the ratio of the ingredients to be measured, respectively. The first-order local gradient is suppressed using a pre-defined set of gradient penalty factors to obtain the suppression result; The suppression result and the potential field of the exposed layer are coupled and mapped to obtain the predicted inhaled nanodose. The suppression result and the density layer potential field are coupled and mapped to obtain the predicted sintering density; The objective function construction module is used to construct an objective function based on the predicted inhaled nano-dose and the predicted sintering density, and to establish the constraints of the objective function. Specifically, constructing the objective function includes: The combined term predicting the inhaled nano-dose is used as the cost term of the objective function, and the combined term predicting the sintering density is used as the benefit term of the objective function. Construct an objective function based on cost and benefit terms; The objective function is as follows: In the formula, It is the objective function. It is the volume fraction of ultrafine powder under the tested ingredient ratio. This is the predicted inhaled nanodose under the tested ingredient ratio. It is the predicted sintering density under the tested ingredient ratio. It is the exposure loss weight. It is the density gain weight; The parameter optimization module is used to optimize the parameters of the objective function under constraints to obtain the optimal parameters of the objective function. The management plan development module is used to develop management plans for powder material preparation based on optimal parameters.

2. The monitoring system for the preparation of metal composite ceramic powder materials according to claim 1, characterized in that, A set of potential sources is generated based on the actual inhaled nano-dose, ultrafine powder volume fraction, and actual sintering density measured values ​​of the workers, including: For each batch i, the concentration of particles smaller than 100 nm in the powder material preparation workshop is multiplied by the worker's timely ventilation to obtain the worker's instantaneous inhalation volume. The instantaneous inhalation volume is then multiplied by the sealing factor of the protective equipment to obtain the actual inhaled nano-dose. For each batch i, the volume fraction of ultrafine powder and the actual measured value of sintering density are collected in real time by the workers. A potential source set is generated based on the actual inhaled nano-dose, the volume fraction of ultrafine powder, and the actual measured values ​​of sintering density.

3. The monitoring system for the preparation of metal composite ceramic powder materials according to claim 1, characterized in that, Power-law mappings were performed on the inhaled nano-dose and the measured values ​​of sintering density to obtain the potential strength of the exposed layer and the potential strength of the dense layer, including: The exposure layer potential of the inhaled nano-dose is calculated based on a preset power-law mapping relationship; The density layer potential strength is calculated based on the pre-set power-law mapping relationship to determine the measured sintering density.

4. The monitoring system for the preparation of metal composite ceramic powder materials according to claim 1, characterized in that, Establish the constraints for the objective function, including: Constraints for the objective function are established based on the upper limit of occupational exposure safety for predicted inhaled nano-dose and the minimum process requirements for predicted sintering density.

5. The monitoring system for the preparation of metal composite ceramic powder materials according to claim 1, characterized in that, Develop a management plan for powder material preparation based on optimal parameters, including: The optimal personnel scheduling scheme and optimal air supply and exhaust volume are determined based on the optimal ultrafine powder volume fraction in the optimal parameters. A management plan is developed based on the optimal ultrafine powder volume fraction, the optimal personnel scheduling plan, and the optimal air supply and exhaust volume. The management plan includes: batching powder materials according to the optimal ultrafine powder volume fraction, arranging workers to perform operations according to the optimal personnel scheduling plan, and supplying and exhausting air according to the optimal air supply and exhaust volume.

Images