Method, device, terminal equipment and storage medium for controlling a powder production process

Abstract

The application is suitable for the technical field of control, and provides a powder production process control method and device, terminal equipment and computer readable storage medium, which comprises the following steps: acquiring real-time particle size distribution in a powder production process; if a particle size deviation value between the real-time particle size distribution and a target particle size distribution is greater than a preset threshold value, calculating a real-time influence weight corresponding to a process parameter; wherein the real-time influence weight is used to represent a real-time influence degree of the process parameter on the particle size distribution; the process parameter is a parameter of a production equipment participating in powder production; generating a real-time control strategy according to the real-time influence weight corresponding to the process parameter; and controlling the production equipment to run according to the real-time control strategy. Through the above method, the control efficiency and control precision of the powder production process can be effectively improved.

Description

Technical Field

[0001] This application belongs to the field of control technology, and in particular relates to a control method, apparatus, terminal equipment and computer-readable storage medium for a powder production process. Background Technology

[0002] In modern industrial production, the particle size distribution of powder materials directly affects product quality and production efficiency, especially in food production, chemical industry, pharmaceuticals, and materials manufacturing, where particle size control is considered a crucial step. During production, particle size distribution is influenced by a combination of process parameters, such as grinding intensity and raw material input speed.

[0003] Current control methods typically employ fixed parameter tuning, meaning that the effect of each process parameter on particle size distribution is assumed to be constant. This approach fails to accurately identify which parameters are most critical under the current conditions, leading to a blind control strategy. This blindness may cause the system to optimize in the wrong direction, wasting time and resources, resulting in low control efficiency and accuracy. Summary of the Invention

[0004] This application provides a method, apparatus, terminal equipment, and computer-readable storage medium for controlling a powder production process, which can effectively improve the control efficiency and accuracy of the powder production process.

[0005] In a first aspect, embodiments of this application provide a method for controlling a powder production process, including:

[0006] To obtain the real-time particle size distribution during the powder production process;

[0007] If the particle size deviation between the real-time particle size distribution and the target particle size distribution is greater than a preset threshold, then the real-time influence weight corresponding to the process parameter is calculated; wherein, the real-time influence weight is used to characterize the degree of real-time influence of the process parameter on the particle size distribution; the process parameter is the parameter of the production equipment involved in powder production;

[0008] A real-time control strategy is generated based on the real-time impact weights corresponding to the process parameters.

[0009] The production equipment is controlled to operate according to the real-time control strategy.

[0010] In this embodiment, monitoring the particle size deviation between the real-time particle size distribution and the target particle size distribution during the powder production process is equivalent to monitoring the powder production process through particle size distribution. When the particle size deviation is large, a control strategy is generated based on the degree of influence of process parameters on particle size distribution. In this way, it is possible to identify which process parameters are most critical in the current state during the control process, making the control strategy more accurate and targeted, thereby helping to improve control efficiency and control accuracy.

[0011] In one possible implementation of the first aspect, the real-time influence weights corresponding to the calculated process parameters include:

[0012] Obtain a parameter set, which includes the process parameters of all production equipment involved in powder production;

[0013] The process parameters included in the parameter set are arranged and combined to obtain multiple parameter groups; wherein each parameter group includes at least two process parameters, and the combination of process parameters included in different parameter groups is different;

[0014] Calculate the interaction effect of each parameter group on the granularity distribution to obtain the interaction weight corresponding to each parameter group;

[0015] Based on the interaction weights corresponding to each parameter group, calculate the real-time impact weight of each process parameter in the parameter set.

[0016] In this embodiment, the individual and combined effects of multiple process parameters on particle size distribution are considered, as well as different combinations of process parameters. This allows the final determined real-time influence weights to more accurately reflect the degree of influence of process parameters on particle size distribution, providing reliable data for subsequent control.

[0017] Optionally, obtaining the parameter set may include: obtaining the process parameters of all production equipment involved in powder production; preprocessing the obtained process parameters to obtain a processed parameter set. The preprocessing process may include outlier handling, standardization, and data alignment.

[0018] In one possible implementation of the first aspect, the step of arranging and combining the process parameters included in the parameter set to obtain multiple parameter groups includes:

[0019] The individual influence of each process parameter in the parameter set on particle size distribution is calculated to obtain the single-factor weight corresponding to each process parameter;

[0020] Filter out process parameters in the parameter set whose single-factor weight is less than a first preset weight value to obtain the filtered parameter set;

[0021] The process parameters included in the filtered parameter set are arranged and combined to obtain multiple parameter groups.

[0022] Compared to directly performing permutations and combinations, this implementation first filters multiple process parameters based on the degree of influence of each parameter on the granularity distribution. Since some process parameters have little or no impact on the granularity distribution, filtering out these parameters not only reduces the disturbance caused by parameter tuning but also reduces the computational load of interaction weights. Furthermore, filtering based on the individual influence of each process parameter can eliminate the influence of other process parameters, thereby reducing the possibility of accidental deletion.

[0023] In one possible implementation of the first aspect, calculating the individual influence of each process parameter in the parameter set on the particle size distribution to obtain the single-factor weight corresponding to each process parameter includes:

[0024] Obtain the historical granularity distribution and the historical parameter values ​​of the process parameters within a preset time window;

[0025] Calculate the first feature data of the historical granularity distribution;

[0026] Calculate the second feature data of the historical parameter values;

[0027] The correlation between the first feature data and the second feature data is calculated to obtain the single-factor weights corresponding to the process parameters.

[0028] In this embodiment, by calculating the correlation between the characteristic data of historical granularity distribution and historical parameter values, the inherent relationship between the changing trends of process parameters and granularity distribution can be uncovered, thereby reflecting the degree of influence of process parameters on granularity distribution and providing reliable data for subsequent control. Furthermore, since characteristic data can more clearly and specifically reflect data characteristics, compared with directly using historical data for calculation, calculation using characteristic data can more accurately reflect the influence of process parameters on granularity distribution, while also reducing the amount of data processing.

[0029] In one possible implementation of the first aspect, calculating the interaction effect of each parameter group on the granularity distribution to obtain the interaction weight corresponding to each parameter group includes:

[0030] The process parameters in the parameter group are divided into a first discrete variable and a first continuous variable;

[0031] Discretize the first continuous variable into a second discrete variable;

[0032] The interaction effect of the parameter set on the granularity distribution is calculated based on the first discrete variable and the second discrete variable to obtain the first weight;

[0033] Fit the first discrete variable to a second continuous variable;

[0034] The interaction effect of the parameter set on the granularity distribution is calculated based on the first continuous variable and the second continuous variable to obtain the second weight;

[0035] The interaction weights corresponding to the parameter group are calculated based on the first weight and the second weight.

[0036] In this embodiment, the method of determining the interaction weights corresponding to the parameter group based on the first weight and the second weight involves two calculations, which can reduce weighting errors caused by inaccuracies during the process of changing from continuous to discrete or from discrete to continuous. Furthermore, the above-mentioned method of calculating interaction weights considers the interactive effects between variables of different forms, and can more accurately reflect the influence of process parameters on particle size distribution.

[0037] Optionally, the weighted value of the interaction weight corresponding to each process parameter can be calculated as the real-time influence weight corresponding to each process parameter.

[0038] Optionally, the maximum interaction weight corresponding to each process parameter can be used as the real-time influence weight corresponding to each process parameter.

[0039] Optionally, the largest interaction weight can be used as the real-time influence weight of each process parameter in the corresponding parameter group, and then the real-time influence weight of the undetermined process parameter can be determined based on the second largest interaction weight, and so on (equivalent to determining the real-time influence weight of each process parameter on a parameter group basis).

[0040] Optionally, the real-time influence weight of each process parameter can be determined by combining the single-factor weight of each process parameter with its corresponding interaction weight.

[0041] In one possible implementation of the first aspect, generating a real-time control strategy based on the real-time influence weights corresponding to the process parameters includes:

[0042] Obtain the target value of the unit control quantity corresponding to the particle size deviation value;

[0043] The priority of the process parameters is determined based on the real-time impact weights corresponding to the process parameters;

[0044] Obtain the priority multiple corresponding to the process parameters;

[0045] The real-time adjustment amount of the process parameter is calculated based on the target value of the unit control quantity, the multiple corresponding to the priority of the process parameter, the real-time influence weight corresponding to the process parameter, and the granularity deviation value; wherein, the real-time control strategy includes the real-time adjustment amount of the process parameter.

[0046] In this embodiment, the real-time adjustment amount is calculated based on the priority multiple of the process parameters and the real-time influence weight of the process parameters. This is equivalent to taking into account the real-time influence of the process parameters on the particle size distribution. The real-time adjustment amount calculated in this way is more targeted, making the control more precise.

[0047] Optionally, when the particle size deviation value is greater than the first value, all process parameters are adjusted. When the particle size deviation value is less than the second value, process parameters with a real-time impact weight less than the preset value are filtered out, and the remaining process parameters are adjusted.

[0048] In one example, the real-time adjustment can be calculated using the formula △K×H×(particle size deviation value / real-time impact weight). Here, △K is the unit control quantity, and H is the multiple corresponding to the priority of the process parameter.

[0049] In one possible implementation of the first aspect, obtaining the target value of the unit control quantity corresponding to the granularity deviation value includes:

[0050] Constraints are generated based on the current parameter value of the first parameter; wherein, the first parameter is the parameter in the parameter set excluding the second parameter, and the second parameter is the process parameter whose real-time influence weight is greater than the second preset weight.

[0051] A target optimization function is constructed based on the constraints and control objectives; wherein the decision variable of the target optimization function is the unit control variable;

[0052] The target value of the unit control quantity is obtained by performing optimization calculations based on the target optimization function.

[0053] In this embodiment of the application, by optimizing the calculation, the unit adjustment amount can be obtained more accurately, which is conducive to achieving more precise control.

[0054] Secondly, embodiments of this application provide a control device for a powder production process, comprising:

[0055] The acquisition unit is used to acquire the real-time particle size distribution during the powder production process;

[0056] The calculation unit is used to calculate the real-time influence weight of the process parameters if the particle size deviation between the real-time particle size distribution and the target particle size distribution is greater than a preset threshold; wherein, the real-time influence weight is used to characterize the degree of real-time influence of the process parameters on the particle size distribution; the process parameters are the parameters of the production equipment involved in powder production;

[0057] The generation unit is used to generate a real-time control strategy based on the real-time influence weights corresponding to the process parameters.

[0058] The control unit is used to control the operation of the production equipment according to the real-time control strategy.

[0059] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the control method for the powder production process as described in any one of the first aspects above.

[0060] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the control method for the powder production process as described in any one of the first aspects above.

[0061] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the control method for the powder production process described in any of the first aspects.

[0062] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a schematic flowchart of a powder production process control method provided in an embodiment of this application;

[0065] Figure 2 This is a schematic diagram of the structure of a control device for a powder production process according to an embodiment of this application;

[0066] Figure 3This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0067] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0068] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0069] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0070] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0071] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0072] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0073] In modern industrial production, the particle size distribution of powder materials directly affects product quality and production efficiency, especially in food production, chemical industry, pharmaceuticals, and materials manufacturing, where particle size control is considered a crucial step. During production, particle size distribution is influenced by a combination of process parameters, such as grinding intensity and raw material input speed.

[0074] Current control methods typically employ fixed parameter tuning, meaning that the effect of each process parameter on particle size distribution is assumed to be constant. This approach fails to accurately identify which parameters are most critical under the current conditions, leading to a blind control strategy. This blindness may cause the system to optimize in the wrong direction, wasting time and resources, resulting in low control efficiency and accuracy.

[0075] Based on this, embodiments of this application provide a control method for a powder production process. In this embodiment, monitoring the particle size deviation between the real-time particle size distribution and the target particle size distribution during the powder production process is equivalent to monitoring the powder production process through particle size distribution. When the particle size deviation is large, a control strategy is generated based on the degree of influence of process parameters on the particle size distribution. This method identifies which process parameters are most critical in the current state during the control process, making the control strategy more precise and targeted, thereby helping to improve control efficiency and accuracy.

[0076] See Figure 1 This is a schematic flowchart of a powder production process control method provided in an embodiment of this application. It is intended as an example and not a limitation. The method may include the following steps:

[0077] S101, obtain the real-time particle size distribution during the powder production process.

[0078] Particle size distribution is a key indicator describing the proportion of particles of different sizes in a powder or particle group, and it directly affects the quality of the finished powder product.

[0079] Optionally, particle size distribution can be characterized by characteristic particle sizes, which reflect the overall size of the particles. For example, D10 represents the particle size distribution at the 10% cumulative distribution, indicating that 10% of the particles are smaller than this value, reflecting the proportion of fine particles. D50 represents the particle size distribution at the 50% cumulative distribution, indicating that 50% of the particles are smaller than this value, typically representing the average size of the powder. D90 represents the particle size distribution at the 90% cumulative distribution, indicating that 90% of the particles are smaller than this value, reflecting the proportion of coarse particles.

[0080] Optionally, particle size distribution can also be characterized by distribution width, which reflects the degree of particle dispersion. For example, one parameter of distribution width is span (D90-D10) / D50; the smaller the span value, the more concentrated the distribution. Another parameter of distribution width is the coefficient of variation (standard deviation / average particle size) × 100%; the smaller the coefficient of variation, the more uniform the particle size.

[0081] In some implementations, real-time particle size distribution can be obtained using laser diffraction. The principle is based on the phenomenon of light scattering by particles. For example, the instrument collects the intensity signals of scattered light using laser detectors distributed at different angles, and calculates the proportion of particles of different sizes using optical models (such as Mie theory), ultimately generating a particle size distribution curve. Laser detection offers advantages such as high speed, wide range, and good repeatability.

[0082] In other implementations, real-time particle size distribution can be obtained through image analysis. For example, microscopic images of particles are acquired using an imaging device, and image processing algorithms are used to identify the size of individual particles and statistically determine the particle size distribution. Unlike laser methods, image analysis can observe information such as particle size and morphology, making it suitable for irregular particles and detection scenarios requiring morphological characterization.

[0083] The real-time granularity distribution in this embodiment can be a granularity distribution obtained according to a preset period. The preset period can be set according to control requirements. It is understood that a smaller preset period results in a higher frequency of obtaining the real-time granularity distribution and higher control accuracy, but also a greater computational load; conversely, a larger preset period results in a lower frequency of obtaining the real-time granularity distribution and lower control accuracy, but a smaller computational load.

[0084] S102, if the particle size deviation between the real-time particle size distribution and the target particle size distribution is greater than a preset threshold, then calculate the real-time influence weight corresponding to the process parameters.

[0085] The target granularity distribution can be preset according to production needs. It is understood that the real-time granularity distribution uses the same granularity distribution index as the target granularity distribution. For example, if the target granularity distribution is represented by a characteristic particle size, the real-time granularity distribution will also be represented by the same characteristic particle size. If the target granularity distribution is represented by a distribution width, the real-time granularity distribution will also be represented by a distribution width.

[0086] Of course, the target granularity distribution can be represented by various indicators. For example, the target granularity distribution can be represented by the D50 characteristic particle size and the coefficient of variation, and the real-time granularity distribution can also be represented by the D50 characteristic particle size and the coefficient of variation. Accordingly, calculating the granularity deviation value includes: calculating the first deviation value between the D50 characteristic particle size of the real-time granularity distribution and the D50 characteristic particle size of the target granularity distribution; calculating the second deviation value between the coefficient of variation of the real-time granularity distribution and the coefficient of variation of the target granularity distribution; and calculating the granularity deviation value based on the first and second deviation values.

[0087] Different weights can be assigned to different indicators, representing their importance. Then, the deviation values ​​corresponding to each indicator are weighted and summed to obtain the granularity deviation value. For example, continuing the previous example, the weight of the first deviation value is 0.4, and the weight of the second deviation value is 0.6. The granularity deviation value is calculated as 0.4 × first deviation value + 0.6 × second deviation value.

[0088] In this embodiment, process parameters refer to the parameters of the production equipment involved in powder production, such as feed rate, pulverizer speed, grinding pressure, spray drying conditions, etc. The real-time influence weight is used to characterize the degree of real-time influence of process parameters on particle size distribution.

[0089] Optionally, the real-time influence weight of only one process parameter can be calculated, which is equivalent to controlling the particle size distribution based solely on the real-time impact of that single process parameter. Alternatively, the real-time influence weight of multiple process parameters can be calculated, which is equivalent to controlling the particle size distribution based on the individual real-time impact of each process parameter. Compared to the two methods, controlling using multiple process parameters results in more precise control.

[0090] In one embodiment, S102 may include:

[0091] S201, Obtain the parameter set, which includes the process parameters of all production equipment involved in powder production.

[0092] Each piece of production equipment can be equipped with sensors to monitor its process parameters. For example, a feeder can be equipped with a speed sensor to monitor the feed rate; a pulverizer can be equipped with a speed sensor to monitor the pulverizer speed; a grinder can be equipped with a pressure sensor to monitor the grinding pressure; and a sprayer can be equipped with a temperature / humidity sensor to monitor the temperature / humidity.

[0093] Optionally, obtaining the parameter set may include: obtaining the process parameters of each of the production equipment involved in powder production; preprocessing the obtained process parameters to obtain a processed parameter set. The preprocessing process may include:

[0094] Outlier handling involves obtaining the standard value range for each process parameter. If a process parameter's value exceeds its corresponding standard value range, the parameter is either deleted or replaced with a value within the standard range (such as the median). This method reduces the impact of outliers on control accuracy.

[0095] Standardization involves normalizing each process parameter to obtain normalized process parameters. Normalization eliminates the influence of different process parameters on their magnitude, making the data comparable on the same scale.

[0096] Data alignment involves using linear interpolation to fill in the sampled data for each process parameter, then obtaining a set of parameters (including multiple process parameters) with timestamp alignment from the filled data as a parameter set. For example, sensor A has a sampling frequency of 10Hz, and sensor B has a sampling frequency of 1Hz. Linear interpolation is used to fill the sampling data from sensor B with the sampling time points of sensor A (e.g., if sensor B's sampling value at 10:00:00 is 5, and its sampling value at 10:00:01 is 7, then 10:00:00.5 is interpolated to 6). Then, the sampling values ​​from sensor A and sensor B corresponding to the same sampling point are grouped into the same parameter set. Since the sampling frequencies of sensors in different production equipment are different, this data alignment method can obtain time-aligned process parameters, providing reliable data for subsequent control.

[0097] S202, based on the process parameters included in the parameter set, arrange and combine them to obtain multiple parameter groups.

[0098] Each parameter group includes at least two process parameters, and the combination of process parameters in different parameter groups is different.

[0099] In one implementation, the process parameters included in the parameter set are directly arranged and combined. For example, the parameter set includes three process parameters: feed rate, mill speed, and grinding pressure. Arranging and combining these parameters yields four parameter groups: the first group includes feed rate and mill speed; the second group includes feed rate and grinding pressure; the third group includes mill speed and grinding pressure; and the fourth group includes feed rate, mill speed, and grinding pressure.

[0100] In another implementation, the individual influence of each process parameter in the parameter set on the particle size distribution is calculated to obtain the single-factor weight corresponding to each process parameter; process parameters in the parameter set whose single-factor weight is less than the first preset weight are filtered to obtain the filtered parameter set; the process parameters included in the filtered parameter set are arranged and combined to obtain multiple parameter groups.

[0101] For example, the parameter set includes four process parameters: feed rate, mill speed, grinding pressure, and spray drying conditions. If the single-factor weight of the spray drying conditions is less than the first preset weight corresponding to that process parameter, then the spray drying conditions are filtered out. The filtered parameter set includes three process parameters: feed rate, mill speed, and grinding pressure. These three process parameters are then arranged and combined to obtain multiple parameter groups.

[0102] Compared to directly performing permutations and combinations, this implementation first filters multiple process parameters based on the degree of influence of each parameter on the granularity distribution. Since some process parameters have little or no impact on the granularity distribution, filtering out these parameters not only reduces the disturbance caused by parameter tuning but also reduces the computational load of interaction weights. Furthermore, filtering based on the individual influence of each process parameter can eliminate the influence of other process parameters, thereby reducing the possibility of accidental deletion.

[0103] Optionally, the calculation method for single-factor weights may include:

[0104] Obtain the historical granularity distribution and historical parameter values ​​of process parameters within a preset time window;

[0105] Calculate the first characteristic data of the historical granularity distribution;

[0106] Calculate the second feature data of historical parameter values;

[0107] Calculate the correlation between the first feature data and the second feature data to obtain the single-factor weights corresponding to the process parameters.

[0108] The first feature data can be D10 / D50 / D90, standard deviation of the distribution, distribution span, or coefficient of variation, etc. Only one feature can be counted, or multiple features can be counted. When counting multiple features, the multiple features can be concatenated to obtain the first feature data, or a feature group including multiple features can be constructed (the data in this feature group is denoted as the first feature data), or the multiple features can be weighted and summed to obtain the first feature data.

[0109] The second characteristic data can be statistical data of historical parameter values, such as average, variance, or rate of change. Only one statistical data point can be calculated, or multiple statistical data points can be calculated. When calculating multiple statistical data points, a data set can be constructed, and the data in this set are denoted as the second characteristic data.

[0110] In one example of correlation calculation, the correlation between the first and second feature data can be calculated using methods such as Pearson correlation coefficient, Spearman rank correlation coefficient, or Kendall rank correlation coefficient. Alternatively, a trained neural network model used to calculate the correlation of the input data can be used to calculate the correlation between the first and second feature data.

[0111] In another example of correlation calculation, when both the first feature data and the second feature data include multiple features, optionally, the correlation between each first feature data and each second feature data can be calculated separately, and all correlations can be weighted to obtain a single-factor weight. Alternatively, the interaction effects of multiple second feature data on each first feature data can be calculated, and the interaction effects corresponding to each first feature data can be weighted to obtain a single-factor weight.

[0112] In this embodiment, by calculating the correlation between the characteristic data of historical granularity distribution and historical parameter values, the inherent relationship between the changing trends of process parameters and granularity distribution can be uncovered, thereby reflecting the degree of influence of process parameters on granularity distribution and providing reliable data for subsequent control. Furthermore, since characteristic data can more clearly and specifically reflect data characteristics, compared with directly using historical data for calculation, calculation using characteristic data can more accurately reflect the influence of process parameters on granularity distribution, while also reducing the amount of data processing.

[0113] S203, calculate the interaction effect of each parameter group on the granularity distribution, and obtain the interaction weight corresponding to each parameter group.

[0114] It is understandable that single-factor weights reflect the degree of influence of a certain process parameter on particle size distribution alone, while interaction weights reflect the influence of multiple process parameters on particle size distribution together.

[0115] In one embodiment, S203 may include: dividing the process parameters in the parameter group into a first discrete variable and a first continuous variable; discretizing the first continuous variable into a second discrete variable; calculating the interaction effect of the parameter group on the particle size distribution based on the first discrete variable and the second discrete variable to obtain a first weight; fitting the first discrete variable into a second continuous variable; calculating the interaction effect of the parameter group on the particle size distribution based on the first continuous variable and the second continuous variable to obtain a second weight; and calculating the interaction weight corresponding to the parameter group based on the first weight and the second weight.

[0116] Optionally, discretization can be performed by taking the numerical values ​​from the continuous variable corresponding to the time points of the discrete variable. Alternatively, discretization can also be performed by taking the statistical values ​​of data within a preset window corresponding to the time points of the discrete variable from the continuous variable. For example, if the time point of the discrete variable is 10:00:00, 10 random data points from the continuous variable within the time window of 09:59:00-10:01:00 can be taken, and the statistical values ​​(such as mean, variance, maximum, or minimum values) of these 10 data points can be calculated.

[0117] Optionally, the fitting method can be to fit discrete variables into continuous variables through interpolation, linear fitting, or least squares methods.

[0118] One implementation of the first weight is to calculate the correlation between every two data points corresponding to timestamps in the first and second discrete variables, and then weight all the calculated correlations to obtain the first weight.

[0119] Optionally, the second weight can be calculated using methods such as Pearson correlation coefficient, Spearman rank correlation coefficient, or Kendall rank correlation coefficient.

[0120] Optionally, the interaction weights corresponding to the parameter group can be obtained by weighted summation of the first and second weights.

[0121] It is understood that in other embodiments, the interaction weights corresponding to the parameter group can be determined solely based on the first weight, or solely based on the second weight. Compared to this approach, the method described above, which determines the interaction weights corresponding to the parameter group based on both the first and second weights, reduces weighting errors caused by inaccuracies during the continuous-to-discrete or discrete-to-continuous transition process through two calculations. Furthermore, the aforementioned method of calculating interaction weights considers the interactive effects between variables of different forms, thus more accurately reflecting the influence of process parameters on particle size distribution.

[0122] S204, calculate the real-time impact weight of each process parameter in the parameter set based on the interaction weight corresponding to each parameter group.

[0123] Optionally, the weighted value of the interaction weight corresponding to each process parameter can be calculated as the real-time influence weight corresponding to each process parameter.

[0124] Optionally, the maximum interaction weight corresponding to each process parameter can be used as the real-time influence weight corresponding to each process parameter.

[0125] Optionally, the largest interaction weight can be used as the real-time influence weight of each process parameter in the corresponding parameter group, and then the real-time influence weight of the undetermined process parameter can be determined based on the second largest interaction weight, and so on (equivalent to determining the real-time influence weight of each process parameter on a parameter group basis).

[0126] Optionally, the real-time influence weight of each process parameter can be determined by combining the single-factor weight of each process parameter with its corresponding interaction weight.

[0127] In the embodiments described in steps S201-S204, the individual and combined effects of multiple process parameters on particle size distribution are considered, as well as different combinations of process parameters. This allows the final determined real-time influence weights to more accurately reflect the degree of influence of process parameters on particle size distribution, providing reliable data for subsequent control.

[0128] S103 generates a real-time control strategy based on the real-time influence weights corresponding to the process parameters.

[0129] In one embodiment, S103 may include:

[0130] Obtain the target value of the unit control quantity corresponding to the particle size deviation value;

[0131] The priority of process parameters is determined based on their real-time impact weights.

[0132] Obtain the priority multiple corresponding to the process parameters;

[0133] The real-time adjustment of the process parameters is calculated based on the target value of the unit control quantity, the priority multiple of the process parameters, the real-time influence weight of the process parameters, and the granularity deviation value; wherein, the real-time control strategy includes the real-time adjustment of the process parameters.

[0134] Optionally, the process parameters can be sorted in descending order of their real-time impact weight, with the process parameter having the largest real-time impact weight having the highest priority and the process parameter having the smallest real-time impact weight having the lowest priority.

[0135] Optionally, process parameters with a real-time impact weight less than a preset value can be filtered out. This is equivalent to adjusting only process parameters with a larger real-time impact weight.

[0136] Understandably, the higher the priority of a process parameter, the larger its corresponding multiplier. Conversely, the lower the priority of a process parameter, the smaller its corresponding multiplier.

[0137] Optionally, when the particle size deviation value is greater than the first value, all process parameters are adjusted. When the particle size deviation value is less than the second value, process parameters with a real-time impact weight less than the preset value are filtered out, and the remaining process parameters are adjusted.

[0138] In one example, the real-time adjustment can be calculated using the formula △K×H×(particle size deviation value / real-time impact weight). Here, △K is the unit control quantity, and H is the multiple corresponding to the priority of the process parameter.

[0139] The unit control quantity is the minimum adjustment amount for control. For example, when the granularity deviation is large, the corresponding unit control quantity is also large, which means that the adjustment amount for each control is large, in order to quickly narrow the gap between the real-time granularity distribution and the target granularity distribution, thus improving control efficiency. Conversely, when the granularity deviation is small, the corresponding unit control quantity is small, which means that the adjustment amount for each control is small, in order to accurately narrow the gap between the real-time granularity distribution and the target granularity distribution, thus improving control accuracy.

[0140] In this embodiment, the real-time adjustment amount is calculated based on the priority multiple of the process parameters and the real-time influence weight of the process parameters. This is equivalent to taking into account the real-time influence of the process parameters on the particle size distribution. The real-time adjustment amount calculated in this way is more targeted, making the control more precise.

[0141] In one embodiment, target values ​​for unit control quantities corresponding to different granularity deviation values ​​can be preset.

[0142] In another embodiment, the method for obtaining the target value of the unit control quantity may include:

[0143] Constraints are generated based on the current parameter value of the first parameter; wherein, the first parameter is the parameter in the parameter set excluding the second parameter, and the second parameter is the process parameter whose real-time influence weight is greater than the second preset weight.

[0144] A target optimization function is constructed based on the constraints and control objectives; wherein the decision variable of the target optimization function is the unit control variable;

[0145] The target value of the unit control quantity is obtained by performing optimization calculations based on the target optimization function.

[0146] The control objectives may include granularity distribution objectives, maximum efficiency, minimum cost, or maximum output, etc.

[0147] For example, assume the control target is a particle size D50 = 10 μm. The parameter set includes stirring speed num1, feed rate num2, and reaction temperature num3. Among them, the real-time influence weight of stirring speed num1 on D50 is greater than the second preset weight value, so stirring speed num1 is the second parameter, and feed rate num2 and reaction temperature num3 are the first parameters.

[0148] Assuming the current feed rate parameter is 50 kg / h, and its maximum value is 60 kg / h, then the constraint condition generated based on the feed rate is num2 ≤ 60 kg / h. The current reaction temperature parameter num3 is 30℃, and its temperature range is 10-60℃. The constraint condition generated based on the reaction temperature is 10 ≤ num3 ≤ 60℃.

[0149] The objective function generated based on the above control objectives and constraints is:

[0150] minf(x) = w1 × |D50 实际 -10∣+w2×energy consumption increment(x);

[0151] st num2≤60kg / h; and, 10≤num3≤60;

[0152] Where x is the unit control quantity, w1 and w2 are weights, and energy consumption increment (x) is the energy consumption increment corresponding to the unit control quantity x.

[0153] It should be noted that the above are merely examples of objective optimization functions. In practical applications, objective functions for process parameters can be constructed according to production needs. This application does not limit the specific form of the objective optimization function.

[0154] In this embodiment of the application, by optimizing the calculation, the unit adjustment amount can be obtained more accurately, which is conducive to achieving more precise control.

[0155] S104 controls the operation of production equipment according to the real-time control strategy.

[0156] As described in S103, the process parameters are adjusted according to the real-time adjustment amount of each process parameter, and the operation of the production equipment is controlled according to the adjusted process parameters. For example, if the feed rate before adjustment is 1 kg / h and the feed rate after adjustment is 1.1 kg / h, then the feeder is controlled to feed at a rate of 1.1 kg / h.

[0157] In this embodiment, monitoring the particle size deviation between the real-time particle size distribution and the target particle size distribution during the powder production process is equivalent to monitoring the powder production process through particle size distribution. When the particle size deviation is large, a control strategy is generated based on the degree of influence of process parameters on particle size distribution. In this way, it is possible to identify which process parameters are most critical in the current state during the control process, making the control strategy more accurate and targeted, thereby helping to improve control efficiency and control accuracy.

[0158] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0159] Corresponding to the control method of the powder production process described in the above embodiments, Figure 2 This is a structural block diagram of the control device for the powder production process provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0160] Reference Figure 2 The device 2 includes:

[0161] Acquisition unit 21 is used to acquire the real-time particle size distribution during the powder production process.

[0162] The calculation unit 22 is used to calculate the real-time influence weight of the process parameters if the particle size deviation between the real-time particle size distribution and the target particle size distribution is greater than a preset threshold; wherein, the real-time influence weight is used to characterize the degree of real-time influence of the process parameters on the particle size distribution; the process parameters are the parameters of the production equipment involved in powder production.

[0163] The generation unit 23 is used to generate a real-time control strategy based on the real-time influence weights corresponding to the process parameters.

[0164] Control unit 24 is used to control the operation of the production equipment according to the real-time control strategy.

[0165] Optionally, the computing unit 22 is also used for:

[0166] Obtain a parameter set, which includes the process parameters of all production equipment involved in powder production;

[0167] The process parameters included in the parameter set are arranged and combined to obtain multiple parameter groups; wherein each parameter group includes at least two process parameters, and the combination of process parameters included in different parameter groups is different;

[0168] Calculate the interaction effect of each parameter group on the granularity distribution to obtain the interaction weight corresponding to each parameter group;

[0169] Based on the interaction weights corresponding to each parameter group, calculate the real-time impact weight of each process parameter in the parameter set.

[0170] Optionally, the computing unit 22 is also used for:

[0171] The individual influence of each process parameter in the parameter set on particle size distribution is calculated to obtain the single-factor weight corresponding to each process parameter;

[0172] Filter out process parameters in the parameter set whose single-factor weight is less than a first preset weight value to obtain the filtered parameter set;

[0173] The process parameters included in the filtered parameter set are arranged and combined to obtain multiple parameter groups.

[0174] Optionally, the computing unit 22 is also used for:

[0175] Obtain the historical granularity distribution and the historical parameter values ​​of the process parameters within a preset time window;

[0176] Calculate the first feature data of the historical granularity distribution;

[0177] Calculate the second feature data of the historical parameter values;

[0178] The correlation between the first feature data and the second feature data is calculated to obtain the single-factor weights corresponding to the process parameters.

[0179] Optionally, the computing unit 22 is also used for:

[0180] The process parameters in the parameter group are divided into a first discrete variable and a first continuous variable;

[0181] Discretize the first continuous variable into a second discrete variable;

[0182] The interaction effect of the parameter set on the granularity distribution is calculated based on the first discrete variable and the second discrete variable to obtain the first weight;

[0183] Fit the first discrete variable to a second continuous variable;

[0184] The interaction effect of the parameter set on the granularity distribution is calculated based on the first continuous variable and the second continuous variable to obtain the second weight;

[0185] The interaction weights corresponding to the parameter group are calculated based on the first weight and the second weight.

[0186] Optionally, the generating unit 23 is also used for:

[0187] Obtain the target value of the unit control quantity corresponding to the particle size deviation value;

[0188] The priority of the process parameters is determined based on the real-time impact weights corresponding to the process parameters;

[0189] Obtain the priority multiple corresponding to the process parameters;

[0190] The real-time adjustment amount of the process parameter is calculated based on the target value of the unit control quantity, the multiple corresponding to the priority of the process parameter, the real-time influence weight corresponding to the process parameter, and the granularity deviation value; wherein, the real-time control strategy includes the real-time adjustment amount of the process parameter.

[0191] Optionally, the generating unit 23 is also used for:

[0192] Constraints are generated based on the current parameter value of the first parameter; wherein, the first parameter is the parameter in the parameter set excluding the second parameter, and the second parameter is the process parameter whose real-time influence weight is greater than the second preset weight.

[0193] A target optimization function is constructed based on the constraints and control objectives; wherein the decision variable of the target optimization function is the unit control variable;

[0194] The target value of the unit control quantity is obtained by performing optimization calculations based on the target optimization function.

[0195] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0196] in addition, Figure 2 The control device for the powder production process shown can be a software unit, a hardware unit, or a combination of software and hardware built into existing terminal equipment. It can also be integrated into the terminal equipment as an independent component, or exist as an independent terminal equipment.

[0197] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0198] Figure 3 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. For example... Figure 3 As shown, the terminal device 3 in this embodiment includes: at least one processor 30 ( Figure 3 (Only one is shown in the image) a processor, a memory 31, and a computer program 32 stored in the memory 31 and executable on the at least one processor 30, wherein the processor 30 executes the computer program 32 to implement the steps in any of the above-described control method embodiments for the powder production process.

[0199] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. This terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 3 This is merely an example of terminal device 3 and does not constitute a limitation on terminal device 3. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0200] The processor 30 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0201] In some embodiments, the memory 31 may be an internal storage unit of the terminal device 3, such as a hard disk or memory of the terminal device 3. In other embodiments, the memory 31 may be an external storage device of the terminal device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal device 3. Furthermore, the memory 31 may include both internal and external storage units of the terminal device 3. The memory 31 is used to store the operating system, applications, boot loader, data, and other programs, such as the program code of the computer program. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0202] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps in the above-described method embodiments.

[0203] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments.

[0204] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / terminal equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0205] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0206] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0207] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0208] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0209] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for controlling a powder production process, characterized in that, include: To obtain the real-time particle size distribution during the powder production process; If the particle size deviation between the real-time particle size distribution and the target particle size distribution is greater than a preset threshold, then the real-time influence weight corresponding to the process parameter is calculated; wherein, the real-time influence weight is used to characterize the degree of real-time influence of the process parameter on the particle size distribution; the process parameter is the parameter of the production equipment involved in powder production; A real-time control strategy is generated based on the real-time impact weights corresponding to the process parameters. The production equipment is controlled to operate according to the real-time control strategy. The real-time impact weights corresponding to the calculated process parameters include: Obtain a parameter set, which includes the process parameters of all production equipment involved in powder production; The process parameters included in the parameter set are arranged and combined to obtain multiple parameter groups; wherein each parameter group includes at least two process parameters, and the combination of process parameters included in different parameter groups is different; Calculate the interaction effect of each parameter group on the granularity distribution to obtain the interaction weight corresponding to each parameter group; Based on the interaction weights corresponding to each parameter group, calculate the real-time impact weight of each process parameter in the parameter set. The step of calculating the interaction influence of each parameter group on the granularity distribution to obtain the interaction weight corresponding to each parameter group includes: The process parameters in the parameter group are divided into a first discrete variable and a first continuous variable; Discretize the first continuous variable into a second discrete variable; The interaction effect of the parameter set on the granularity distribution is calculated based on the first discrete variable and the second discrete variable to obtain the first weight; Fit the first discrete variable to a second continuous variable; The interaction effect of the parameter set on the granularity distribution is calculated based on the first continuous variable and the second continuous variable to obtain the second weight; The interaction weights corresponding to the parameter group are calculated based on the first weight and the second weight.

2. The method for controlling the powder production process as described in claim 1, characterized in that, The process parameters included in the parameter set are arranged and combined to obtain multiple parameter groups, including: The individual influence of each process parameter in the parameter set on particle size distribution is calculated to obtain the single-factor weight corresponding to each process parameter; Filter out process parameters in the parameter set whose single-factor weight is less than a first preset weight value to obtain the filtered parameter set; The process parameters included in the filtered parameter set are arranged and combined to obtain multiple parameter groups.

3. The method for controlling the powder production process as described in claim 2, characterized in that, The step of calculating the individual impact of each process parameter in the parameter set on particle size distribution to obtain the single-factor weight corresponding to each process parameter includes: Obtain the historical granularity distribution and the historical parameter values ​​of the process parameters within a preset time window; Calculate the first feature data of the historical granularity distribution; Calculate the second feature data of the historical parameter values; The correlation between the first feature data and the second feature data is calculated to obtain the single-factor weights corresponding to the process parameters.

4. The method for controlling the powder production process as described in claim 1, characterized in that, The step of generating a real-time control strategy based on the real-time influence weights corresponding to the process parameters includes: Obtain the target value of the unit control quantity corresponding to the particle size deviation value; The priority of the process parameters is determined based on the real-time impact weights corresponding to the process parameters; Obtain the priority multiple corresponding to the process parameters; The real-time adjustment amount of the process parameter is calculated based on the target value of the unit control quantity, the multiple corresponding to the priority of the process parameter, the real-time influence weight corresponding to the process parameter, and the granularity deviation value; wherein, the real-time control strategy includes the real-time adjustment amount of the process parameter.

5. The method for controlling the powder production process as described in claim 4, characterized in that, The step of obtaining the target value of the unit control quantity corresponding to the particle size deviation value includes: Constraints are generated based on the current parameter value of the first parameter; wherein, the first parameter is the parameter in the parameter set excluding the second parameter, and the second parameter is the process parameter whose real-time influence weight is greater than the second preset weight. A target optimization function is constructed based on the constraints and control objectives; wherein the decision variable of the target optimization function is the unit control variable; The target value of the unit control quantity is obtained by performing optimization calculations based on the target optimization function.

6. A control device for a powder production process, characterized in that, include: The acquisition unit is used to acquire the real-time particle size distribution during the powder production process; The calculation unit is used to calculate the real-time influence weight of the process parameters if the particle size deviation between the real-time particle size distribution and the target particle size distribution is greater than a preset threshold; wherein, the real-time influence weight is used to characterize the degree of real-time influence of the process parameters on the particle size distribution; the process parameters are the parameters of the production equipment involved in powder production; The generation unit is used to generate a real-time control strategy based on the real-time influence weights corresponding to the process parameters. A control unit is used to control the operation of the production equipment according to the real-time control strategy. The computing unit is further used for: A parameter set is obtained, which includes the process parameters of all production equipment involved in powder production; the process parameters included in the parameter set are arranged and combined to obtain multiple parameter groups; each parameter group includes at least two process parameters, and the combination of process parameters in different parameter groups is different; the interaction influence of each parameter group on particle size distribution is calculated to obtain the interaction weight corresponding to each parameter group; based on the interaction weight corresponding to each parameter group, the real-time influence weight corresponding to each process parameter in the parameter set is calculated; The computing unit is further used for: The process parameters in the parameter group are divided into a first discrete variable and a first continuous variable; Discretize the first continuous variable into a second discrete variable; The interaction effect of the parameter set on the granularity distribution is calculated based on the first discrete variable and the second discrete variable to obtain the first weight; Fit the first discrete variable to a second continuous variable; The interaction effect of the parameter set on the granularity distribution is calculated based on the first continuous variable and the second continuous variable to obtain the second weight; The interaction weights corresponding to the parameter group are calculated based on the first weight and the second weight.

7. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 5.

Images