A process chain prediction model processing method and device, electronic equipment and storage medium

Abstract

This invention discloses a processing method, apparatus, electronic device, and storage medium for a process chain prediction model. The method includes: acquiring process parameters corresponding to multiple process units in a production process flow; constructing a unit prediction model for each process unit based on its process parameters; and obtaining a process chain prediction model for any process unit in the production process flow based on its unit prediction model and the unit prediction models of related process units. This invention can comprehensively consider multiple process units, fully integrate the process parameters of all process units in the production process flow, and pay attention to the interdependencies between process units, thereby improving the understanding and prediction of the entire production process flow, realizing the model association between the product and all relevant process parameters, and providing a robust and comprehensive control strategy to cope with fluctuations and deviations that occur during the production process.

Description

Technical Field

[0001] This invention relates to the field of pharmaceutical manufacturing technology, and in particular to a method, apparatus, electronic device, and storage medium for processing a process chain prediction model. Background Technology

[0002] Downstream production of protein drugs is a complex and delicate process involving multiple key steps in the extraction and purification of the target protein from cell culture medium. These steps aim to improve product purity and ensure the quality and performance of the final product. However, conventional downstream processes present several challenges to downstream process control that is geared towards product quality and performance.

[0003] First, the downstream process flow consists of multiple independent process units connected in series, each with different process and time constraints. This series connection leads to complex interdependencies between units, with the output of each unit directly serving as the input for the next. Therefore, any unit operation is affected by the previous one, and this cumulative effect makes it difficult to accurately predict and control the quality of the final product. Second, product quality is the cumulative effect of clarified cell culture medium across multiple process units. As the production process progresses through multiple stages, the degree of influence of process parameters and process units on product quality may vary. This variation increases the difficulty of process control, as precise control of different process parameters is required at different stages. Furthermore, fluctuations in cell culture medium quality or disturbances in any single process unit will propagate throughout the production process, affecting the quality and process performance of the final product. This propagation effect leads to differences in product quality and process performance between different batches, further increasing the complexity of process control.

[0004] The commonly used approach at present is to model each individual process unit independently. However, this approach has several problems. First, it lacks a holistic understanding and consideration of the production process, making it unable to accurately predict and control the cumulative effects of product quality and process performance. Second, models of individual process units can only predict the impact of parameter changes on product quality and process performance within a single step, failing to identify the impact on the final product quality and process performance. Therefore, predictions and risk assessments conducted using a single model are often inaccurate and cannot provide a robust and comprehensive control strategy to address fluctuations and deviations that occur during production. Summary of the Invention

[0005] This invention provides a method, apparatus, electronic device, and storage medium for processing a process chain prediction model, in order to solve the problem that predictions and risk assessments conducted using models for individual process units are inaccurate and cannot provide a robust and comprehensive control strategy to cope with fluctuations and deviations that occur during the production process.

[0006] According to one aspect of the present invention, a method for processing a process chain prediction model is provided, comprising:

[0007] Obtain the process parameters corresponding to multiple process units in the production process flow;

[0008] For each of the aforementioned process units, a unit prediction model for the process unit is constructed based on the process parameters of the process unit.

[0009] For any process unit in the production process, a process chain prediction model corresponding to the process unit is obtained based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0010] Wherein, for non-first process units, the associated process unit is at least one process unit in the production process flow that is located before the non-first process unit, and the process chain prediction model corresponding to the first process unit is the unit prediction model of the first process unit.

[0011] According to another aspect of the present invention, a processing apparatus for a process chain prediction model is provided, comprising:

[0012] The process parameter acquisition module is used to acquire the process parameters corresponding to multiple process units in the production process flow.

[0013] The unit prediction model construction module is used to construct a unit prediction model for each process unit based on the process parameters of the process unit.

[0014] The process chain prediction model determination module is used to obtain the process chain prediction model corresponding to any process unit in the production process flow based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0015] Wherein, for non-first process units, the associated process unit is at least one process unit in the production process flow that is located before the non-first process unit, and the process chain prediction model corresponding to the first process unit is the unit prediction model of the first process unit.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the processing method of the process chain prediction model according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the processing method of the process chain prediction model according to any embodiment of the present invention.

[0021] The technical solution of this invention involves obtaining process parameters corresponding to multiple process units in a production process flow; for each process unit, constructing a unit prediction model based on the process parameters; and for any process unit in the production process flow, obtaining a process chain prediction model corresponding to that process unit based on the unit prediction model and the unit prediction models of related process units. This approach comprehensively considers multiple process units, fully integrates the process parameters of all process units in the production process flow, and focuses on the interdependencies between process units, thereby improving the understanding and prediction of the entire production process flow. It achieves model association between the product and all relevant process parameters, enabling simultaneous analysis of the impact of multiple process units on the final product quality. This avoids the risk that optimizing a single process unit may lead to a decline in the performance of other process units, providing a robust and comprehensive control strategy to cope with fluctuations and deviations that occur during the production process.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of a process chain prediction model processing method provided in Embodiment 1 of the present invention;

[0025] Figure 2 This is a flowchart of a process chain prediction model processing method provided in Embodiment 2 of the present invention;

[0026] Figure 3This is a flowchart of a process chain prediction model processing method provided in Embodiment 3 of the present invention;

[0027] Figure 4 This is a comparison chart of the prediction results of the traditional single-step model and the prediction results of the process chain prediction model provided in Embodiment 3 of the present invention;

[0028] Figure 5 This is a flowchart of a process chain prediction model processing method provided in Embodiment 4 of the present invention;

[0029] Figure 6 This is a flowchart of a process chain prediction model processing method provided in Embodiment 5 of the present invention;

[0030] Figure 7 This is a schematic diagram of the structure of a processing device for a process chain prediction model provided in Embodiment Six of the present invention;

[0031] Figure 8 This is a schematic diagram of the structure of an electronic device provided in Embodiment 7 of the present invention. Detailed Implementation

[0032] To enable those skilled in the art to better understand the present invention, 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0034] Example 1

[0035] Figure 1This is a flowchart of a process chain prediction model processing method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where the product quality and process performance of each process unit in a production process are predicted, and the production process is controlled based on product quality and process performance. This method can be executed by a process chain prediction model processing device, which can be implemented in hardware and / or software, and can be configured in electronic devices such as computers and servers. Figure 1 As shown, the method includes:

[0036] S110. Obtain the process parameters corresponding to multiple process units in the production process flow.

[0037] In this context, a process unit refers to an operational unit within a production process that performs specific technological operations according to a defined process flow. For example, taking the production process of monoclonal antibody stock solution as an example, the downstream production process includes four process units: capture chromatography, depth filtration, anion exchange chromatography, and concentration and liquid exchange. Process parameters refer to various factors that affect the production process and product quality; specifically, process parameters include, but are not limited to, temperature, pressure, time, and flow rate.

[0038] In this embodiment of the invention, process parameters and predicted indicators for each process unit in the production process are collected. The predicted indicators are used to characterize product quality attributes or process performance; therefore, the predicted indicators can be product quality attributes, such as host cell protein residue, or process performance data, such as product recovery rate. It should be noted that the collected process parameters can be divided into a modeling set and a validation set. The modeling set data is used to establish the process chain model, and the validation set data is used to validate the process chain model.

[0039] In some embodiments, for each process unit, the process parameters of the process unit can be converted into a process parameter matrix, which can be represented by X. j This represents a process parameter matrix, and can also be represented using Y. j Let represent the prediction index matrix. Here, 1 ≤ j ≤ k, and k represents the number of process units. It can be understood that if the number of experiments for process unit j is n... j Then the process parameter matrix X j The size is n j ×i j Predictive index matrix Y j The size is n j ×q j For example, taking the production process of monoclonal antibody stock solution as an example, the number of experiments, process parameters, and predicted indicators for the four process units in the production process of monoclonal antibody stock solution are shown in Table 1:

[0040] Table 1

[0041]

[0042] S120. For each process unit, construct a unit prediction model for the process unit based on the process parameters of the process unit.

[0043] Among them, the unit prediction model is used to predict the prediction indicators of the process unit, and the unit prediction model can be a regression module.

[0044] It should be noted that in the production process, the unit prediction model of the first process unit represents the mapping relationship between the prediction index of the first process unit and the first influencing factor; the first influencing factor includes the process parameters in the first process unit; the unit prediction model of non-first process units represents the mapping relationship between the prediction index of non-first process units and the second influencing factor; the second influencing factor includes the process parameters of non-first process units and the prediction index of the process unit preceding the non-first process unit.

[0045] In this embodiment of the invention, for the first process unit, a unit prediction model for the first process unit is established based on the process parameters of the first process unit; specifically, since the second-order terms or interaction terms of the process parameters can also have a significant impact on the prediction index, the process parameter matrix X1 can be expanded to... Used for model fitting. Where X1 is a process parameter item, Let X1 be a quadratic term of the process parameters, and X1·X1 be an interaction term of the process parameters. It can be understood that if the size of the process parameter matrix X1 is n1×i1, then the process parameter extended matrix... The size is

[0046] The cell prediction model for the first process unit is: Where Y1 is the prediction index matrix for the first process unit (size n1×q1, where q is the prediction index and q1 is the prediction index for the first process unit). It is the process parameter extension matrix of the first process unit (size is...) )), It is the regression coefficient matrix (size is 1). ), ∈1 error term matrix (size is n1×q1).

[0047] In this embodiment of the invention, for non-first process units, a unit prediction model is established based on the process parameters of the non-first process unit and the prediction index of the preceding process unit. Specifically, similar to the method for constructing the unit prediction matrix of the first process unit, the process parameter matrix also needs to be expanded to obtain an expanded process parameter matrix. The expanded process parameter matrix includes process parameter terms, quadratic terms of process parameters, interaction terms of process parameters, and the prediction index of the preceding process unit. The process parameter matrix X can be... j Expanded into an extended matrix of process parameters Where X1 is a process parameter item, X1 is a quadratic term of the process parameters, X1·X1 is an interaction term of the process parameters, and Y is a quadratic term of the process parameters. j-1 This is a predictive indicator for the previous process unit. It can be understood that X... j The size is n j ×i j , The size is The "+1" is because the prediction index Y from the previous process unit was added. j-1 .

[0048] The cell prediction model for non-first process units is: Among them, Y j It is the prediction index matrix of the j-th process unit (of size n) j ×q j ), It is the extended matrix of process parameters for the j-th process unit. It is the regression coefficient matrix (size is 1). ), ∈ j Error term matrix (size n) j ×q j ).

[0049] In this embodiment of the invention, for each constructed unit prediction model of a process unit, the regression coefficients of the unit prediction model can be estimated using the least squares method to obtain the unit prediction model of the process unit.

[0050] S130. For any process unit in the production process, obtain the process chain prediction model corresponding to the process unit based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0051] It should be noted that for non-first process units, the associated process unit is at least one process unit in the production process flow that is located before the non-first process unit, and the process chain prediction model corresponding to the first process unit is the unit prediction model of the first process unit.

[0052] In this embodiment of the invention, for the first process unit, the unit prediction model of the first process unit is used as the process chain prediction model corresponding to the first process unit; for non-first process units, the unit prediction models of associated process units and the unit prediction models of non-first process units are merged to obtain the process chain prediction model corresponding to the non-first process unit.

[0053] Based on the above embodiments, optionally, obtaining the process chain prediction model corresponding to the process unit based on the unit prediction model of the process unit and the unit prediction model of the associated process unit includes: for any non-first process unit, based on the execution order of each process unit, using the unit prediction model of the previous process unit as the prediction index of the previous process unit, merging it into the unit prediction model of the next unit, until the unit prediction model of the associated process unit is merged into the unit prediction model of the non-first process unit, thereby obtaining the process chain prediction model corresponding to the non-first process unit.

[0054] In this embodiment of the invention, for any non-first process unit, the unit prediction model of the previous process unit is recursively used as the prediction index of the previous process unit and merged into the unit prediction model of the next process unit, until the unit prediction models of related process units are merged into the unit prediction model of the non-first process unit, thus obtaining the process chain prediction model corresponding to the non-first process unit. The process chain prediction model corresponding to the non-first process unit can be expressed as:

[0055]

[0056] Among them, g j It is a joint function that contains the process parameter extension matrix of the j-th process unit and the unit prediction model of the associated process units.

[0057] The technical solution of this embodiment obtains the process parameters corresponding to multiple process units in the production process flow; for each process unit, a unit prediction model of the process unit is constructed based on the process parameters of the process unit; for any process unit in the production process flow, a process chain prediction model corresponding to the process unit is obtained based on the unit prediction model of the process unit and the unit prediction models of related process units. This approach comprehensively considers multiple process units, fully integrates the process parameters of all process units in the production process flow, and focuses on the interdependencies between process units, thereby improving the understanding and prediction of the entire production process flow. It achieves model association between the product and all relevant process parameters, enabling simultaneous analysis of the impact of multiple process units on the final product quality. This avoids the risk that optimizing a single process unit may lead to a decline in the performance of other process units, providing a robust and comprehensive control strategy to cope with fluctuations and deviations that occur during the production process.

[0058] Example 2

[0059] Figure 2 This is a flowchart of a process chain prediction model processing method provided in Embodiment 2 of the present invention. Based on the above embodiments, the method may optionally include: obtaining the expected process parameters of each process unit; and using the process chain prediction model corresponding to each process unit to predict the expected process parameters of at least one process unit to obtain the prediction index of each process unit.

[0060] like Figure 2 As shown, the method includes:

[0061] S210. Obtain the process parameters corresponding to multiple process units in the production process flow.

[0062] S220. For each process unit, construct a unit prediction model for the process unit based on the process parameters of the process unit.

[0063] S230. For any process unit in the production process, obtain the process chain prediction model corresponding to the process unit based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0064] S240. Obtain the expected process parameters of each process unit; use the process chain prediction model corresponding to each process unit to perform prediction processing on the expected process parameters of at least one process unit to obtain the prediction index of each process unit.

[0065] The predicted process parameters can be obtained based on historical data, experience, or optimization algorithms. In this embodiment of the invention, the predicted process parameters of each process unit are substituted into the process chain prediction model corresponding to each process unit, and the predicted process parameters of at least one process unit are predicted to obtain the predicted index of each process unit.

[0066] The technical solution of this embodiment obtains the expected process parameters of each process unit; and uses the process chain prediction model corresponding to each process unit to predict the expected process parameters of at least one process unit, thereby obtaining the predicted indicators of each process unit. Based on these predicted indicators, the overall performance and potential risks of the production process are evaluated.

[0067] Example 3

[0068] Figure 3 This is a flowchart of a process chain prediction model processing method provided in Embodiment 3 of the present invention. In this embodiment, the unit prediction model of any process unit includes multiple initial parameter items, which include one or more of process parameter items, quadratic terms of the process parameters, and interaction terms of the process parameters. After constructing the unit prediction model of the process unit based on the process parameters of the process unit, the method further includes: determining the significance index corresponding to each of the multiple initial parameter items; filtering the multiple initial parameter items through the significance index corresponding to each of the multiple initial parameter items to obtain target parameter items; and updating the unit prediction model based on the target parameter items. Figure 3 As shown, the method includes:

[0069] S310. Obtain the process parameters corresponding to multiple process units in the production process flow.

[0070] S320. For each of the process units, construct a unit prediction model for the process unit based on the process parameters of the process unit.

[0071] S330. Determine the significance index corresponding to each of the plurality of initial parameter items; filter the plurality of initial parameter items using the significance index corresponding to each of the plurality of initial parameter items to obtain target parameter items; update the unit prediction model based on the target parameter items.

[0072] The significance index is used to evaluate the significance of the initial parameter items, facilitating the selection of these items. In this embodiment, a significance index corresponding to each of the multiple initial parameter items is determined based on a preset testing method. The selection is then performed based on the significance index of each initial parameter item, retaining parameter items that have a significant impact on the prediction result (i.e., target parameter items) and removing those with minor or insignificant impact. The unit prediction model is updated based on the target parameter items. The preset testing method includes, but is not limited to, T-tests and F-tests. For example, using the T-test, the significance index corresponding to each initial parameter item is calculated. Initial parameter items with significance indices lower than a preset significance level threshold are retained, while those with significance indices higher than the preset significance level threshold are removed, thereby updating the unit prediction model. The preset significance level threshold can be 0.05.

[0073] For example, taking the production process of monoclonal antibody stock solution as an example, the number of key process parameters for the four process units finally determined are 3 (process parameters: a1-a3), 4 (process parameters a4-a7), 1 (process parameter a8), and 5 (process parameters a9-a7). 13 ), and the predicted indicators (host cell protein residues) of the four process units were expressed in B j This indicates that 1≤j≤4.

[0074] The unit prediction models for the four process units are as follows:

[0075] The unit prediction model expression for the first process unit is: B1 = β 1.0 +β 1.1 *a1+β 1.2 *a2+β 1.3 *a3+β 1.4 *a1*a2.

[0076] The unit prediction model expression for the second process unit is: B2 = β 2.0 +β 2.1 *a4+β 2.2 *a5+β 2.3 *a6+β 2.4 *a7+β 2.5 *B1.

[0077] The unit prediction model expression for the third process unit is: B3 = β 3.1 +β 3.2 *a8+β 3.3 *B2.

[0078] The unit prediction model expression for the fourth process unit is: B4 = β 4.0 +β 4.1 *a9+β 4.2*a 10 +β 4.3 *a 11 +β 4.4 *a 12 +β 4.5 *a 13 +β 4.6 *B3.

[0079] S340. For any process unit in the production process flow, obtain the process chain prediction model corresponding to the process unit based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0080] For example, taking the production process of monoclonal antibody stock solution as an example, the expression of the process chain prediction model corresponding to the fourth process unit is:

[0081] B4 = β 4.0 +β 4.1 *a9+β 4.2 *a 10 +β 4.3 *a 11 +β 4.4 *a 12 +β 4.5 *a 13 +

[0082] β 4.6 *B3(β 3.1 +β 3.2 *a8+β 3.3 *B2(β 2.0 +β 2.1 *a4+β 2.2 *a5+β 2.3 *a6+

[0083] β 2.4 *a7+β 2.5 *(β 1.0 +β 1.1 *a1+β 1.2 *a2+β 1.3 *a3+β 1.4 *a1*a2))).

[0084] Figure 4 This is a comparison chart of the prediction results of the traditional single-step model and the prediction results of the process chain prediction model provided in Embodiment 3 of the present invention. Taking the production process of monoclonal antibody stock solution as an example, the prediction results of the traditional single-step model and the prediction results of the process chain prediction model in the production process of monoclonal antibody stock solution are as follows: Figure 4 As shown.

[0085] The technical solution of this embodiment involves determining the significance index corresponding to each of the plurality of initial parameter items; filtering the plurality of initial parameter items using the significance index to obtain target parameter items; and updating the unit prediction model based on the target parameter items. Updating the unit prediction model while retaining the initial parameter items that significantly affect the prediction results allows the updated model to better capture key information in process parameters, thereby improving the accuracy and reliability of predictions. Furthermore, it simplifies the unit prediction model and improves its generalization ability.

[0086] Example 4

[0087] Figure 5 This is a flowchart of a processing method for a process chain prediction model provided in Embodiment 4 of the present invention. Based on the above embodiments, the method further includes: obtaining the numerical range of process parameters in the process chain prediction model; sampling based on the numerical range of the process parameters to obtain multiple sets of simulated parameter values ​​for the process parameters in the process chain prediction model; predicting the multiple sets of simulated parameter values ​​based on the process chain prediction model to obtain prediction indices corresponding to the multiple sets of simulated parameter values; and adjusting the numerical range of the process parameters when the prediction indices corresponding to the multiple sets of simulated parameter values ​​meet the numerical range adjustment conditions.

[0088] like Figure 5 As shown, the method includes:

[0089] S410. Obtain the process parameters corresponding to multiple process units in the production process flow.

[0090] S420. For each process unit, construct a unit prediction model for the process unit based on the process parameters of the process unit.

[0091] S430. For any process unit in the production process, obtain the process chain prediction model corresponding to the process unit based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0092] S440. Obtain the numerical range of process parameters in the process chain prediction model; sample based on the numerical range of process parameters to obtain multiple sets of simulated parameter values ​​of process parameters in the process chain prediction model; predict the multiple sets of simulated parameter values ​​based on the process chain prediction model to obtain prediction indicators corresponding to the multiple sets of simulated parameter values; adjust the numerical range of process parameters when the prediction indicators corresponding to the multiple sets of simulated parameter values ​​meet the numerical range adjustment conditions.

[0093] The numerical range refers to the range of process parameters in the process chain prediction model. In this embodiment of the invention, the numerical range of process parameters in the process chain prediction model can be determined based on the research scope of process characterization, such as actual production experience, process requirements, or experimental data. It should be noted that the distribution of values ​​within the numerical range is uniform to ensure that the probability of each value being selected within the numerical range is equal. Further, sampling is performed within the numerical range of the process parameters to obtain multiple sets of simulated parameter values ​​in the process chain prediction model. The number of simulations is set, and simulation predictions are performed on each set of simulated parameter values ​​based on the process chain prediction model to obtain the prediction indicators corresponding to each set of simulated parameter values. Further, the prediction indicators corresponding to each set of simulated parameter values ​​are verified based on the acceptance standard range. If the proportion of the prediction indicators corresponding to each set of simulated parameter values ​​within the acceptance standard range is greater than or equal to a preset proportion, then the numerical range of the process parameters is considered to be a verified acceptable operating range, and no adjustment of the numerical range is required. If the proportion of the prediction indicators corresponding to each set of simulated parameter values ​​within the acceptance standard range is less than a preset proportion, then the numerical range of the process parameters is narrowed. For example, the acceptance criteria range can be the mean of the predicted indicator ± 3 times the standard deviation, and the preset ratio can be 95%.

[0094] It should be noted that this can be achieved by adjusting the upper and lower limits of the parameters, changing the distribution type of the parameters, or re-optimizing the parameters. The adjusted numerical range should be verified by simulation again until it meets the requirements of the acceptance criteria.

[0095] Example 5

[0096] Figure 6 This is a flowchart of a process chain prediction model processing method provided in Embodiment 5 of the present invention. Based on the above embodiments, the method further includes: obtaining simulated parameter values ​​of process parameters in each process unit; predicting the simulated parameter values ​​of process parameters of at least one process unit through the process chain prediction model of each process unit to obtain the prediction index corresponding to each process unit; and increasing the acceptance range corresponding to non-end process units when the prediction index of the end process unit in the production process meets the preset acceptance range.

[0097] like Figure 6 As shown, the method includes:

[0098] S510. Obtain the process parameters corresponding to multiple process units in the production process flow.

[0099] S520. For each of the process units, construct a unit prediction model for the process unit based on the process parameters of the process unit.

[0100] S530. For any process unit in the production process, obtain the process chain prediction model corresponding to the process unit based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0101] S540. Obtain the simulated parameter values ​​of the process parameters in each process unit; predict the simulated parameter values ​​of the process parameters of at least one process unit using the process chain prediction model of each process unit to obtain the prediction index corresponding to each process unit; if the prediction index of the terminal process unit in the production process meets the preset acceptance range, increase the acceptance range corresponding to the non-terminal process unit.

[0102] In this embodiment of the invention, the numerical range of process parameters in the process chain prediction model can be obtained; based on the numerical range of process parameters, sampling is performed to obtain multiple sets of simulated parameter values ​​for process parameters in the process chain prediction model; through the process chain prediction model of each process unit, the simulated parameter values ​​of process parameters for at least one process unit are predicted to obtain the prediction index corresponding to each process unit. If the prediction index of the terminal process unit in the production process flow meets the preset acceptance range, it indicates that the current production process flow is controllable in terms of prediction index and may have some optimization space, and the acceptance range corresponding to non-terminal process units can be increased. It is understood that if the prediction index of the terminal process unit is stable and meets expectations, then even if there are some fluctuations in the preceding process units, it will not have a significant impact on the final product. Increasing the acceptance range helps to reduce strict restrictions in the production process, improve production efficiency, and maintain stable product quality.

[0103] Example 6

[0104] Figure 7 This is a schematic diagram of the processing device for a process chain prediction model provided in Embodiment Six of the present invention. Figure 7 As shown, the device includes:

[0105] The process parameter acquisition module 610 is used to acquire the process parameters corresponding to multiple process units in the production process flow.

[0106] The unit prediction model construction module 620 is used to construct a unit prediction model for each process unit based on the process parameters of the process unit.

[0107] The process chain prediction model determination module 630 is used to obtain the process chain prediction model corresponding to any process unit in the production process flow based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0108] Wherein, for non-first process units, the associated process unit is at least one process unit in the production process flow that is located before the non-first process unit, and the process chain prediction model corresponding to the first process unit is the unit prediction model of the first process unit.

[0109] The technical solution of this embodiment obtains the process parameters corresponding to multiple process units in the production process flow; for each process unit, a unit prediction model of the process unit is constructed based on the process parameters of the process unit; for any process unit in the production process flow, a process chain prediction model corresponding to the process unit is obtained based on the unit prediction model of the process unit and the unit prediction models of related process units. This allows for comprehensive consideration of multiple process units, fully integrating the process parameters of all process units in the production process flow, while also focusing on the interdependencies between process units. This improves the understanding of the entire production process flow, enables model association between the product and all relevant process parameters, and allows for simultaneous analysis of the impact of multiple process units on the final product quality. It avoids the risk that optimizing a single process unit may lead to a decline in the performance of other process units, providing a robust and comprehensive control strategy to cope with fluctuations and deviations that occur during the production process.

[0110] Based on the above embodiments, optionally, the device further includes a predicted process parameter prediction module, used to obtain the predicted process parameters of each process unit; and to perform prediction processing on the predicted process parameters of at least one process unit through the process chain prediction model corresponding to each process unit to obtain the prediction index of each process unit.

[0111] Based on the above embodiments, optionally, in the production process flow, the unit prediction model of the first process unit represents the mapping relationship between the prediction index of the first process unit and the first influencing factor; the first influencing factor includes the process parameters in the first process unit;

[0112] The unit prediction model for the non-first process unit represents the mapping relationship between the prediction index of the non-first process unit and the second influencing factor, which includes the process parameters of the non-first process unit and the prediction index of the preceding process unit.

[0113] Based on the above embodiments, optionally, the process chain prediction model determination module 630 is specifically used to, for any non-first process unit, based on the execution order of each process unit, take the unit prediction model of the previous process unit as the prediction index of the previous process unit and merge it into the unit prediction model of the next unit, until the unit prediction model of the associated process unit is merged into the unit prediction model of the non-first process unit, so as to obtain the process chain prediction model corresponding to the non-first process unit.

[0114] Based on the above embodiments, optionally, the unit prediction model of any of the process units includes multiple initial parameter items, the initial parameter items including one or more of process parameter items, quadratic terms of the process parameters, and interaction terms of the process parameters; the device further includes a unit prediction model update module, used to determine the significance index corresponding to each of the multiple initial parameter items; to filter the multiple initial parameter items through the significance index corresponding to each of the multiple initial parameter items to obtain target parameter items; and to update the unit prediction model based on the target parameter items.

[0115] Optionally, based on the above embodiments, the device further includes a process parameter numerical range adjustment module, used to obtain the numerical range of process parameters in the process chain prediction model; sample based on the numerical range of process parameters to obtain multiple sets of simulated parameter values ​​of process parameters in the process chain prediction model; predict the multiple sets of simulated parameter values ​​based on the process chain prediction model to obtain prediction indicators corresponding to the multiple sets of simulated parameter values ​​respectively; and adjust the numerical range of process parameters when the prediction indicators corresponding to the multiple sets of simulated parameter values ​​meet the numerical range adjustment conditions.

[0116] Based on the above embodiments, optionally, the device further includes an acceptance range adjustment module, used to obtain the simulated parameter values ​​of the process parameters in each process unit; predict the simulated parameter values ​​of the process parameters of at least one process unit through the process chain prediction model of each process unit to obtain the prediction index corresponding to each process unit; and increase the acceptance range corresponding to the non-end process unit when the prediction index of the end process unit in the production process meets the preset acceptance range.

[0117] The processing apparatus for the process chain prediction model provided in the embodiments of the present invention can execute the processing method for the process chain prediction model provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0118] Example 7

[0119] Figure 8This is a schematic diagram of the structure of an electronic device provided in Embodiment 7 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0120] like Figure 8 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0121] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0122] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the process chain prediction model processing method.

[0123] In some embodiments, the process chain prediction model processing method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the process chain prediction model processing method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the process chain prediction model processing method by any other suitable means (e.g., by means of firmware).

[0124] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0125] Computer programs used for implementing the process chain prediction model of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0126] Example 8

[0127] Embodiment 8 of the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute a processing method for a process chain prediction model, the method comprising:

[0128] Obtain the process parameters corresponding to multiple process units in the production process flow;

[0129] For each process unit, a unit prediction model for the process unit is constructed based on the process parameters of the process unit.

[0130] For any process unit in the production process, a process chain prediction model corresponding to the process unit is obtained based on the unit prediction model of the process unit and the unit prediction model of the associated process units.

[0131] For non-first process units, the associated process unit is at least one process unit in the production process flow that is located before the non-first process unit, and the process chain prediction model corresponding to the first process unit is the unit prediction model of the first process unit.

[0132] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0133] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0134] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0135] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0136] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0137] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for processing a process chain prediction model, characterized in that, include: Obtain the process parameters corresponding to multiple process units in the production process flow; For each of the aforementioned process units, a unit prediction model for the process unit is constructed based on the process parameters of the process unit. For any process unit in the production process, a process chain prediction model corresponding to the process unit is obtained based on the unit prediction model of the process unit and the unit prediction model of the associated process units. Wherein, for non-first process units, the associated process unit is at least one process unit in the production process flow that is located before the non-first process unit, and the process chain prediction model corresponding to the first process unit is the unit prediction model of the first process unit.

2. The method according to claim 1, characterized in that, The method further includes: Obtain the expected process parameters for each of the aforementioned process units; By using the process chain prediction model corresponding to each process unit, the expected process parameters of at least one process unit are predicted to obtain the prediction index of each process unit.

3. The method according to claim 1, characterized in that, In the production process, the unit prediction model of the first process unit represents the mapping relationship between the prediction index of the first process unit and the first influencing factor; the first influencing factor includes the process parameters in the first process unit. The unit prediction model for the non-first process unit represents the mapping relationship between the prediction index of the non-first process unit and the second influencing factor, which includes the process parameters of the non-first process unit and the prediction index of the preceding process unit.

4. The method according to claim 3, characterized in that, The process chain prediction model corresponding to the process unit is obtained by using the unit prediction model based on the process unit and the unit prediction model of the associated process units, including: For any of the non-first process units, based on the execution order of each process unit, the unit prediction model of the previous process unit is used as the prediction index of the previous process unit and merged into the unit prediction model of the next unit, until the unit prediction models of the associated process units are merged into the unit prediction model of the non-first process unit, thus obtaining the process chain prediction model corresponding to the non-first process unit.

5. The method according to claim 1, characterized in that, The unit prediction model of any of the process units includes multiple initial parameter terms, which include one or more of the process parameter terms, the quadratic terms of the process parameters, and the interaction terms of the process parameters. After constructing the unit prediction model for the process unit based on its process parameters, the method further includes: Determine the significance index corresponding to each of the multiple initial parameter items; The target parameter is obtained by filtering the multiple initial parameter items using the significance index corresponding to each of the multiple initial parameter items; The unit prediction model is updated based on the target parameter.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the numerical range of process parameters in the process chain prediction model; Based on the numerical range of the process parameters, multiple sets of simulated parameter values ​​of the process parameters in the process chain prediction model are obtained by sampling. Based on the process chain prediction model, the multiple sets of simulated parameter values ​​are predicted respectively to obtain the prediction index corresponding to each set of simulated parameter values; when the prediction index corresponding to each set of simulated parameter values ​​meets the numerical range adjustment condition, the numerical range of the process parameters is adjusted.

7. The method according to claim 1, characterized in that, The method further includes: Obtain the simulated parameter values ​​of the process parameters in each of the aforementioned process units; By using the process chain prediction model of each process unit, the simulated parameter values ​​of the process parameters of at least one process unit are predicted, and the prediction index corresponding to each process unit is obtained. If the predicted indicators of the end process unit in the production process flow meet the preset acceptance range, the acceptance range corresponding to the non-end process unit is increased.

8. A processing device for a process chain prediction model, characterized in that, include: The process parameter acquisition module is used to acquire the process parameters corresponding to multiple process units in the production process flow. The unit prediction model construction module is used to construct a unit prediction model for each process unit based on the process parameters of the process unit. The process chain prediction model determination module is used to obtain the process chain prediction model corresponding to any process unit in the production process flow based on the unit prediction model of the process unit and the unit prediction model of the associated process units. Wherein, for non-first process units, the associated process unit is at least one process unit in the production process flow that is located before the non-first process unit, and the process chain prediction model corresponding to the first process unit is the unit prediction model of the first process unit.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the processing method of the process chain prediction model according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the processing method of the process chain prediction model according to any one of claims 1-7.

Images