Lyocell fiber intelligent production scheduling and quality prediction method and system

Abstract

The invention relates to the field of fiber production and quality prediction, in particular to an intelligent production scheduling and quality prediction method and system for lyocell fibers, and the method comprises the following steps: generating a batching rule table of the lyocell fibers by using a central batching method according to the polymerization degree requirement and batching principle of the lyocell fibers, generating an inventory table in combination with an inventory list of the pulp raw materials; a production plan is made based on the inventory table, so that a stock solution workshop worker prepares the Lyocell fiber pulp porridge according to the production plan; and constructing a relation model of the raw material attributes, the process parameters and the quality of the pulp porridge by using the two-stage quality prediction model to realize quality prediction of the Lyocell fiber pulp porridge. According to the method, intelligent production scheduling of the lyocell fibers can be realized based on a central batching method and inventory data, whole-course tracking from raw materials to swelling products is realized, then the quality prediction of the lyocell fiber pulp porridge is realized by using the two-stage quality prediction model, and the quality prediction effect of the lyocell pulp porridge is improved.

Description

Technical Field

[0001] The present application relates to the field of fiber production and quality prediction, and specifically to a method and system for intelligent production scheduling and quality prediction of lyocell fibers. Background Art

[0002] Currently, the formulation of Lyocell production plans is mostly based on manual formulation, which is highly random and cannot form a regular production schedule. It is difficult to achieve full tracking of the raw materials to the swelling products, and it is difficult to grasp the swelling quality of Lyocell fibers.

[0003] However, precise control of swelling quality is a crucial prerequisite for high-quality lyocell fiber production. Currently, swelling quality can only be controlled based on manual experience or inferred from the quality of subsequent production processes, lacking a reasonable means of predicting and controlling swelling quality. A swelling quality prediction model can be constructed based on full-process tracking data from raw materials to swelling products. Currently, machine learning models such as Gaussian process regression (GPR), random forest (RF), and support vector machine regression (SVM) have achieved significant results in quality prediction. However, a single model cannot fully meet the accuracy, robustness, and uncertainty requirements for lyocell pulp quality prediction. Summary of the Invention

[0004] In response to the defects in the prior art, the present invention provides a method and system for intelligent production scheduling and quality prediction of lyocell fibers.

[0005] In order to achieve the above-mentioned objectives, in a first aspect, the present invention provides a method for intelligent production scheduling and quality prediction of lyocell fiber, the method comprising the following steps: according to the requirements for the degree of polymerization and the batching principle of lyocell fiber, using the central batching method to generate a batching rule table for lyocell fiber; using the batching rule table and the inventory list of pulp raw materials for producing lyocell fiber, counting the inventory table that meets the batching rule table; formulating a production plan based on the inventory table, and then enabling the staff of the stock solution workshop to prepare lyocell fiber porridge according to the production plan; using a two-stage quality prediction model to construct a relationship model between raw material properties, process parameters and porridge quality, and realizing quality prediction of the lyocell fiber porridge. The present invention can realize intelligent production scheduling of lyocell fiber based on the central batching method and inventory data, realize full-process tracking from raw materials to swelling products, and then use the two-stage quality prediction model to realize quality prediction of lyocell fiber porridge, thereby improving the quality prediction effect of lyocell porridge.

[0006] Optionally, the central polymerization degree of material B is equal to the polymerization degree requirement, the polymerization degrees of material A and material C are symmetrically arranged with the polymerization degree of material B as the center, and the material A, the material B and the material C are three pulp raw materials for producing lyocell fiber.

[0007] Optionally, the central batching method satisfies the following principles: The degree of polymerization of the material B is taken as the central column of the ingredient rule table, and the degree of polymerization of the central column is equal to the mean of its symmetrical column pair; For each column in the batching rules table, , , is the central aggregation degree, is the lower limit of polymerization degree, is the upper limit of polymerization degree; Extremely poor , is the maximum degree of polymerization of the C material whose pulp code is C4, It is the minimum degree of polymerization of the A material with pulp code A4.

[0008] Optionally, the symmetrical column pair of the central column includes a column of polymerization degree of the A material and a column of polymerization degree of the C material.

[0009] Optionally, the polymerization degree of the material B is used as the central column of the ingredient rule table, and the polymerization degree of the central column is equal to the mean of its symmetric column pair, satisfying the following relationship: , in, is the degree of polymerization of the central column, that is, the degree of polymerization of the B material; is the degree of polymerization of the i-th column in the ingredient rule table, and is the degree of polymerization of the material A, and its pulp code is A(n+1-i); The first The polymerization degree of the column is the polymerization degree of the C material, and its pulp code is Ci; n is the number of types of the A material and the C material.

[0010] Optionally, formulating a production plan based on the inventory table, and then enabling staff in the stock solution workshop to prepare lyocell fiber pulp porridge according to the production plan, comprises the following steps: Generate a total production plan based on the inventory table, and take a predetermined amount of various pulps to the stock solution workshop according to the total production plan; After transporting the various pulps to the raw liquid workshop, they are placed in the line side warehouse according to the order of the various pulps in the ingredient rule table; A workshop production plan is generated based on the pulp inventory in the line-side warehouse, and the staff of the raw liquid workshop takes the material according to the workshop production plan and puts it into the feeding line; Based on the characteristics of the pulp fed into the feeding line and the weighing result, the feeding amounts of the solvent and the auxiliary agent are automatically calculated, thereby generating the lyocell fiber pulp porridge.

[0011] Optionally, the overall production plan is formulated in accordance with the following principles: The total number of material packages of A, B and C is 16. The number of material packages of A and C is the same, and the number of material packages of B is an even number. According to the order of material A, material B and material C in the ingredient rule table, the ingredient combination close to the polymerization degree requirement is given priority, which is (A1, B, C1), (A2, B, C2), ..., (An, B, Cn), where Ai and Ci are pulp codes, and n is the number of types of material A and material B.

[0012] Optionally, the use of a two-stage quality prediction model to construct a relationship model between raw material attributes, process parameters and porridge quality to achieve quality prediction of the lyocell fiber porridge comprises the following steps: Determine the parameters affecting the quality of porridge and the quality indicators of porridge, then construct a data set through a lyocell fiber porridge production experiment, and divide the data set into a training set and a test set; According to the training set, for any of the porridge quality indicators, GPR, RF and SVM are used to construct a first prediction model, a second prediction model and a third prediction model respectively; Using the first prediction model to obtain the predicted value GPR_pre of the porridge quality index in the test set, and then obtaining the standard deviation GPR_std of the predicted value; When GPR_std<0.5, GPR_pre is used as the prediction result of the porridge quality index; When GPR_std>0.5, the second prediction model and the third prediction model are used to obtain the predicted values ​​RF_pre and SVM_pre of the porridge quality index in the test set, respectively, and the mean of RF_pre and SVM_pre is used as the prediction result of the porridge quality index; The prediction results of the porridge quality index are compared with the real data in the test set to test the model performance.

[0013] Optionally, the porridge quality influencing parameters include NMMO input amount, NMMO concentration, temperature, swelling machine speed, pulp weight, pulp polymerization degree, pulp cellulose content, swelling time, PG input amount, PG concentration, HA input amount and HA concentration, and the porridge quality indicators include porridge shear viscosity, porridge NMMO concentration and porridge cellulose content.

[0014] In a second aspect, the present invention provides a lyocell fiber intelligent production scheduling and quality prediction system, the intelligent production scheduling and quality prediction system for lyocell fibers comprising: a data acquisition device, a data output device, a processor and a storage device, the storage comprising a computer-readable storage medium, the computer-readable storage medium storing a computer program, the computer program comprising program instructions, and when the program instructions are executed by the processor, the processor enables the processor to implement a lyocell fiber intelligent production scheduling and quality prediction method provided by the present invention.

[0015] In summary, the present invention has at least the following beneficial effects: 1. This method proposes a central batching method. This method, combined with inventory data, can realize the automatic formulation of production plans, namely smart scheduling. Smart scheduling can obtain a detailed workshop production plan. If materials are added based on this plan and the quality of the corresponding solvent and swelling product is recorded, the entire process from raw materials to swelling products can be tracked, which provides a good data foundation for the subsequent establishment of a swelling quality prediction model.

[0016] 2. This method combines machine learning models such as GPR, RF and SVM, and proposes a two-stage quality prediction model based on multiple machine learning models, which solves the problem that a single model is difficult to fully meet the accuracy, robustness and uncertainty requirements in the quality prediction of lyocell porridge.

[0017] 3. The present invention provides a system compatible with the method, which can improve the efficiency and practicality of intelligent production scheduling and quality prediction of lyocell fibers, and facilitate the promotion of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 Schematic diagram of a process for intelligent production scheduling and quality prediction of lyocell fibers according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the framework of a lyocell fiber intelligent production scheduling and quality prediction system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] Specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the present invention. In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention. However, it will be apparent to one of ordinary skill in the art that these specific details are not necessarily required to practice the present invention. In other instances, well-known circuits, software, or methods are not specifically described to avoid obscuring the present invention.

[0021] Throughout this specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present invention. Therefore, appearances of the phrases "in one embodiment," "in an embodiment," "an example," or "an example" in various places throughout this specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, or characteristics may be combined in any suitable combinations and / or subcombinations in one or more embodiments or examples. Furthermore, those of ordinary skill in the art will appreciate that the figures provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0022] It should be noted in advance that, in an optional embodiment, except for independent explanations, the same symbols or letters appearing in all formulas have the same meanings and values.

[0023] In an alternative embodiment, see Figure 1 The present invention provides a method for intelligent production scheduling and quality prediction of lyocell fibers, the method comprising the following steps: S1. Based on the requirements for the degree of polymerization and the batching principles of lyocell fiber, a batching rule table for lyocell fiber is generated using the central batching method.

[0024] Specifically, in this embodiment, for the three pulps A, B, and C used to produce lyocell fibers, the batching principle is as follows: the central polymerization degree of B is equal to the polymerization degree requirement, and the polymerization degrees of A and C are symmetrically arranged with the polymerization degree of B as the center. The central batching method satisfies the following principles: the polymerization degree of B is used as the central column of the batching rule table, the polymerization degree of the central column is equal to the mean of its symmetrical column pair, and the symmetrical column pair of the central column includes a column of polymerization degree of A and a column of polymerization degree of C; for each column in the batching rule table, , , is the central aggregation degree, is the lower limit of polymerization degree, The upper limit of polymerization degree; , The maximum degree of polymerization of pulp material C with the code name C4, The minimum degree of polymerization of material A with the pulp code A4. The degree of polymerization of the center column satisfies the following relationship:

[0025] Where DP is the degree of polymerization, is the degree of polymerization of the central column, that is, the degree of polymerization of material B; is the degree of polymerization of the i-th column in the ingredient rule table, and is the degree of polymerization of material A, and its pulp code is A(n+1-i); Ingredients Rules The polymerization degree of the column is the polymerization degree of C material, and its pulp code is Ci; n is the number of types of A material and C material.

[0026] More specifically, as shown in Table 1, the degree of polymerization of material B is the center column of the ingredient rule table, and the degrees of polymerization of materials A and C are located on both sides of the degree of polymerization of material B, and the pulp codes of the two are symmetrical about the center column. When the degree of polymerization is required to be 460, the value range of the degree of polymerization of material B is 457~463, and 460 is the center degree of polymerization of material B. According to the degree of polymerization of the center column, it is equal to the average of its symmetrical column pair and The degree of polymerization of material A and material C can be set symmetrically according to the requirements. For example, if the degree of polymerization of material A ranges from 436 to 442, that is, A3, then the degree of polymerization of material C should range from 478 to 484, that is, C3.

[0027] Table 1 Ingredient rules when the degree of polymerization is required to be 460

[0028] S2. Utilizing the ingredient rule table and the inventory list of pulp raw materials for producing lyocell fibers, a stock list that complies with the ingredient rule table is compiled.

[0029] Specifically, in this embodiment, the inventory list of pulp raw materials for producing lyocell fibers includes information such as the degree of polymerization, production batch number, and cellulose content of the pulp. A partial inventory list is shown in Table 2.

[0030] Table 2 Partial inventory list

[0031] After obtaining the inventory list, the production batch number of each pulp raw material can be obtained according to the polymerization degree required in the ingredient rule table, and then the inventory table that meets the ingredient rule table can be obtained, as shown in Table 3.

[0032] Table 3 Inventory table that complies with the ingredient rules

[0033] S3. Formulate a production plan based on the inventory table, and then enable the staff of the stock solution workshop to prepare the lyocell fiber pulp porridge according to the production plan.

[0034] Wherein, step S3 specifically includes the following steps: S31. Generate a total production plan based on the inventory table, and take a predetermined amount of various pulps to the raw liquid workshop according to the total production plan.

[0035] Specifically, in this embodiment, the following principles are followed when formulating the overall production plan: the total number of material packages of the three pulp raw materials, namely, material A, material B, and material C, is 16, the number of material packages of material A and material C is the same, and the number of material packages of material B is an even number; according to the order of material A, material B, and material C in the ingredient rule table, the ingredient combination close to the polymerization degree requirement is given priority, in the order of (A1, B, C1), (A2, B, C2), ..., (An, B, Cn), where Ai and Ci are pulp codes, and n=4.

[0036] Furthermore, for ease of understanding, the following explanation is given for “giving priority to ingredient combinations that are close to the polymerization degree requirements”: Taking the two ingredient combinations (A1, B, C1) and (A2, B, C2) in the row where the center polymerization degree is located as an example, in the ingredient rule table, the polymerization degree requirement is 460, and the polymerization degree of pulp A1 is closer to the polymerization degree of pulp B than that of pulp A2, and the polymerization degree of pulp C1 is closer to the polymerization degree of pulp B than that of pulp C2. Therefore, the ingredient combination (A1, B, C1) is closer to the polymerization degree requirement than the ingredient combination (A2, B, C2). Therefore, when formulating the overall production plan, the feeding order of the ingredient combination (A1, B, C1) should be before the feeding order of the ingredient combination (A2, B, C2).

[0037] More specifically, the overall production plan is shown in Table 4.

[0038] Table 4 Overall production plan

[0039] Generate the overall production plan and take half a month's supply of various pulps to the raw liquid workshop according to the overall production plan.

[0040] S32. After transporting the various pulps to the raw liquid workshop, the pulps are placed in the line-side warehouse according to the order of the pulps in the ingredient rule table.

[0041] Specifically, in this embodiment, the order of pulp storage in each line-side warehouse in the raw liquid workshop is the same as the order of pulp codes in the ingredient rule table.

[0042] S33. A workshop production plan is generated based on the pulp inventory in the line-side warehouse, and the staff of the raw liquid workshop takes the pulp according to the workshop production plan and puts it into the feeding line.

[0043] Specifically, in this embodiment, the method for generating a workshop production plan is similar to the method for generating a master production plan, the only difference being that this method is based on line-side warehouse inventory, while the master production plan is based on total inventory. The generated workshop production plan is shown in Table 5.

[0044] Table 5 Workshop production plan

[0045] S34. Automatically calculating the amount of solvent and additives fed based on the characteristics of the pulp fed from the feeding line and the weighing result, thereby generating the lyocell fiber pulp porridge.

[0046] Specifically, in this embodiment, the staff of the raw liquid workshop takes materials according to the workshop production plan and puts them into the feeding line for de-papering, weighing and other processing, and finally puts them into the swelling machine, and then calculates the input amount of solvent and additives. Then, the pulp, solvent and additives are mixed in the swelling machine, and react within a certain time, at a certain temperature and at a certain speed of the swelling machine rotor to finally generate lyocell fiber pulp porridge.

[0047] The solvent used in this embodiment is NMMO solution, i.e., N-methylmorpholine-N-oxide solution. The adjuvants used in this embodiment are PG solution and HA solution, i.e., propyl gallate and hydroxylamine solution.

[0048] S4. Use a two-stage quality prediction model to construct a relationship model between raw material properties, process parameters and pulp quality, so as to achieve quality prediction of the lyocell fiber pulp.

[0049] Wherein, step S4 specifically includes the following steps: S41. Determine the parameters affecting the quality of porridge and the quality indicators of porridge, and then construct a data set through a lyocell fiber porridge production experiment, and divide the data set into a training set and a test set.

[0050] Specifically, in this embodiment, the parameters affecting the quality of pulp porridge include NMMO input amount, NMMO concentration, temperature, swelling machine speed, pulp weight, pulp polymerization degree, pulp cellulose content, swelling time, PG input amount, PG concentration, HA input amount and HA concentration, and the pulp porridge quality indicators include pulp porridge shear viscosity, pulp porridge NMMO concentration and pulp porridge cellulose content.

[0051] Furthermore, after determining the parameters affecting pulp porridge quality and pulp porridge quality indicators, a data set was constructed through the Lyocell fiber pulp porridge production experiment, and the data set was divided into a training set and a test set in a ratio of 7:3.

[0052] S42. Based on the training set, for any of the porridge quality indicators, use GPR, RF and SVM to construct a first prediction model, a second prediction model and a third prediction model respectively.

[0053] Specifically, in this embodiment, GPR, RF and SVM are Gaussian process regression model, random forest model and support vector machine regression model respectively.

[0054] GPR is a probabilistic regression model based on a Bayesian nonparametric framework. It predicts output values ​​by defining a joint Gaussian distribution for the input data. It not only provides predictions but also quantifies the uncertainty of the predictions, making it suitable for scenarios with small samples, nonlinearities, and the need for confidence interval estimation. GPR is insensitive to noise and can naturally handle multi-output problems, but its computational complexity is relatively high, making it suitable for medium-sized datasets.

[0055] RF builds a model by integrating multiple decision trees. Each tree is independently trained based on a randomly sampled subset of data and features, and the final output is obtained through voting or averaging. RF is highly resistant to overfitting and can handle high-dimensional data and missing values, but the model has weak interpretability and is sensitive to extreme noise or class imbalance.

[0056] SVMs achieve classification or regression by finding an optimal hyperplane and using kernel functions to map data into a high-dimensional space, solving nonlinear problems. While SVMs perform well with small sample sizes and have strong generalization capabilities, kernel function selection and parameter tuning are complex, and computational costs rise significantly with increasing sample size.

[0057] The parameters affecting porridge quality are used as model input, and the porridge quality indicators are used as model output. For any porridge quality indicator, the training set is used to complete the training of GPR, RF and SVM respectively, and the first prediction model, second prediction model and third prediction model of the porridge quality indicator can be obtained.

[0058] S43. Use the first prediction model to obtain the predicted value GPR_pre of the porridge quality index in the test set, and then obtain the standard deviation GPR_std of the predicted value.

[0059] Specifically, in this embodiment, the first prediction model is used to obtain the predicted value GPR_pre of the porridge quality index in the test set, and the standard deviation GPR_std of the predicted value is calculated. This is the first stage of realizing the porridge quality prediction.

[0060] S44. When GPR_std<0.5, GPR_pre is used as the prediction result of the porridge quality index.

[0061] Specifically, in this embodiment, steps S44 and S45 are the second stage of achieving porridge quality prediction. If GPR_std<0.5, it means that the first prediction model is more confident in the prediction result, and GPR_pre can be directly used as the prediction result of the porridge quality index.

[0062] S45. When GPR_std>0.5, the second prediction model and the third prediction model are used respectively to obtain the predicted values ​​RF_pre and SVM_pre of the porridge quality index in the test set, and the average of RF_pre and SVM_pre is used as the prediction result of the porridge quality index.

[0063] Specifically, in this embodiment, when GPR_std>0.5, it means that the first prediction model is not confident in the prediction result. At this time, the mean of RF_pre and SVM_pre is used as the prediction result of the porridge quality index to improve robustness.

[0064] S46. Compare the predicted results of the porridge quality index with the real data in the test set to test the model performance.

[0065] Specifically, in this embodiment, a two-stage quality prediction model based on multiple machine learning models is proposed by combining machine learning models such as GPR, RF and SVM, which solves the problem that a single model is difficult to fully meet the accuracy, robustness and uncertainty requirements in the quality prediction of Lyocell porridge.

[0066] It should be noted that, in some cases, the actions described in the specification can be performed in a different order and still achieve the desired results. In this embodiment, the order of steps given is only to make the embodiment appear clearer and easier to explain, rather than to limit it.

[0067] In an alternative embodiment, see Figure 2 In order to improve the practicality of the method and facilitate the promotion of the method, the present invention also provides a lyocell fiber intelligent production scheduling and quality prediction system, which includes: a data acquisition device 1, a data output device 2, a processor 3 and a storage 4, the storage 4 includes a computer-readable storage medium, the computer-readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by the processor 3, the processor 3 implements the contents described in steps S1 to S4.

[0068] In summary, the present invention has the following beneficial effects: 1. This method proposes a central batching method. This method, combined with inventory data, can realize the automatic formulation of production plans, namely smart scheduling. Smart scheduling can obtain a detailed workshop production plan. If materials are added based on this plan and the quality of the corresponding solvent and swelling product is recorded, the entire process from raw materials to swelling products can be tracked, which provides a good data foundation for the subsequent establishment of a swelling quality prediction model.

[0069] 2. This method combines machine learning models such as GPR, RF and SVM, and proposes a two-stage quality prediction model based on multiple machine learning models, which solves the problem that a single model is difficult to fully meet the accuracy, robustness and uncertainty requirements in the quality prediction of lyocell porridge.

[0070] 3. The present invention provides a system compatible with the method, which can improve the efficiency and practicality of intelligent production scheduling and quality prediction of lyocell fibers, and facilitate the promotion of the method.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. A method for intelligent production scheduling and quality prediction of lyocell fibers, characterized in that: The steps include: According to the requirements of polymerization degree and batching principle of lyocell fiber, the batching rule table of lyocell fiber is generated by using the central batching method; Using the ingredient rule table and the inventory list of pulp raw materials for producing lyocell fiber, a stock list that complies with the ingredient rule table is compiled; Formulate a production plan based on the inventory table, and then enable the staff of the stock solution workshop to prepare the lyocell fiber pulp porridge according to the production plan; A two-stage quality prediction model is used to construct a relationship model between raw material properties, process parameters and pulp quality, so as to realize the quality prediction of the lyocell fiber pulp.

2. The method for intelligent production scheduling and quality prediction of lyocell fiber according to claim 1, wherein: The central polymerization degree of material B is equal to the polymerization degree requirement, and the polymerization degrees of materials A and C are symmetrically arranged with the polymerization degree of material B as the center. Material A, material B and material C are three pulp raw materials for producing lyocell fiber.

3. A lyocell fiber intelligent production scheduling and quality prediction method according to claim 2, characterized in that: The central batching method satisfies the following principles: The degree of polymerization of the material B is taken as the central column of the ingredient rule table, and the degree of polymerization of the central column is equal to the mean of its symmetrical column pair; For each column in the batching rules table, , , is the central aggregation degree, is the lower limit of polymerization degree, is the upper limit of polymerization degree; Extremely poor , is the maximum degree of polymerization of the C material whose pulp code is C4, It is the minimum degree of polymerization of the A material with pulp code A4.

4. The method for intelligent production scheduling and quality prediction of lyocell fiber according to claim 3, wherein: The symmetrical column pair of the central column includes a column of polymerization degree of the A material and a column of polymerization degree of the C material.

5. A lyocell fiber intelligent production scheduling and quality prediction method according to claim 3, characterized in that: The polymerization degree of the material B is taken as the central column of the ingredient rule table, and the polymerization degree of the central column is equal to the mean of its symmetric column pair, satisfying the following relationship: ， in, is the degree of polymerization of the central column, that is, the degree of polymerization of the B material; is the degree of polymerization of the i-th column in the ingredient rule table, and is the degree of polymerization of the material A, and its pulp code is A(n+1-i); The first The polymerization degree of the column is the polymerization degree of the C material, and its pulp code is Ci; n is the number of types of the A material and the C material.

6. A lyocell fiber intelligent production scheduling and quality prediction method according to claim 1, characterized in that: The step of formulating a production plan based on the inventory table and then enabling the staff of the stock solution workshop to prepare the lyocell fiber pulp porridge according to the production plan comprises the following steps: Generate a total production plan based on the inventory table, and take a predetermined amount of various pulps to the stock solution workshop according to the total production plan; After transporting the various pulps to the raw liquid workshop, they are placed in the line side warehouse according to the order of the various pulps in the ingredient rule table; A workshop production plan is generated based on the pulp inventory in the line-side warehouse, and the staff of the raw liquid workshop takes the material according to the workshop production plan and puts it into the feeding line; Based on the characteristics of the pulp fed into the feeding line and the weighing result, the feeding amounts of the solvent and the auxiliary agent are automatically calculated, thereby generating the lyocell fiber pulp porridge.

7. A method for intelligent production scheduling and quality prediction of lyocell fiber according to claim 6, characterized in that: The following principles are followed when formulating the overall production plan: The total number of material packages of A, B and C is 16. The number of material packages of A and C is the same, and the number of material packages of B is an even number. According to the order of material A, material B and material C in the ingredient rule table, the ingredient combination close to the polymerization degree requirement is given priority, which is (A1, B, C1), (A2, B, C2), ..., (An, B, Cn), where Ai and Ci are pulp codes, and n is the number of types of material A and material B.

8. The method for intelligent production scheduling and quality prediction of lyocell fiber according to claim 1, wherein: The method of using a two-stage quality prediction model to construct a relationship model between raw material attributes, process parameters and pulp quality to achieve quality prediction of the lyocell fiber pulp comprises the following steps: Determine the parameters affecting the quality of porridge and the quality indicators of porridge, then construct a data set through a lyocell fiber porridge production experiment, and divide the data set into a training set and a test set; According to the training set, for any of the porridge quality indicators, GPR, RF and SVM are used to construct a first prediction model, a second prediction model and a third prediction model respectively; Using the first prediction model to obtain the predicted value GPR_pre of the porridge quality index in the test set, and then obtaining the standard deviation GPR_std of the predicted value; When GPR_std<0.5, GPR_pre is used as the prediction result of the porridge quality index; When GPR_std>0.5, the second prediction model and the third prediction model are used to obtain the predicted values ​​RF_pre and SVM_pre of the porridge quality index in the test set, respectively, and the mean of RF_pre and SVM_pre is used as the prediction result of the porridge quality index; The prediction results of the porridge quality index are compared with the real data in the test set to test the model performance.

9. The method for intelligent production scheduling and quality prediction of lyocell fiber according to claim 8, characterized in that: The parameters affecting the porridge quality include NMMO input amount, NMMO concentration, temperature, swelling machine speed, pulp weight, pulp polymerization degree, pulp cellulose content, swelling time, PG input amount, PG concentration, HA input amount and HA concentration; the porridge quality indicators include porridge shear viscosity, porridge NMMO concentration and porridge cellulose content.

10. A lyocell fiber intelligent production scheduling and quality prediction system, characterized in that: The intelligent production scheduling and quality prediction system for lyocell fibers includes: a data acquisition device, a data output device, a processor, and a storage device. The storage device includes a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by the processor, the processor implements the intelligent production scheduling and quality prediction method for lyocell fibers as described in any one of claims 1 to 9.

Images