Self-adaptive weight control method, device and system for expanded tobacco shreds before moisture regaining

Abstract

According to the self-adaptive control method, device and system method for the weight of the expanded tobacco shreds before moisture regaining, smooth denoising data preprocessing is carried out on related data, data correlation analysis is carried out on main factors of the weight of the tobacco shreds before moisture regaining and the influence degree of the main factors, and data with high correlation is selected as input of a radial basis function neural network; a before-moisture-regaining tobacco shred weight estimation model based on the radial basis function neural network is established; based on a tobacco shred estimated value output by a tobacco shred weight estimation model, a self-adaptive control method is introduced, the weight conversion proportion of dry ice tobacco shred after the tobacco shred expansion process is approached through a self-adaptive law, and a quantitative band frequency set value is calculated according to the relation of the expanded tobacco shred weight change, the average flow and the control period. The method can effectively improve the estimation precision of the weight of the tobacco shreds before moisture regaining, and realizes the indirect self-adaptive control of the moisture fluctuation of the expanded tobacco shreds before moisture regaining based on the weight estimation of the expanded tobacco shreds before moisture regaining through the self-adaptive control of the quantitative band frequency.

Description

Technical Field

[0001] The present invention relates to a method, a device and a system for adaptively controlling the weight of expanded tobacco before moisture regain, and belongs to the technical field of digital twins for expanded tobacco production. Background Art

[0002] The moisture regeneration of expanded tobacco is an important process in the cigarette production process. The weight of expanded tobacco before moisture regeneration is one of the important factors affecting the moisture content of expanded tobacco, and the moisture content control of expanded tobacco is a key factor in ensuring the quality of tobacco. On the one hand, due to the hysteresis of the conventional method of feedback of tobacco weight through sensors, it is necessary to estimate the weight of expanded tobacco before moisture regeneration in advance; some methods use the change in the tobacco mass in the vibration bin during the control cycle to estimate the tobacco weight, and some methods use the tobacco stock and tobacco blending ratio to estimate the actual tobacco stock in the feeder; however, in the current tobacco weight estimation methods, the estimation modeling of tobacco weight is mainly based on traditional mechanisms and regression methods, and neural networks are rarely used for estimation modeling; and if unprocessed data is directly input into the neural network for training, or data with weak correlation is used as the input of the estimation model, due to problems such as data correlation and data misalignment, it is easy to cause the estimation model neural network training time to be long and the training results to be unsatisfactory, affecting the estimation accuracy. On the other hand, the current mainstream control method for tobacco weight in domestic cigarette production lines is to calculate the flow rate of dry ice tobacco within a certain period of time based on the change in tobacco mass in the vibration bin, and then the workers adjust the quantitative belt frequency according to the operation of the electronic belt scale to control the weight of dry ice tobacco by adjusting the quantitative belt frequency; however, this control method does not have the ability to predict the weight of tobacco and the ability to quickly adapt to different working conditions. It has lags and uncertainties, resulting in large fluctuations in the weight of expanded tobacco in different batches, affecting the control performance. Summary of the invention

[0003] The purpose of the present invention is to provide a method, device and system for adaptively controlling the weight of expanded tobacco before moisture regain, so as to achieve adaptively controlling the weight of expanded tobacco before moisture regain through an indirect adaptive control method.

[0004] In order to achieve the above objectives / solve the above technical problems, the present invention is implemented by adopting the following technical solutions.

[0005] In a first aspect, the present invention provides a method for adaptively controlling the weight of expanded tobacco before moisture reconditioning, comprising:

[0006] Obtain data on factors affecting the weight of expanded tobacco during the expansion process;

[0007] Inputting the data of factors affecting the weight of expanded tobacco into a pre-trained tobacco weight estimation model to obtain a predicted value of the weight of expanded tobacco before rehumidification;

[0008] Based on the predicted weight of expanded tobacco before rehydration, an adaptive law is set to approximate the weight conversion ratio of dry ice tobacco after the expansion process to obtain the predicted weight of dry ice tobacco;

[0009] According to the difference between the set value of dry ice tobacco weight and the predicted value of dry ice tobacco weight, PID adjustment is performed on the data of factors affecting the weight of expanded tobacco during the expansion process, and then the data of factors affecting the weight of expanded tobacco during the expansion process are adjusted to control the weight of dry ice tobacco entering the expansion process so that the weight of expanded tobacco before rehumidification is maintained at the target value.

[0010] Furthermore, the acquisition of data on factors affecting the weight of expanded tobacco shreds during the expansion process specifically includes the following steps:

[0011] The Pearson correlation coefficient method is used to perform correlation analysis on several sample data in the expansion process of a set batch of tobacco, and the sample data and the actual value of the expanded tobacco weight are set as two variables, and the Pearson coefficient of the two variables is defined. The sample data between -1 and 1 are the factors affecting the weight of expanded tobacco;

[0012] ,

[0013] in: is the actual weight value of the i-th expanded tobacco in the set batch, To set the i-th data sample in the batch, is the total number of samples in the same batch, is the sample mean of the weight of expanded tobacco, is the sample mean of the sample data;

[0014] The sample data include: metering belt frequency, sublimation tube hot air temperature, metering belt tobacco height, sublimation tube return air temperature and expanded tobacco moisture before rehumidification.

[0015] Furthermore, the first 70% of the data on factors affecting the quality of expanded tobacco in the batch expansion process are set as a training set, and the last 30% of the data are set as a test set.

[0016] Furthermore, the training of the tobacco weight estimation model includes:

[0017] The K-means clustering algorithm is used to perform cluster analysis on the sample data in the training set, and the sample data in the training set is divided into several clusters. The centroid of each cluster is used as the central vector of the corresponding hidden layer unit;

[0018] Calculate the width parameter of each hidden unit based on the Euclidean distance between the centroids;

[0019] For each sample data in the training set, calculate its activation value on all hidden units based on the Gaussian radial basis function, and combine these activation values ​​into the hidden layer output matrix;

[0020] The weight matrix of the output layer is optimized and solved by the least square method or gradient descent method. The optimization goal is to minimize the nominal error between the predicted value of expanded tobacco quality and the actual value of expanded tobacco quality to less than 5%;

[0021] The performance of the radial basis neural network model is optimized by continuously iteratively adjusting the center vector, width parameter and output layer weight until the value of the loss function converges or reaches a predetermined maximum number of iterations, thereby obtaining the final trained radial basis neural network model;

[0022] Where: Gaussian radial basis function:

[0023] ,

[0024] in, is the jth sample data in the training set, For the The center of the kernel function of the hidden unit,

[0025] For the The width of the kernel function of hidden units, For the The radial basis functions of the hidden units of .

[0026] Furthermore, obtaining the predicted value of dry ice tobacco weight specifically includes the following steps:

[0027] ,

[0028] ,

[0029] in: is the predicted value of the expanded tobacco weight before rehydration, is the predicted weight of dry ice tobacco, is the adaptive law, is the dry ice tobacco conversion parameter, Set a value for the dry ice tobacco weight.

[0030] Further, the PID adjustment is performed on the data of factors affecting the weight of expanded tobacco in the expansion process according to the difference between the set value of the dry ice tobacco weight and the predicted value of the dry ice tobacco weight, and then the data of factors affecting the weight of expanded tobacco in the expansion process is adjusted to control the weight of dry ice tobacco entering the expansion process so that the weight of expanded tobacco before rehumidification is maintained at the target value. Specifically, it includes:

[0031] Quantitative band frequency adjustment, including:

[0032] ,

[0033] in: is the quantitative band frequency change value, The dosing band frequency is set for the dry ice tobacco weight setting value, Set the value for the dry ice tobacco weight. The dosing band frequency set for the predicted dry ice tobacco weight, The frequency of the average quantitative band accumulated in one control cycle. is the dry ice tobacco flow rate in one control cycle,

[0034] ,

[0035] in: is the i-th common quantitative band frequency in production, is a control cycle, m is the number of commonly selected quantitative band frequencies,

[0036] ,

[0037] in: For quantitative belt travel, It is the ratio coefficient of quantitative belt stroke to control cycle or quantitative belt frequency;

[0038] The quantitative belt frequency is PID adjusted. According to the adjusted quantitative belt frequency, all the dry ice tobacco on the quantitative belt is output into the expansion process within one control cycle, and the weight of the expanded tobacco before rehumidification is controlled to be maintained at the target value.

[0039] In a second aspect, the present invention provides a device for adaptively controlling the weight of expanded tobacco before moisture regeneration, wherein the acquisition module is used to acquire data on factors affecting the weight of expanded tobacco during the expansion process;

[0040] An input module is used to input the data of factors affecting the weight of expanded tobacco into a pre-trained tobacco weight estimation model to obtain a predicted value of the weight of expanded tobacco before rehumidification;

[0041] An adaptive module is used to set an adaptive law to approximate the weight conversion ratio of dry ice tobacco after the expansion process based on the predicted weight of expanded tobacco before rehydration, so as to obtain the predicted weight of dry ice tobacco;

[0042] The adjustment module is used to perform PID adjustment on the data of factors affecting the weight of expanded tobacco during the expansion process according to the difference between the set value of the dry ice tobacco weight and the predicted value of the dry ice tobacco weight, thereby adjusting the weight of the dry ice tobacco during the expansion process to maintain it at the set value and controlling the weight of the expanded tobacco before rehumidification to maintain it at the target value.

[0043] In a third aspect, the present invention provides an adaptive control system for the weight of expanded tobacco before moisture reconditioning, characterized in that it comprises:

[0044] Memory, for storing computer programs / instructions;

[0045] A processor is used to execute the computer program / instructions to implement the steps of the above-mentioned method for adaptively controlling the weight of expanded tobacco before moisture reconditioning.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for adaptively controlling the weight of expanded tobacco before rehydration.

[0047] In a fifth aspect, the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned method for adaptively controlling the weight of expanded tobacco before rehydration.

[0048] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention aims at the problem of estimating the weight of expanded tobacco, adopts the Pearson correlation coefficient method to perform data correlation analysis, and calculates the main influencing factors affecting the estimation of the weight of expanded tobacco before rehumidification: quantitative belt frequency, sublimation tube hot air temperature, and metering belt tobacco height; on this basis, a model for estimating the weight of expanded tobacco based on the main influencing factors is proposed, which can preliminarily estimate the weight of expanded tobacco, and then intervene in the quantitative belt frequency in advance according to the adaptive control method, which greatly reduces the lag of the control of the weight of expanded tobacco, effectively ensures the estimation accuracy and control performance of the weight of expanded tobacco before rehumidification, thereby realizing indirect adaptive control of the quality fluctuation of expanded tobacco before rehumidification. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a block diagram of the adaptive feedback control of expanded tobacco weight for this application;

[0050] Figure 2 A data preprocessing flow chart for estimating the weight of expanded tobacco in this application;

[0051] Figure 3 This is an example diagram of tobacco weight estimation for this application. DETAILED DESCRIPTION

[0052] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0053] The term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " generally indicates that the related objects are in an "or" relationship.

[0054] Example 1

[0055] like Figures 1 to 3 An embodiment shown in the figure provides a method for adaptively controlling the weight of expanded tobacco before moisture regain, comprising:

[0056] Obtain data on factors affecting the weight of expanded tobacco during the expansion process;

[0057] The acquired data needs to be preprocessed, including: data screening: the original data of the selected factors affecting the weight of expanded tobacco contains a large amount of invalid data and irrelevant variable data, which are streamlined and screened and re-sorted in batches;

[0058] Time-delay translation: time-delay translation of data such as tobacco height and tobacco flow rate of metering belt;

[0059] Eliminate abnormal data from batch data: Take batch production data as the unit, and eliminate abnormal values ​​in the two time periods of each batch, when the production line just starts working and when the production line is about to end working; for example, when the quantitative band frequency data stays at a fixed value for a long time, only the first data is retained.

[0060] The method of obtaining the data of factors affecting the weight of expanded tobacco shreds during the expansion process specifically comprises the following steps:

[0061] The Pearson correlation coefficient method is used to perform correlation analysis on several sample data in the expansion process of a set batch of tobacco, and the sample data and the actual value of the expanded tobacco weight are set as two variables, and the Pearson coefficient of the two variables is defined. The sample data between -1 and 1 are the factors affecting the weight of expanded tobacco;

[0062] ,

[0063] in: is the actual weight value of the i-th expanded tobacco in the set batch, To set the i-th data sample in the batch, is the total number of samples in the same batch, is the sample mean of the weight of expanded tobacco, is the sample mean of the sample data;

[0064] The sample data includes: metering belt frequency, sublimation tube hot air temperature, metering belt tobacco height, sublimation tube return air temperature and expanded tobacco moisture before rehumidification;

[0065] According to the sample data and the actual value of the expanded tobacco weight, the data with strong correlation are selected as the input of the radial basis function neural network (RBFNN), including: the frequency of the metering belt, the hot air temperature of the sublimation tube, and the height of the tobacco in the metering belt.

[0066] The first 70% of the data on factors affecting the quality of expanded tobacco in the process of batch expansion are taken as the training set, and the last 30% of the data are taken as the test set.

[0067] Inputting the data of factors affecting the weight of expanded tobacco into a pre-trained tobacco weight estimation model to obtain a predicted value of the weight of expanded tobacco before rehumidification;

[0068] The training of the tobacco weight estimation model includes:

[0069] The K-means clustering algorithm is used to perform cluster analysis on the sample data in the training set, and the sample data in the training set is divided into several clusters. The centroid of each cluster is used as the central vector of the corresponding hidden layer unit;

[0070] Calculate the width parameter of each hidden unit based on the Euclidean distance between the centroids;

[0071] For each sample data in the training set, calculate its activation value on all hidden units based on the Gaussian radial basis function, and combine these activation values ​​into the hidden layer output matrix;

[0072] The weight matrix of the output layer is optimized and solved by the least square method or the gradient descent method. The optimization goal is to minimize the nominal error between the predicted value of the expanded tobacco cut mass and the actual value of the expanded tobacco cut mass to less than 5%, and the nominal error between the model estimated value and the actual value is calculated and tested using the test set;

[0073] The performance of the radial basis neural network (RBFNN) model is optimized by continuously iteratively adjusting the center vector, width parameter and output layer weight until the value of the loss function converges or reaches the predetermined maximum number of iterations, and the final trained radial basis neural network model is obtained; RBFNN has a three-layer structure of input layer, hidden layer and output layer. The input layer and hidden layer only play the role of data transmission, and the output layer is the linear weighted sum of hidden units. The hidden layer activation function is the radial basis function, and the Gaussian radial basis function is selected:

[0074] ,

[0075] in, is the jth sample data in the training set, For the The center of the kernel function of the hidden unit, For the The width of the kernel function of hidden units, For the The radial basis functions of the hidden units of ;

[0076] Based on the predicted weight value of expanded tobacco before rehydration, an adaptive law is set to approximate the weight conversion ratio of dry ice tobacco after the expansion process, and the predicted weight value of dry ice tobacco is obtained, which specifically includes the following steps:

[0077] ,

[0078] ,

[0079] in: is the predicted value of the expanded tobacco weight before rehydration, is the predicted weight of dry ice tobacco, is the adaptive law, is the dry ice tobacco conversion parameter, Set a value for the dry ice tobacco weight;

[0080] According to the difference between the set value of dry ice tobacco weight and the predicted value of dry ice tobacco weight, PID adjustment is performed on the data of factors affecting the weight of expanded tobacco in the expansion process, and then the data of factors affecting the weight of expanded tobacco in the expansion process are adjusted to control the weight of dry ice tobacco entering the expansion process, so that the weight of expanded tobacco before rehumidification is maintained at the target value, including:

[0081] Quantitative band frequency adjustment values, including:

[0082] ,

[0083] in: is the quantitative band frequency change value, The dosing band frequency is set for the dry ice tobacco weight setting value, Set the value for the dry ice tobacco weight. The dosing band frequency set for the predicted dry ice tobacco weight, The frequency of the average quantitative band accumulated in one control cycle. is the dry ice tobacco flow rate in one control cycle,

[0084] ,

[0085] in: is the i-th common quantitative band frequency in production, is a control cycle, m is the number of commonly selected quantitative band frequencies,

[0086] ,

[0087] in: For quantitative belt travel, It is the ratio coefficient of quantitative belt stroke to control cycle or quantitative belt frequency;

[0088] The quantitative belt frequency is PID adjusted. According to the adjusted quantitative belt frequency, all the dry ice tobacco on the quantitative belt is output into the expansion process within one control cycle, and the weight of the expanded tobacco before rehumidification is controlled to be maintained at the target value.

[0089] Example 2

[0090] This embodiment provides a device for adaptively controlling the weight of expanded tobacco before moisture regain, wherein the acquisition module is used to acquire data on factors affecting the weight of expanded tobacco during the expansion process;

[0091] An input module is used to input the data of factors affecting the weight of expanded tobacco into a pre-trained tobacco weight estimation model to obtain a predicted value of the weight of expanded tobacco before rehumidification;

[0092] An adaptive module is used to set an adaptive law to approximate the weight conversion ratio of dry ice tobacco after the expansion process based on the predicted weight of expanded tobacco before rehydration, so as to obtain the predicted weight of dry ice tobacco;

[0093] The adjustment module is used to perform PID adjustment on the data of factors affecting the weight of expanded tobacco in the expansion process according to the difference between the set value of the dry ice tobacco weight and the predicted value of the dry ice tobacco weight, and then adjust the data of factors affecting the weight of expanded tobacco in the expansion process, and control the weight of dry ice tobacco entering the expansion process, so that the weight of expanded tobacco before rehumidification is maintained at the target value.

[0094] Example 3

[0095] This embodiment provides an adaptive control system for the weight of expanded tobacco before moisture reconditioning, which is characterized by comprising:

[0096] Memory, for storing computer programs / instructions;

[0097] A processor is used to execute the computer program / instructions to implement the steps of the above-mentioned method for adaptively controlling the weight of expanded tobacco before moisture reconditioning.

[0098] Example 4

[0099] This embodiment provides a computer-readable storage medium on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of the above-mentioned method for adaptively controlling the weight of expanded tobacco before moisture regain are implemented.

[0100] Example 5

[0101] This embodiment provides a computer program product, including a computer program / instruction, which implements the steps of the above-mentioned method for adaptively controlling the weight of expanded tobacco before moisture regain when the computer program / instruction is executed by a processor.

[0102] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0104] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0105] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0106] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for adaptively controlling the weight of expanded tobacco before moisture reconditioning, characterized in that: include: Obtain data on factors affecting the weight of expanded tobacco during the expansion process; Inputting the data of factors affecting the weight of expanded tobacco into a pre-trained tobacco weight estimation model to obtain a predicted value of the weight of expanded tobacco before rehumidification; Based on the predicted weight of expanded tobacco before rehydration, an adaptive law is set to approximate the weight conversion ratio of dry ice tobacco after the expansion process to obtain the predicted weight of dry ice tobacco; According to the difference between the set value of dry ice tobacco weight and the predicted value of dry ice tobacco weight, PID adjustment is performed on the data of factors affecting the weight of expanded tobacco during the expansion process, and then the data of factors affecting the weight of expanded tobacco during the expansion process are adjusted to control the weight of dry ice tobacco entering the expansion process so that the weight of expanded tobacco before rehumidification is maintained at the target value.

2. The method for adaptively controlling the weight of expanded tobacco before moisture reconditioning according to claim 1, characterized in that: The method of obtaining the data of factors affecting the weight of expanded tobacco shreds during the expansion process specifically comprises the following steps: The Pearson correlation coefficient method is used to perform correlation analysis on several sample data in the expansion process of a set batch of tobacco, and the sample data and the actual value of the expanded tobacco weight are set as two variables, and the Pearson coefficient of the two variables is defined. The sample data between -1 and 1 are the factors affecting the weight of expanded tobacco; ， in: is the actual weight value of the i-th expanded tobacco in the set batch, To set the i-th data sample in the batch, is the total number of samples in the same batch, is the sample mean of the weight of expanded tobacco, is the sample mean of the sample data; The sample data include: metering belt frequency, sublimation tube hot air temperature, metering belt tobacco height, sublimation tube return air temperature and expanded tobacco moisture before rehumidification.

3. The method for adaptively controlling the weight of expanded tobacco before moisture reconditioning according to claim 2, characterized in that: The first 70% of the data on factors affecting the quality of expanded tobacco in the process of batch expansion are taken as the training set, and the last 30% of the data are taken as the test set.

4. The method for adaptively controlling the weight of expanded tobacco before moisture reconditioning according to claim 3, characterized in that: The training of the tobacco weight estimation model includes: The K-means clustering algorithm is used to perform cluster analysis on the sample data in the training set, and the sample data in the training set is divided into several clusters. The centroid of each cluster is used as the central vector of the corresponding hidden layer unit; Calculate the width parameter of each hidden unit based on the Euclidean distance between the centroids; For each sample data in the training set, calculate its activation value on all hidden units based on the Gaussian radial basis function, and combine these activation values ​​into the hidden layer output matrix; The weight matrix of the output layer is optimized and solved by the least square method or gradient descent method. The optimization goal is to minimize the nominal error between the predicted value of expanded tobacco quality and the actual value of expanded tobacco quality to less than 5%; The performance of the radial basis neural network model is optimized by continuously iteratively adjusting the center vector, width parameter and output layer weight until the value of the loss function converges or reaches a predetermined maximum number of iterations, thereby obtaining the final trained radial basis neural network model; Where: Gaussian radial basis function: ， in, is the jth sample data in the training set, For the The center of the kernel function of the hidden unit, For the The width of the kernel function of hidden units, For the The radial basis functions of the hidden units of .

5. The method for adaptively controlling the weight of expanded tobacco before moisture reconditioning according to claim 4, characterized in that: The method of obtaining the predicted value of dry ice tobacco weight specifically comprises the following steps: ， ， in: is the predicted value of the expanded tobacco weight before rehydration, is the predicted weight of dry ice tobacco, is the adaptive law, is the dry ice tobacco conversion parameter, Set a value for the dry ice tobacco weight.

6. The method for adaptively controlling the weight of expanded shredded tobacco before moisture reconditioning according to claim 5, characterized in that: The method of performing PID adjustment on the data of factors affecting the weight of expanded tobacco in the expansion process according to the difference between the set value of the dry ice tobacco weight and the predicted value of the dry ice tobacco weight, and then adjusting the data of factors affecting the weight of expanded tobacco in the expansion process, and controlling the weight of dry ice tobacco entering the expansion process, so that the weight of expanded tobacco before rehumidification is maintained at the target value, specifically comprises the following steps: Adjustment of the quantitative band frequency of factors affecting the weight of expanded tobacco, including: ， in: is the quantitative band frequency change value, The dosing band frequency is set for the dry ice tobacco weight setting value, Set the value for the dry ice tobacco weight. The dosing band frequency set for the predicted dry ice tobacco weight, The frequency of the average quantitative band accumulated in one control cycle. is the dry ice tobacco flow rate in one control cycle, ， in: is the i-th common quantitative band frequency in production, is a control cycle, m is the number of commonly selected quantitative band frequencies, ， in: For quantitative belt travel, It is the ratio coefficient of quantitative belt stroke to control cycle or quantitative belt frequency; The quantitative belt frequency is PID adjusted. According to the adjusted quantitative belt frequency, all the dry ice tobacco on the quantitative belt is output into the expansion process within one control cycle, and the weight of the expanded tobacco before rehumidification is controlled to be maintained at the target value.

7. An adaptive control device for the weight of expanded tobacco before moisture regeneration, characterized in that: Acquisition module: used to obtain data on factors affecting the weight of expanded tobacco during the expansion process; An input module is used to input the data of factors affecting the weight of expanded tobacco into a pre-trained tobacco weight estimation model to obtain a predicted value of the weight of expanded tobacco before rehumidification; An adaptive module is used to set an adaptive law to approximate the weight conversion ratio of dry ice tobacco after the expansion process based on the predicted weight of expanded tobacco before rehydration, so as to obtain the predicted weight of dry ice tobacco; The adjustment module is used to perform PID adjustment on the data of factors affecting the weight of expanded tobacco in the expansion process according to the difference between the set value of the dry ice tobacco weight and the predicted value of the dry ice tobacco weight, and then adjust the data of factors affecting the weight of expanded tobacco in the expansion process, and control the weight of dry ice tobacco entering the expansion process, so that the weight of expanded tobacco before rehumidification is maintained at the target value.

8. An adaptive control system for the weight of expanded tobacco before moisture reconditioning, characterized in that: include: Memory, for storing computer programs / instructions; A processor is used to execute the computer program / instructions to implement the steps of the method for adaptively controlling the weight of expanded tobacco before moisture reconditioning according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the method for adaptively controlling the weight of expanded tobacco before moisture regeneration as described in any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that The computer program / instructions, when executed by a processor, implement the steps of the method for adaptively controlling the weight of expanded tobacco before moisture reconditioning as described in any one of claims 1 to 6.

Images