Tobacco loosening water adding flow control method, device, equipment and medium

Abstract

The invention discloses a tobacco loosening water adding flow control method and device, equipment and a medium, and relates to the technical field of computers, historical production data of tobacco primary processing is preprocessed, and a brand sample data set is obtained; dividing the trademark sample data set to obtain a training set and a test set, performing principal component analysis on the training set, and extracting main feature values; constructing each loose water adding flow prediction model corresponding to each mark based on a multiple regression model, inputting the main characteristic value and the test set into the loose water adding flow prediction model, outputting a training prediction value and a test prediction value, and performing error verification on the training prediction value and the test prediction value; and if the verification is passed, inputting the characteristic parameters of the to-be-predicted tobacco into the loose water-adding flow prediction model of the corresponding mark to obtain each loose water-adding flow prediction value, and performing loose water-adding flow control on the to-be-predicted tobacco by using each loose water-adding flow prediction value, thereby solving the problem of water-adding flow control hysteresis in the loose process of tobacco primary processing. The tobacco shred process quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and medium for controlling the flow rate of loose tobacco water addition. Background Art

[0002] The water flow control in the loosening process is a key control link in the production of tobacco shreds. Its accuracy and stability will directly affect the moisture control effect at the inlet of the drying shreds, and thus affect the process quality of the finished tobacco shreds. At present, the control method of the water flow of the loosening and moistening machine in the industry is basically to achieve manual adjustment based on the moisture feedback at the inlet of the drying shreds, which has serious lag, thus affecting the quality of batch shred products.

[0003] As can be seen from the above, how to solve the hysteresis problem of water flow control in the tobacco shred loosening process and improve the process quality of tobacco shreds is a problem to be solved in this field. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a method, device, equipment and medium for controlling the flow rate of water added to loose tobacco, which can solve the hysteresis problem of water addition flow rate control in the loosening process of tobacco shreds and improve the process quality of shredded tobacco. The specific scheme is as follows:

[0005] In a first aspect, the present application discloses a method for controlling the flow rate of loose tobacco water addition, comprising:

[0006] Acquire historical production data of tobacco silk making, and pre-process the historical production data to obtain a brand sample data set; the pre-processing includes data screening, abnormal data processing and standardization processing;

[0007] Dividing the brand sample data set to obtain a training set and a test set, and performing principal component analysis on the training set to extract main eigenvalues;

[0008] Based on the multivariate regression model, each loose water addition flow prediction model corresponding to each brand is constructed, and the main characteristic value and the test set are respectively input into the loose water addition flow prediction model to output the corresponding training prediction value and test prediction value, and the training prediction value and the test prediction value are error checked;

[0009] If the error check passes, the characteristic parameters of the tobacco to be predicted are input into the loose water addition flow prediction model of the corresponding brand to obtain each loose water addition flow prediction value, and the loose water addition flow prediction value is used to control the loose water addition flow of the tobacco to be predicted.

[0010] Optionally, the obtaining of historical production data of tobacco shreds includes:

[0011] The particle size of tobacco is collected according to production batches to obtain historical production data of tobacco silk making; the historical production data includes production batch number, production brand, batch mean of loosening process data, batch mean of feeding process data, batch mean of leaf storage process data, batch mean of moisture at the drying silk inlet and batch mean of environmental data.

[0012] Optionally, the preprocessing of the historical production data to obtain a brand sample data set includes:

[0013] generating a brand basic data set based on the historical production data;

[0014] The brand basic data set is screened, and abnormal data processing is performed on the screened brand basic data set based on the abnormal data determination formula of the Laida criterion to obtain the brand basic data set after abnormal data processing;

[0015] The brand basic data set after abnormal data processing is standardized by using a deviation standardization method to obtain a brand sample data set.

[0016] Optionally, the brand sample data set is divided to obtain a training set and a test set, and principal component analysis is performed on the training set to extract main eigenvalues, including:

[0017] Randomly extracting and dividing the brand sample data set according to a preset data set division ratio to obtain a training set and a test set;

[0018] The principal component analysis method is used to reduce the dimension of the training set to extract the main eigenvalues.

[0019] Optionally, the inputting the main characteristic value and the test set into the loose water addition flow prediction model respectively includes:

[0020] Input each characteristic parameter in the main characteristic value into the loose water addition flow prediction model, and input each characteristic parameter in the test set into the loose water addition flow prediction model; the characteristic parameters include loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade filament segment ambient temperature, blade filament segment ambient humidity and filament drying inlet moisture.

[0021] Optionally, the performing error checking on the training prediction value and the test prediction value includes:

[0022] Acquire the training actual value corresponding to the training set and the test actual value corresponding to the test set;

[0023] The error between the training prediction value and the training actual value is calculated, and the error between the test prediction value and the test actual value is calculated to complete error checking.

[0024] Optionally, the method for controlling the flow rate of loose tobacco water addition further comprises:

[0025] Acquire new production data of tobacco shreds, and combine the new production data with the historical production data to obtain production data for updating the loose water addition flow prediction model;

[0026] A process of preprocessing the generated data is executed so as to update each loose water addition flow prediction model in real time to obtain an updated loose water addition flow prediction model.

[0027] In a second aspect, the present application discloses a tobacco loose water adding flow control device, comprising:

[0028] A preprocessing module, used to obtain historical production data of tobacco silk making, and preprocess the historical production data to obtain a brand sample data set; the preprocessing includes data screening, abnormal data processing and standardization processing;

[0029] A principal component analysis module is used to divide the brand sample data set to obtain a training set and a test set, and perform principal component analysis on the training set to extract main eigenvalues;

[0030] A prediction module, for constructing loose water addition flow prediction models corresponding to each brand based on a multivariate regression model, inputting the main characteristic values ​​and the test set into the loose water addition flow prediction models respectively, so as to output corresponding training prediction values ​​and test prediction values, and performing error checking on the training prediction values ​​and the test prediction values;

[0031] The loose water addition flow control module is used to input the characteristic parameters of the tobacco to be predicted into the loose water addition flow prediction model of the corresponding brand if the error check passes, so as to obtain each loose water addition flow prediction value, and use each loose water addition flow prediction value to control the loose water addition flow of the tobacco to be predicted. In a third aspect, the present application discloses an electronic device, comprising:

[0032] Memory, used to store computer programs;

[0033] The processor is used to execute the computer program to implement the aforementioned tobacco loosening and water adding flow control method.

[0034] In a fourth aspect, the present application discloses a computer storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the steps of the aforementioned tobacco loosening water addition flow control method are implemented.

[0035] It can be seen that the present application provides a method for controlling the loose water addition flow rate of tobacco, including obtaining historical production data of tobacco silk making, preprocessing the historical production data to obtain a brand sample data set; the preprocessing includes data screening, abnormal data processing and standardization processing; dividing the brand sample data set to obtain a training set and a test set, performing principal component analysis on the training set to extract main eigenvalues; constructing loose water addition flow prediction models corresponding to each brand based on a multivariate regression model, inputting the main eigenvalues ​​and the test set into the loose water addition flow prediction model respectively to output corresponding training prediction values ​​and test prediction values, and performing error checking on the training prediction values ​​and the test prediction values; if the error check passes, inputting the characteristic parameters of the tobacco to be predicted into the loose water addition flow prediction model of the corresponding brand to obtain each loose water addition flow prediction value, and using each loose water addition flow prediction value to control the loose water addition flow rate of the tobacco to be predicted. The present application pre-processes historical production data to obtain a brand sample data set, divides the brand sample data set to obtain a training set and a test set, and constructs loose water addition flow prediction models corresponding to each brand based on a multivariate regression model. The main characteristic values ​​extracted from the training set and the test set are input into the loose water addition flow prediction model to obtain training prediction values ​​and test prediction values. The training prediction values ​​and the test prediction values ​​are error-checked. If the error check passes, the characteristic parameters of the tobacco to be predicted are input to obtain each loose water addition flow prediction value, thereby realizing the loose water addition flow control of the tobacco to be predicted. The present application constructs a loose water addition flow prediction model and combines the prediction model to solve the lag problem of water addition flow control in the loose process of tobacco shredding, thereby improving the process quality of shredded tobacco. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0037] Figure 1 This is a flow chart of a method for controlling the flow rate of loose water addition to tobacco disclosed in this application;

[0038] Figure 2 This is a schematic structural diagram of a tobacco loose water adding flow control device disclosed in this application;

[0039] Figure 3 A structural diagram of an electronic device provided for this application. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] The water flow control in the loosening process is a key control link in the production of tobacco shreds. Its accuracy and stability will directly affect the moisture control effect at the drying inlet, and thus affect the process quality of the finished tobacco shreds. At present, the control method of the water flow of the loosening and rehumidification machine in the industry is basically to achieve manual adjustment based on the moisture feedback at the drying inlet, which has serious lags, thus affecting the quality of batch shred products. As can be seen from the above, how to solve the lag problem of water flow control in the loosening process of tobacco shreds and improve the process quality of tobacco shreds is a problem to be solved in this field.

[0042] See also Figure 1 As shown, the embodiment of the present invention discloses a method for controlling the flow rate of loose tobacco water addition, which may specifically include:

[0043] Step S11: Acquire historical production data of tobacco shreds, and pre-process the historical production data to obtain a brand sample data set; the pre-processing includes data screening, abnormal data processing, and standardization processing.

[0044] In this embodiment, the particle size of tobacco is collected according to the production batch to obtain the historical production data of tobacco silk making; the historical production data includes the production batch number, the production brand, the batch mean of the loosening process data, the batch mean of the feeding process data, the batch mean of the leaf storage process data, the batch mean of the moisture at the drying inlet and the batch mean of the environmental data; then a brand basic data set is generated based on the historical production data; the brand basic data set is screened, and the screened brand basic data set is subjected to abnormal data processing based on the abnormal data judgment formula of the Laida criterion to obtain the brand basic data set after abnormal data processing; the brand basic data set after abnormal data processing is standardized by using the deviation standardization method to obtain a brand sample data set; the preprocessing includes data screening, abnormal data processing and standardization.

[0045] Taking the acquisition of historical production data related to a time span of one year as an example, the historical production data includes: production batch number, production brand, batch mean of loose process data (including: batch mean of loose inlet moisture, batch mean of loose water addition flow, batch mean of loose steam flow, batch mean of loose moisture exhaust air volume), batch mean of feeding process data (including: batch mean of feeding liquid temperature, batch mean of feeding steam flow, batch mean of feeding moisture exhaust air volume, batch mean of feeding cylinder temperature), batch mean of storage blade process data (including: The batch mean of temperature between leaf storage, the batch mean of humidity between leaf storage), the batch mean of moisture at the drying silk inlet, the batch mean of environmental data (including: the batch mean of leaf segment temperature, the batch mean of leaf segment humidity, the batch mean of leaf silk segment temperature, the batch mean of leaf silk segment humidity) and other characteristic parameters; then, the brand basic data set is generated based on the historical production data, and the brand basic data set is preprocessed to obtain a sample data set; the preprocessing includes data screening, abnormal data processing, and standardization processing in sequence, and the brand basic data set after preprocessing is used as the brand sample data set.

[0046] The specific preprocessing process is as follows: data screening, the brand basic data set is screened out according to the batch granularity to obtain different brand basic data sets, such as: Brand A {[batch number, batch process data 1, 2, 3...],...}, Brand B {[batch number, batch process data 1, 2, 3...],...}; abnormal data processing, based on the abnormal data judgment formula of the Laida criterion, it is determined whether the data in the filtered brand basic data set is abnormal data, and the batch data set containing abnormal data is eliminated as a whole. The abnormal data judgment formula is:

[0047] ;

[0048] Among them, x is a characteristic parameter of a certain type of production data, is the average value of the characteristic parameters of this type of production data, is the standard deviation of the characteristic parameters of this type of production data;

[0049] Standardization processing, using the deviation standardization method to standardize the brand basic data set after abnormal data processing, the formula is:

[0050] ;

[0051] in, It is the standardized result of a characteristic parameter of a sample data;

[0052] After completing data screening, abnormal data processing and standardization, the initial brand sample data set is obtained. Then, according to different brands, the characteristic parameters with high correlation with loose water addition flow are selected from the characteristic parameters in the initial brand sample data set; Pearson correlation analysis is often used to evaluate the linear correlation between each characteristic parameter and the loose water addition flow. and samples , its correlation coefficient The calculation formula is as follows:

[0053] ;

[0054] By calculation, delete The initial brand sample data set is further screened based on the characteristic parameters of the loose water flow, loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade shred ambient temperature, blade shred ambient humidity, and shred drying inlet moisture. Finally, the brand sample data set is obtained. The brand sample data set format is: Brand A {[Batch number, loose water flow, loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade shred ambient temperature, blade shred ambient humidity, shred drying inlet moisture],…}, Brand B {[Batch number, loose water flow, loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade shred ambient temperature, blade shred ambient humidity, shred drying inlet moisture],…}.

[0055] Step S12: Divide the brand sample data set to obtain a training set and a test set, and perform principal component analysis on the training set to extract main eigenvalues.

[0056] In this embodiment, the brand sample data set is randomly selected and divided according to a preset data set division ratio to obtain a training set and a test set; the principal component analysis method is used to reduce the data dimension of the training set to extract the main eigenvalues.

[0057] For example, randomly select and divide the data into training set and test set according to the ratio of 80% and 20%, perform data dimension reduction on the training set based on principal component analysis, and extract the main eigenvalues ​​of the sample data. Usually the cumulative contribution rate is greater than 80%, and it can be considered that the m principal components can represent all the features in sample n. In order to extract the accuracy of all the features of the principal components, the cumulative contribution rate is set to 90%. Here, 4 principal components are extracted. The expressions of the principal components are: .

[0058] Step S13: construct loose water addition flow prediction models corresponding to each brand based on the multivariate regression model, input the main characteristic values ​​and the test set into the loose water addition flow prediction models respectively to output corresponding training prediction values ​​and test prediction values, and perform error checking on the training prediction values ​​and test prediction values.

[0059] In this embodiment, loose water addition flow prediction models corresponding to each brand are constructed based on the multivariate regression model, and each characteristic parameter in the main characteristic value is input into the loose water addition flow prediction model, and each characteristic parameter in the test set is input into the loose water addition flow prediction model to output corresponding training prediction values ​​and test prediction values, obtain training actual values ​​corresponding to the training set, and test actual values ​​corresponding to the test set, calculate the error between the training prediction value and the training actual value, and calculate the error between the test prediction value and the test actual value to complete error verification; the characteristic parameters include loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade silk segment ambient temperature, blade silk segment ambient humidity and wire drying inlet moisture.

[0060] Specifically, a prediction model for the loose water addition flow rate of each brand is constructed based on the multivariate regression model. The multivariate regression formula is:

[0061] ;

[0062] The output is the loose water addition flow rate, the input is the main eigenvalue, and the model construction results are shown in Table 1:

[0063] Table 1

[0064]

[0065] All are close to 1, and the model fitting effect is good. VIF (variance inflation factor) is all less than 10, so the model has no multicollinearity problem and the model is well constructed. The prediction model expression of each brand is:

[0066] ;

[0067] According to the main characteristic values ​​and the prediction model expression, the model conversion is carried out, and the prediction model represented by the sample data of each brand is obtained as follows:

[0068] ;

[0069] Among them, Y is the loose water flow rate, are the characteristic parameters in the sample data, including loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade shred segment ambient temperature, blade shred segment ambient humidity, and shred drying inlet moisture.

[0070] Step S14: If the error check passes, the characteristic parameters of the tobacco to be predicted are input into the loose water addition flow prediction model of the corresponding brand to obtain each loose water addition flow prediction value, and the loose water addition flow prediction value is used to control the loose water addition flow of the tobacco to be predicted.

[0071] This application evaluates the performance of the model through error verification, and uses the mean absolute error as a verification index, which measures the difference between the model prediction value and the actual value. The results are all less than the industry standard specific threshold (usually set to 0.1), the error verification is passed, the model verification is passed, and the loose water flow prediction model is directly used. The characteristic parameters of the tobacco to be predicted are substituted into the loose water flow prediction model of the corresponding brand, and the loose water flow prediction value of the brand can be obtained; wherein, the characteristic parameters of the tobacco to be predicted are: loose inlet moisture, loose steam flow, leaf segment ambient temperature, leaf segment ambient humidity, leaf shred segment ambient temperature, leaf shred segment ambient humidity, and drying shred inlet moisture; wherein, the drying shred inlet moisture is the production process data of the subsequent process of the batch, so it can be identified as the expected drying shred inlet moisture value, which can be achieved by setting the expected drying shred inlet moisture value and substituting the remaining characteristic parameters to obtain the loose water flow prediction value.

[0072] In addition, the tobacco loose water flow control method proposed in the present application also includes: obtaining new production data of tobacco shreds, and combining the new production data with the historical production data to obtain production data for updating the loose water flow prediction model; executing a process of preprocessing the generated data so as to update each loose water flow prediction model in real time to obtain an updated loose water flow prediction model. The present application can also realize an automatic learning function for the loose water flow prediction model.

[0073] The loose water addition flow control method of tobacco proposed in this application is applied to the loose water addition flow control system, which includes a production process data acquisition module, a sample database module, a data preprocessing module, a model publishing and operation module, and a system interaction module. The specific functions are: the production process data acquisition module, the collected data includes the production brand, the expected moisture value of the drying silk inlet, and the characteristic parameters of the brand and the batch, the collection frequency is once every 1 second, and the collection cycle is once every 3 minutes; the sample database module is implemented using the SQL server database system, and the SQL (Structured Query Language, Structured Query Language) script realizes the interaction of sample data; the data preprocessing module preprocesses the periodically collected data, removes abnormal data, and calculates the mean of each process parameter as the input parameter of the corresponding brand prediction model; the model publishing and operation module converts the mathematical model into an algorithm program, embeds the device PLC (programmable logic controller) control program, inputs the parameters after data preprocessing, and outputs the loose water addition flow prediction value. And the predicted value is input into the loose control program as the water addition flow setting value; the system interaction module includes an input box for the expected moisture value of the drying wire inlet, an output box for the loose water addition flow prediction, a loose water addition flow prediction value curve, etc.

[0074] In the present embodiment, historical production data of tobacco silk making is obtained, and the historical production data is preprocessed to obtain a brand sample data set; the preprocessing includes data screening, abnormal data processing and standardization processing; the brand sample data set is divided to obtain a training set and a test set, and the training set is subjected to principal component analysis to extract main eigenvalues; loose water addition flow prediction models corresponding to each brand are constructed based on a multivariate regression model, and the main eigenvalues ​​and the test set are respectively input into the loose water addition flow prediction model to output corresponding training prediction values ​​and test prediction values, and the training prediction values ​​and the test prediction values ​​are error checked; if the error check passes, the characteristic parameters of the tobacco to be predicted are input into the loose water addition flow prediction model of the corresponding brand to obtain each loose water addition flow prediction value, and the loose water addition flow prediction value is used to control the loose water addition flow of the tobacco to be predicted. The present application pre-processes historical production data to obtain a brand sample data set, divides the brand sample data set to obtain a training set and a test set, and constructs loose water addition flow prediction models corresponding to each brand based on a multivariate regression model. The main characteristic values ​​extracted from the training set and the test set are input into the loose water addition flow prediction model to obtain training prediction values ​​and test prediction values. The training prediction values ​​and the test prediction values ​​are error-checked. If the error check passes, the characteristic parameters of the tobacco to be predicted are input to obtain each loose water addition flow prediction value, thereby realizing the loose water addition flow control of the tobacco to be predicted. The present application constructs a loose water addition flow prediction model and combines the prediction model to solve the lag problem of water addition flow control in the loose process of tobacco shredding, thereby improving the process quality of shredded tobacco.

[0075] See also Figure 2 As shown, the embodiment of the present invention discloses a flow control device for loosening tobacco and adding water, which may specifically include:

[0076] The preprocessing module 11 is used to obtain historical production data of tobacco shreds and preprocess the historical production data to obtain a brand sample data set; the preprocessing includes data screening, abnormal data processing and standardization processing;

[0077] A principal component analysis module 12 is used to divide the brand sample data set to obtain a training set and a test set, and perform principal component analysis on the training set to extract main eigenvalues;

[0078] A prediction module 13 is used to construct loose water addition flow prediction models corresponding to each brand based on a multivariate regression model, input the main characteristic value and the test set into the loose water addition flow prediction model respectively, output corresponding training prediction values ​​and test prediction values, and perform error checking on the training prediction values ​​and the test prediction values;

[0079] The loose water addition flow control module 14 is used to input the characteristic parameters of the tobacco to be predicted into the loose water addition flow prediction model of the corresponding brand if the error check is passed, so as to obtain each loose water addition flow prediction value, and use each loose water addition flow prediction value to control the loose water addition flow of the tobacco to be predicted.

[0080] In the present embodiment, historical production data of tobacco silk making is obtained, and the historical production data is preprocessed to obtain a brand sample data set; the preprocessing includes data screening, abnormal data processing and standardization processing; the brand sample data set is divided to obtain a training set and a test set, and the training set is subjected to principal component analysis to extract main eigenvalues; loose water addition flow prediction models corresponding to each brand are constructed based on a multivariate regression model, and the main eigenvalues ​​and the test set are respectively input into the loose water addition flow prediction model to output corresponding training prediction values ​​and test prediction values, and the training prediction values ​​and the test prediction values ​​are error checked; if the error check passes, the characteristic parameters of the tobacco to be predicted are input into the loose water addition flow prediction model of the corresponding brand to obtain each loose water addition flow prediction value, and the loose water addition flow prediction value is used to control the loose water addition flow of the tobacco to be predicted. The present application pre-processes historical production data to obtain a brand sample data set, divides the brand sample data set to obtain a training set and a test set, and constructs loose water addition flow prediction models corresponding to each brand based on a multivariate regression model. The main characteristic values ​​extracted from the training set and the test set are input into the loose water addition flow prediction model to obtain training prediction values ​​and test prediction values. The training prediction values ​​and the test prediction values ​​are error-checked. If the error check passes, the characteristic parameters of the tobacco to be predicted are input to obtain each loose water addition flow prediction value, thereby realizing the loose water addition flow control of the tobacco to be predicted. The present application constructs a loose water addition flow prediction model and combines the prediction model to solve the lag problem of water addition flow control in the loose process of tobacco shredding, thereby improving the process quality of shredded tobacco.

[0081] In some specific embodiments, the preprocessing module 11 may specifically include:

[0082] The historical production data acquisition module is used to collect the particle size of tobacco according to production batches to obtain the historical production data of tobacco silk making; the historical production data includes production batch number, production brand, batch mean of loosening process data, batch mean of feeding process data, batch mean of leaf storage process data, batch mean of moisture at the drying silk inlet and batch mean of environmental data.

[0083] In some specific embodiments, the preprocessing module 11 may specifically include:

[0084] A basic data set generation module, used to generate a brand basic data set based on the historical production data;

[0085] A screening and abnormal data processing module, used for screening the brand basic data set, and performing abnormal data processing on the screened brand basic data set based on the abnormal data determination formula of the Laida criterion, so as to obtain the brand basic data set after abnormal data processing;

[0086] The standardization processing module is used to standardize the brand basic data set after abnormal data processing by using a deviation standardization method to obtain a brand sample data set.

[0087] In some specific embodiments, the principal component analysis module 12 may specifically include:

[0088] A random sampling and partitioning module is used to randomly sample and partition the brand sample data set according to a preset data set partitioning ratio to obtain a training set and a test set;

[0089] The data dimension reduction module is used to perform data dimension reduction on the training set using principal component analysis to extract main eigenvalues.

[0090] In some specific embodiments, the prediction module 13 may specifically include:

[0091] A characteristic parameter input module is used to input each characteristic parameter in the main characteristic value into the loose water addition flow prediction model, and input each characteristic parameter in the test set into the loose water addition flow prediction model; the characteristic parameters include loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade wire segment ambient temperature, blade wire segment ambient humidity and wire drying inlet moisture.

[0092] In some specific embodiments, the prediction module 13 may specifically include:

[0093] An actual value acquisition module, used to acquire training actual values ​​corresponding to the training set and test actual values ​​corresponding to the test set;

[0094] The error calculation module is used to calculate the error between the training prediction value and the training actual value, and calculate the error between the test prediction value and the test actual value to complete error checking.

[0095] In some specific embodiments, the tobacco loose water adding flow control device may further include:

[0096] A new production data acquisition module, used to acquire new production data of tobacco shreds, and combine the new production data with the historical production data to obtain production data for updating the loose water addition flow prediction model;

[0097] The prediction model updating module is used to execute the process of preprocessing the generated data so as to update each loose water addition flow prediction model in real time to obtain an updated loose water addition flow prediction model.

[0098] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is used to store a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the tobacco loosening water flow control method performed by the electronic device disclosed in any of the aforementioned embodiments.

[0099] In this embodiment, the power supply 23 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and the external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0100] In addition, the memory 22, as a carrier for storing resources, can be a read-only memory, a random access memory, a disk or an optical disk, etc. The resources stored thereon include an operating system 221, a computer program 222 and data 223, etc. The storage method can be temporary storage or permanent storage.

[0101] The operating system 221 is used to manage and control the hardware devices and computer programs 222 on the electronic device 20, so as to realize the operation and processing of the data 223 in the memory 22 by the processor 21, which can be Windows, Unix, Linux, etc. In addition to including a computer program that can be used to complete the tobacco loose water flow control method performed by the electronic device 20 disclosed in any of the aforementioned embodiments, the computer program 222 can further include a computer program that can be used to complete other specific tasks. In addition to including data transmitted from an external device received by the tobacco loose water flow control device, the data 223 can also include data collected by its own input and output interface 25, etc.

[0102] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0103] Furthermore, an embodiment of the present application also discloses a computer-readable storage medium, in which a computer program is stored. When the computer program is loaded and executed by a processor, the steps of the tobacco loosening water addition flow control method disclosed in any of the aforementioned embodiments are implemented.

[0104] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0105] The above is a detailed introduction to a method, device, equipment and storage medium for controlling the flow rate of loose tobacco water addition provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for controlling the flow rate of loose tobacco water addition, characterized in that: include: Acquire historical production data of tobacco silk making, and pre-process the historical production data to obtain a brand sample data set; The preprocessing includes data screening, abnormal data processing and standardization processing; Dividing the brand sample data set to obtain a training set and a test set, and performing principal component analysis on the training set to extract main eigenvalues; Based on the multivariate regression model, each loose water addition flow prediction model corresponding to each brand is constructed, and the main characteristic value and the test set are respectively input into the loose water addition flow prediction model to output the corresponding training prediction value and test prediction value, and the training prediction value and the test prediction value are error checked; If the error check passes, the characteristic parameters of the tobacco to be predicted are input into the loose water addition flow prediction model of the corresponding brand to obtain each loose water addition flow prediction value, and the loose water addition flow prediction value is used to control the loose water addition flow of the tobacco to be predicted.

2. The method for controlling the flow rate of loose tobacco water addition according to claim 1, characterized in that: The acquisition of historical production data of tobacco silk production includes: The particle size of tobacco is collected according to production batches to obtain historical production data of tobacco silk making; the historical production data includes production batch number, production brand, batch mean of loosening process data, batch mean of feeding process data, batch mean of leaf storage process data, batch mean of moisture at the drying silk inlet and batch mean of environmental data.

3. The method for controlling the flow rate of loose tobacco water addition according to claim 1, characterized in that: The preprocessing of the historical production data to obtain a brand sample data set includes: generating a brand basic data set based on the historical production data; The brand basic data set is screened, and abnormal data processing is performed on the screened brand basic data set based on the abnormal data determination formula of the Laida criterion to obtain the brand basic data set after abnormal data processing; The brand basic data set after abnormal data processing is standardized by using a deviation standardization method to obtain a brand sample data set.

4. The method for controlling the flow rate of loose tobacco water addition according to claim 1, characterized in that: The step of dividing the brand sample data set to obtain a training set and a test set, and performing principal component analysis on the training set to extract main eigenvalues ​​includes: Randomly extracting and dividing the brand sample data set according to a preset data set division ratio to obtain a training set and a test set; The principal component analysis method is used to reduce the dimension of the training set to extract the main eigenvalues.

5. The method for controlling the flow rate of loose tobacco water addition according to claim 1, characterized in that: The inputting the main characteristic value and the test set into the loose water addition flow prediction model respectively includes: Input each characteristic parameter in the main characteristic value into the loose water addition flow prediction model, and input each characteristic parameter in the test set into the loose water addition flow prediction model; the characteristic parameters include loose inlet moisture, loose steam flow, blade segment ambient temperature, blade segment ambient humidity, blade filament segment ambient temperature, blade filament segment ambient humidity and filament drying inlet moisture.

6. The method for controlling the flow rate of loose tobacco water addition according to claim 1, characterized in that: The error checking of the training prediction value and the test prediction value comprises: Acquire the training actual value corresponding to the training set and the test actual value corresponding to the test set; The error between the training prediction value and the training actual value is calculated, and the error between the test prediction value and the test actual value is calculated to complete error checking.

7. The method for controlling the flow rate of loose tobacco water addition according to claim 1, characterized in that: Also includes: Acquire new production data of tobacco shreds, and combine the new production data with the historical production data to obtain production data for updating the loose water addition flow prediction model; A process of preprocessing the generated data is executed so as to update each loose water addition flow prediction model in real time to obtain an updated loose water addition flow prediction model.

8. A flow control device for loosening tobacco and adding water, characterized in that: include: A preprocessing module, used to obtain historical production data of tobacco silk making, and preprocess the historical production data to obtain a brand sample data set; The preprocessing includes data screening, abnormal data processing and standardization processing; A principal component analysis module is used to divide the brand sample data set to obtain a training set and a test set, and perform principal component analysis on the training set to extract main eigenvalues; A prediction module, for constructing loose water addition flow prediction models corresponding to each brand based on a multivariate regression model, inputting the main characteristic values ​​and the test set into the loose water addition flow prediction models respectively, so as to output corresponding training prediction values ​​and test prediction values, and performing error checking on the training prediction values ​​and the test prediction values; The loose water addition flow control module is used to input the characteristic parameters of the tobacco to be predicted into the loose water addition flow prediction model of the corresponding brand if the error check is passed, so as to obtain each loose water addition flow prediction value, and use each loose water addition flow prediction value to control the loose water addition flow of the tobacco to be predicted.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor is used to execute the computer program to implement the tobacco loosening and watering flow control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store computer programs; wherein, when the computer program is executed by a processor, the tobacco loosening and watering flow control method according to any one of claims 1 to 7 is implemented.

Images