CN115251445B - Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine - Google Patents
Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine Download PDFInfo
- Publication number
- CN115251445B CN115251445B CN202210973084.1A CN202210973084A CN115251445B CN 115251445 B CN115251445 B CN 115251445B CN 202210973084 A CN202210973084 A CN 202210973084A CN 115251445 B CN115251445 B CN 115251445B
- Authority
- CN
- China
- Prior art keywords
- outlet
- conditioning
- machine
- tobacco
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A24—TOBACCO; CIGARS; CIGARETTES; SIMULATED SMOKING DEVICES; SMOKERS' REQUISITES
- A24B—MANUFACTURE OR PREPARATION OF TOBACCO FOR SMOKING OR CHEWING; TOBACCO; SNUFF
- A24B3/00—Preparing tobacco in the factory
- A24B3/04—Humidifying or drying tobacco bunches or cut tobacco
-
- A—HUMAN NECESSITIES
- A24—TOBACCO; CIGARS; CIGARETTES; SIMULATED SMOKING DEVICES; SMOKERS' REQUISITES
- A24B—MANUFACTURE OR PREPARATION OF TOBACCO FOR SMOKING OR CHEWING; TOBACCO; SNUFF
- A24B9/00—Control of the moisture content of tobacco products, e.g. cigars, cigarettes, pipe tobacco
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Manufacture Of Tobacco Products (AREA)
Abstract
A control method for the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine. The invention collects production process data and environment data within one month as historical data; acquiring the latest n=15 batches of data; establishing a linear regression model I to obtain the water content of the leaf storage cabinet; establishing a probability distribution model to obtain a set value of the moisture content of tobacco leaves at the outlet of the damping machine; establishing a linear regression model II, and solving to obtain an initial value of the loose conditioning water adding accumulation; establishing a machine learning prediction model, and solving to obtain a predicted value of the moisture content of tobacco leaves at the outlet of the damping machine; adopting an optimization algorithm to carry out optimization pushing on loosening, conditioning and water adding flow in each tobacco shred output stage; and regulating the loosening and conditioning water adding flow according to the deviation between the moisture content of tobacco leaves at the outlet of the damping machine and the actual value of the moisture content of tobacco leaves at the outlet of the damping machine, which is output by the prediction model. The method has the advantages of improving the prediction accuracy of the prediction model, improving the control effect, reducing the input of manpower and achieving the purposes of improving the quality of the tobacco leaves subjected to moisture regain and reducing the cost.
Description
Technical Field
The invention relates to a control method of the moisture content of tobacco leaves, in particular to a control method of the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine, and belongs to the technical field of tobacco production and processing equipment.
Background
In each link of tobacco production, the moisture control of tobacco leaf conditioning links comprises loosening conditioning and water adding flow, and directly plays an important role in the quality control of tobacco shreds in the subsequent conditioning links. Loose conditioning is the first core process of tobacco processing, and the stability of the moisture content of tobacco leaves at a loose conditioning outlet has direct influence on the process indexes of the subsequent processes. Due to the influences of factors such as physical property change of tobacco flakes, incoming material flow fluctuation, measurement hysteresis, environmental temperature and humidity and the like, fluctuation exists in the water content of tobacco leaves at the outlet of the loosening and conditioning process in actual production, larger deviation is generated between the water content of tobacco leaves at the conditioning outlet and a standard value, the control effect is poor, the water content of the tobacco leaves is unstable, manual intervention is needed, and the production cost is increased.
Disclosure of Invention
In order to overcome the defects in the prior loosening and conditioning process of tobacco leaf processing, the invention provides a control method for the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine.
The technical scheme adopted for solving the technical problems is as follows: a control method of the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine comprises the following steps:
s1, acquiring production process data and environment data within one month, preprocessing the data and storing the data into a database as historical data.
The method for cleaning and preprocessing the historical data comprises the steps of eliminating null values and eliminating abnormal values.
And S2, acquiring the latest N=15 batches of data, and calculating the average value of each batch of sample data.
And S3, establishing a linear regression model I of the water content of the leaf storage cabinet outlet by adopting a linear regression method, and solving to obtain the water content of the leaf storage cabinet outlet.
S4, taking the water content of the tobacco leaves at the outlet of the tobacco storage cabinet obtained by solving the water content difference and the linear regression model I as a variable, establishing a probability distribution model taking the set value of the water content of tobacco leaves at the outlet of the moisture regain machine as a target value, and solving to obtain the set value of the water content of tobacco leaves at the outlet of the moisture regain machine.
The water content difference is the difference value between the water content of the tobacco leaves in the tobacco leaves storage cabinet and the set value of the water content of tobacco leaves at the outlet of the damping machine in the historical data.
S5, using the probability distribution model to solve the obtained set value of the moisture content of tobacco leaves at the outlet of the damping machine, the ambient temperature of the damping machine, the ambient humidity of the damping machine and the cumulative amount of the slicing damping electronic scale as variables, adopting a linear regression method to establish a linear regression model II which takes the initial value of the cumulative amount of loose damping water as a target value, and solving to obtain the initial value of the cumulative amount of loose damping water.
S6, taking the moisture content of tobacco leaves at the outlet of the damping machine as a target value, selecting loose damping water adding flow as a controllable variable in an input value, adopting a correlation analysis method to select the amount of a correlation coefficient r more than or equal to 0.5 as a correlation variable,
s7, taking the correlation variable in the step S6 as input, taking the moisture content of tobacco leaves at the outlet of the target value damping machine as output, and training by adopting a machine learning algorithm to obtain a machine learning prediction model.
And solving and obtaining a predicted value of the moisture regain machine outlet tobacco leaf moisture content based on a machine learning-established moisture regain machine outlet tobacco leaf moisture content prediction model.
S8, taking the set value of the moisture content of tobacco leaves at the outlet of the damping machine, which is obtained according to the probability distribution model in the step S4, as a best-pushing target value of a prediction model, and adopting an optimization algorithm to push loose damping and water flow of each tobacco leaf output stage.
S9, converting the initial value of the accumulated water quantity of the loose conditioning into the initial value of the water quantity of the loose conditioning according to the initial value of the accumulated water quantity of the loose conditioning obtained by the linear regression model II in the step S5 through the formula (1) in the step S6, and taking the initial value of the water quantity of the loose conditioning when the production is started.
Regulating the loosening and conditioning water adding flow according to the deviation between the moisture content of tobacco leaves at the outlet of the damping machine and the actual value of the moisture content of tobacco leaves at the outlet of the damping machine output by the prediction model,
when the absolute value of the deviation is less than 0.3%, the moisture content of the conditioning outlet is kept unchanged;
and when the absolute value of the deviation is more than or equal to 0.3%, solving the optimal water adding flow by adopting the optimization algorithm in the step S8.
Further, in step S1, the production process data includes the moisture content of tobacco shreds at the inlet of the heating and humidifying device, the moisture content of the tobacco leaves at the outlet of the leaf storage cabinet, the moisture regaining machine outlet moisture content set value, the cumulative amount of the slicing moisture regaining electronic scale, the flow of the loose moisture regaining inlet scale, the actual opening of the moisture regaining water adding film valve, the opening of the heating and humidifying film valve, the flow of the loose moisture regaining machine, the flow of loose moisture regaining steam, the temperature of loose moisture regaining hot air, the temperature of loose moisture regaining fresh air, the pressure of the loose moisture regaining discharge cover, the loose moisture regaining return air temperature, the flow of the feeding machine inlet scale, and the cumulative amount of the slicing moisture regaining electronic scale.
The environmental data comprises the environmental temperature of the cut-tobacco dryer, the environmental humidity of the cut-tobacco dryer, the environmental temperature of the moisture regain machine and the environmental humidity of the moisture regain machine.
Further, in the step S2, the average value of the recent n=15 batches of production process data and environmental data is obtained as fitting data of the linear regression model i, the probability distribution model, and the linear regression model ii.
Further, in the step S3, the expression of the linear regression model I taking the water content of the leaf storage cabinet as the target value is that,
Y 1 ＝2.19486-0.07020X 1 -0.05841X 2 +1.18721X 3 (2)
wherein:
Y 1 -leaf storage cabinet outlet water content
X 1 -the ambient temperature of the cut-tobacco drier
X 2 -environmental humidity of cut-tobacco drier
X 3 -the moisture content of tobacco shreds at the inlet of the heating and humidifying equipment.
Further, in step S5, the expression of the linear regression model II targeting the initial value of the integrated amount of loose conditioning water addition is that,
Y 2 ＝-3395.52679+11.93938X 4 -0.65249X 5 +109.58232X 6 +0.12824X 7 (3)
wherein:
Y 2 initial value of accumulated amount of loosening, conditioning and water adding
X 4 -environmental temperature of the conditioning machine
X 5 -moisture regain machine ambient humidity
X 6 -the set value of the moisture content of tobacco leaves at the outlet of the moisture regain machine
X 7 -slice conditioning electronic scale accumulation.
Further, in step S6, the correlation coefficient r < 0.3 is no correlation, r < 0.3 is low correlation, r < 0.5 is significant correlation, r < 0.8 is high correlation, r < 1.0 is high correlation;
the correlation variable includes: the method comprises the following steps of weighing flow rate of a loosening and conditioning inlet, actual opening of a conditioning water adding film valve, opening of a warming and wetting film valve, flow rate of loosening and conditioning water adding, flow rate of loosening and conditioning steam, temperature of loosening and conditioning hot air, temperature of loosening and conditioning fresh air, pressure of a loosening and conditioning discharging cover, temperature of loosening and conditioning return air, ambient temperature of a conditioning machine, ambient humidity of the conditioning machine, weighing flow rate of an inlet of a charging machine and cumulative amount of a slicing conditioning electronic scale.
And (3) carrying out pairwise analysis on all variables in the data by adopting a correlation analysis method, and selecting a correlation variable with a target value of the moisture content of tobacco leaves at the outlet of the damping machine and a correlation coefficient r more than or equal to 0.5 as a variable of the prediction model in the step S7.
The collected production process data and environmental data are stored in a database. Preferably, the database updates the data at preset time intervals, and updates the data of the tobacco leaf moisture content prediction model at the outlet of the damping machine by using the updated database.
Further, in step S8, the optimal loose conditioning water flow rate is solved using the optimization algorithm.
Further, in the production process of the step S9, the nodes are divided into four stages of initial stage, first stage, second stage and third stage according to the total yield of tobacco shreds of 0%, 25%, 50%, 75% and 100%,
when the total tobacco shred yield is more than or equal to 0 percent and less than 25 percent, solving the obtained initial value of the loose conditioning water adding accumulation by using a linear regression model II as the water adding flow value in the initial stage;
when the total tobacco shred yield is 25 percent or less and the tobacco shred yield is less than or equal to 50 percent of the total tobacco shred yield, taking a set value of the moisture regain machine outlet tobacco shred water content obtained by solving a probability distribution model as a pushing target value, adopting an optimization algorithm to push the prediction model, and solving a loose conditioning water adding flow pushing optimal value at the moment of 25 percent of the total tobacco shred yield as a water adding flow pushing optimal value in the first stage;
when the total tobacco shred yield is 50 percent or less and the tobacco shred yield is less than 75 percent of the total tobacco shred yield, taking a set value of the moisture regain machine outlet tobacco shred water content obtained by solving a probability distribution model as a pushing target value, adopting an optimization algorithm to push the prediction model, and solving a loose conditioning water adding flow pushing optimal value at the moment of 50 percent of the total tobacco shred yield as a water adding flow pushing optimal value of a second stage;
when the total tobacco shred yield is 75 percent or less and is less than 100 percent, taking a set value of the tobacco shred water content at the outlet of the damping machine, which is obtained by solving the probability distribution model, as a best pushing target value, adopting an optimization algorithm to push the prediction model, and solving to obtain a loose damping water adding flow best pushing value at the moment of 75 percent of the total tobacco shred yield, which is used as a water adding flow best pushing value in the third stage.
The method has the advantages that a variable with relatively large influence on the system is selected by adopting a correlation analysis method, and data preprocessing is carried out on historical data; the prediction model is built by building a linear regression model I, a probability distribution model and a linear regression model II and based on a machine learning algorithm, when the deviation between a prediction target value and an actual target value output by the prediction model is more than or equal to 0.3%, the prediction model is optimized by utilizing an optimization algorithm, the optimal loose conditioning water adding flow is solved by utilizing the optimization algorithm and the machine learning prediction model through a parameter optimization scheme, and the water adding flow value is automatically regulated to provide the control precision of the moisture content of tobacco leaves at the outlet of the conditioning machine, so that the prediction accuracy of the prediction model is improved, the control effect is improved, the input of manpower is reduced, and the purposes of improving the quality of the conditioning tobacco leaves and reducing the cost are achieved.
Drawings
FIG. 1 is a flow chart of linear regression and probability distribution model modeling in accordance with the present invention.
FIG. 2 is a flow chart of modeling a machine learning model in accordance with the present invention.
FIG. 3 is a diagram of the production run phase control logic of the present invention.
Detailed Description
The invention is further described below with reference to the drawings and examples. However, it should be understood by those skilled in the art that the present invention is not limited to the specific embodiments listed and should be included within the scope of the present invention as long as the spirit of the present invention is satisfied.
See fig. 1-3. The invention relates to a control method for the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine, which comprises the following steps:
s1, acquiring production process data and environment data within one month, preprocessing the data and storing the data into a database as historical data.
The method for cleaning and preprocessing the historical data comprises the steps of eliminating null values and eliminating abnormal values.
Further, the production process data comprises the moisture content of tobacco shreds at an inlet of the heating and humidifying equipment, the moisture content of a tobacco storage cabinet, a moisture content set value at an outlet of a moisture regaining machine, a cumulative amount of a slicing moisture regaining electronic scale, a flow rate of a loose moisture regaining inlet scale, an actual opening degree of a moisture regaining water adding film valve, an opening degree of the heating and humidifying film valve, a flow rate of the loose moisture regaining machine, a flow rate of loose moisture regaining steam, a temperature of loose moisture regaining hot air, a temperature of loose moisture regaining fresh air, a pressure of a loose moisture regaining discharge cover, a temperature of loose moisture regaining return air, a flow rate of a feeding machine inlet scale and a cumulative amount of the slicing moisture regaining electronic scale.
Further, the environmental data comprises the environmental temperature of the cut-tobacco dryer, the environmental humidity of the cut-tobacco dryer, the environmental temperature of the moisture regain machine and the environmental humidity of the moisture regain machine.
And S2, acquiring the latest N=15 batches of data, and calculating the average value of each batch of sample data.
And S3, establishing a linear regression model I of the water content of the leaf storage cabinet outlet by adopting a linear regression method, and solving to obtain the water content of the leaf storage cabinet outlet.
The expression of the linear regression model I taking the water content of the leaf storage cabinet as the target value is that,
Y 1 ＝2.19486-0.07020X 1 -0.05841X 2 +1.18721X 3 (2)
wherein:
Y 1 -leaf storage cabinet outlet water content
X 1 -the ambient temperature of the cut-tobacco drier
X 2 -environmental humidity of cut-tobacco drier
X 3 The water content of tobacco shreds at the inlet of a heating and humidifying device (HT for short).
S4, taking the water content of the tobacco leaves at the outlet of the tobacco storage cabinet obtained by solving the water content difference and the linear regression model I as a variable, establishing a probability distribution model taking the set value of the water content of tobacco leaves at the outlet of the moisture regain machine as a target value, and solving to obtain the set value of the water content of tobacco leaves at the outlet of the moisture regain machine.
The water content difference is the difference value between the water content of the tobacco leaves in the tobacco leaves storage cabinet and the set value of the water content of tobacco leaves at the outlet of the damping machine in the historical data.
S5, using the probability distribution model to solve the obtained set value of the moisture content of tobacco leaves at the outlet of the damping machine, the ambient temperature of the damping machine, the ambient humidity of the damping machine and the cumulative amount of the slicing damping electronic scale as variables, adopting a linear regression method to establish a linear regression model II which takes the initial value of the cumulative amount of loose damping water as a target value, and solving to obtain the initial value of the cumulative amount of loose damping water.
The expression of the linear regression model II which aims at the initial value of the loose conditioning water adding accumulation amount is as follows,
Y 2 ＝-3395.52679+11.93938X 4 -0.65249X 5 +109.58232X 6 +0.12824X 7 (3)
wherein:
Y 2 initial value of accumulated amount of loosening, conditioning and water adding
X 4 -environmental temperature of the conditioning machine
X 5 -moisture regain machine ambient humidity
X 6 -the set value of the moisture content of tobacco leaves at the outlet of the moisture regain machine
X 7 -slice conditioning electronic scale accumulation.
S6, taking the moisture content of tobacco leaves at the outlet of the damping machine as a target value, selecting loose damping water adding flow as a controllable variable in an input value, adopting a correlation analysis method to select the amount of a correlation coefficient r more than or equal to 0.5 as a correlation variable,
when the correlation coefficient r is less than 0.3, no correlation exists, r is more than or equal to 0.3 and less than 0.5, low correlation is achieved, r is more than or equal to 0.5 and less than or equal to 0.8, obvious correlation is achieved, and r is more than or equal to 0.8 and less than or equal to 1.0, high correlation is achieved.
The correlation variable includes: the loosening and conditioning inlet nominal flow, the actual opening of the conditioning water adding film valve, the opening of the warming and wetting film valve, the loosening and conditioning water adding flow, the loosening and conditioning steam flow, the loosening and conditioning hot air temperature, the loosening and conditioning fresh air temperature, the loosening and conditioning discharge cover pressure, the loosening and conditioning return air temperature, the conditioning machine environment humidity, the feeding machine inlet nominal flow, the slice conditioning electronic scale cumulative amount and the like.
S7, taking the correlation variable in the step S6 as input, taking the moisture content of tobacco leaves at the outlet of the target value damping machine as output, and training by adopting a machine learning algorithm to obtain a machine learning prediction model;
and solving and obtaining a predicted value of the moisture regain machine outlet tobacco leaf moisture content based on a machine learning-established moisture regain machine outlet tobacco leaf moisture content prediction model.
And carrying out pairwise analysis on all variables in the data by adopting a correlation analysis method, and selecting a correlation variable with a target value of the moisture content of tobacco leaves at the outlet of the damping machine and a correlation coefficient r more than or equal to 0.5 as a variable of a prediction model.
S8, taking the set value of the moisture content of tobacco leaves at the outlet of the damping machine, which is obtained according to the probability distribution model in the step S4, as a best-pushing target value of a prediction model, and adopting an optimization algorithm to push loose damping and water flow of each tobacco leaf output stage.
And solving the optimal loose conditioning water adding flow by using the optimization algorithm.
S9, converting the initial value of the accumulated water quantity of the loose conditioning into the initial value of the water quantity of the loose conditioning according to the initial value of the accumulated water quantity of the loose conditioning obtained by the linear regression model II in the step S5 through the formula (1) in the step S6, and taking the initial value of the water quantity of the loose conditioning when the production is started.
And regulating the loosening and conditioning water adding flow according to the deviation between the moisture content of tobacco leaves at the outlet of the damping machine and the actual value of the moisture content of tobacco leaves at the outlet of the damping machine, which is output by the prediction model. When the absolute value of the deviation is less than 0.3%, the moisture content of the conditioning outlet is kept unchanged; and when the absolute value of the deviation is more than or equal to 0.3%, solving the optimal water adding flow by adopting the optimization algorithm in the step S8.
In the production process, the nodes are divided into four stages of initial stage, first stage, second stage and third stage according to the total yield of tobacco shreds of 0%, 25%, 50%, 75% and 100%,
when the total tobacco shred yield is more than or equal to 0 percent and less than 25 percent, solving the obtained initial value of the loose conditioning water adding accumulation by using a linear regression model II as the water adding flow value in the initial stage;
when the total tobacco shred yield is 25 percent or less and the tobacco shred yield is less than or equal to 50 percent of the total tobacco shred yield, taking a set value of the moisture regain machine outlet tobacco shred water content obtained by solving a probability distribution model as a pushing target value, pushing the prediction model by adopting an optimization algorithm, and solving to obtain a loose conditioning water adding flow pushing optimal value at the moment of 25 percent of the total tobacco shred yield, wherein the pushing optimal value is taken as a water adding flow pushing optimal value in the first stage;
when the total tobacco shred yield is 50 percent or less and the tobacco shred yield is less than or equal to 75 percent of the total tobacco shred yield, taking a set value of the moisture regain machine outlet tobacco shred water content obtained by solving a probability distribution model as a pushing target value, pushing the prediction model by adopting an optimization algorithm, and solving to obtain a loose conditioning water adding flow pushing optimal value at the moment of 50 percent of the total tobacco shred yield, wherein the pushing optimal value is taken as a water adding flow pushing optimal value of a second stage;
when the total tobacco shred yield is 75 percent or less and is less than 100 percent, the set value of the tobacco shred water content at the outlet of the damping machine, which is obtained by solving the probability distribution model, is used as a optimal pushing target value, an optimization algorithm is adopted to perform optimal pushing on the prediction model, and a loose damping water adding flow optimal pushing value at the moment when the total tobacco shred yield is 75 percent is obtained, wherein the optimal pushing value is used as a water adding flow optimal pushing value in the third stage.
In the production regulation and modeling process, the loose conditioning water adding flow is more visual and more accurate than the loose conditioning water adding accumulation amount.
Further, in the step S2, the average value of the recent n=15 batches of production process data and environmental data is obtained as fitting data of the linear regression model i, the probability distribution model, and the linear regression model ii.
Storing the collected production process data and environmental data into a database; and the database updates the data of the database at preset time intervals, and updates the data of the tobacco leaf moisture content prediction model at the outlet of the damping machine by utilizing the updated database. Preferably, the preset time interval for updating the data of the database is 200-400 seconds, and more preferably, the preset time interval is 250 seconds, 300 seconds or 350 seconds.
According to the control method of the moisture content of the tobacco leaves at the outlet of the loosening and conditioning machine, a machine learning model is built, and a correlation analysis method is adopted to select variables with relatively large influence on a system; according to the parameter optimization scheme, the moisture regain and water flow of each node is optimized according to the water content of the moisture regain outlet, the controllable variable is automatically adjusted, the input of manpower is reduced, and the purposes of improving the tobacco shred quality and reducing the cost are achieved.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims.
Claims (7)
1. A control method of the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine comprises the following steps:
s1, acquiring production process data and environment data within one month, preprocessing the data and storing the data into a database as historical data;
the method for cleaning and preprocessing the historical data comprises the steps of eliminating null values and eliminating abnormal values;
s2, acquiring the latest N=15 batches of data, and solving the average value of each batch of sample data;
s3, establishing a linear regression model I of the water content of the leaf storage cabinet outlet by adopting a linear regression method, and solving to obtain the water content of the leaf storage cabinet outlet;
s4, taking the water content of the tobacco leaves at the outlet of the tobacco storage cabinet obtained by solving the water content difference and the linear regression model I as a variable, establishing a probability distribution model taking the set value of the water content of tobacco leaves at the outlet of the moisture regain machine as a target value, and solving to obtain the set value of the water content of tobacco leaves at the outlet of the moisture regain machine;
the water content difference is the difference value between the water content of the tobacco leaves in the tobacco leaves storage cabinet and the set value of the water content of tobacco leaves at the outlet of the moisture regain machine in the historical data;
s5, using the probability distribution model to solve the obtained set value of the moisture content of tobacco leaves at the outlet of the damping machine, the ambient temperature of the damping machine, the ambient humidity of the damping machine and the cumulative amount of the slicing damping electronic scale as variables, adopting a linear regression method to establish a linear regression model II which takes the initial value of the cumulative amount of loose damping water as a target value, and solving to obtain the initial value of the cumulative amount of loose damping water;
s6, taking the moisture content of tobacco leaves at the outlet of the damping machine as a target value, selecting loose damping water adding flow as a controllable variable in an input value, adopting a correlation analysis method to select the amount of a correlation coefficient r more than or equal to 0.5 as a correlation variable,
s7, taking the correlation variable in the step S6 as input, taking the moisture content of tobacco leaves at the outlet of the target value damping machine as output, and training by adopting a machine learning algorithm to obtain a machine learning prediction model;
solving a tobacco leaf water content prediction model at the outlet of the damping machine based on machine learning to obtain a predicted value of the tobacco leaf water content at the outlet of the damping machine;
s8, taking the set value of the moisture content of tobacco leaves at the outlet of the damping machine, which is obtained according to the probability distribution model in the step S4, as a best-pushing target value of a prediction model, and adopting an optimization algorithm to push loose damping water flow of each tobacco leaf output stage;
s9, converting the initial value of the accumulated water quantity of the loose conditioning into the initial value of the water quantity of the loose conditioning according to the initial value of the accumulated water quantity of the loose conditioning obtained by the linear regression model II in the step S5 through the formula (1) in the step S6, and taking the initial value of the water quantity of the loose conditioning when the production is started;
regulating the loosening and conditioning water adding flow according to the deviation between the moisture content of tobacco leaves at the outlet of the damping machine and the actual value of the moisture content of tobacco leaves at the outlet of the damping machine output by the prediction model,
when the absolute value of the deviation is less than 0.3%, the moisture content of the conditioning outlet is kept unchanged;
when the absolute value of the deviation is more than or equal to 0.3%, solving the optimal water adding flow by adopting the optimization algorithm in the step S8;
in the step S1 of the process,
the production process data comprises the moisture content of tobacco shreds at an inlet of a heating and humidifying device, the moisture content of a tobacco storage cabinet, a moisture content set value at an outlet of a moisture regaining machine, a cumulative amount of a slicing moisture regaining electronic scale, a flow rate of a loose moisture regaining inlet scale, an actual opening degree of a moisture regaining water adding film valve, an opening degree of a heating and humidifying film valve, a flow rate of the loose moisture regaining machine, a flow rate of loose moisture regaining steam, a temperature of loose moisture regaining hot air, a temperature of loose moisture regaining fresh air, a pressure of a loose moisture regaining discharge cover, a temperature of loose moisture regaining return air, a flow rate of a feeding machine inlet scale and a cumulative amount of the slicing moisture regaining electronic scale;
the environmental data comprise the environmental temperature of the cut-tobacco dryer, the environmental humidity of the cut-tobacco dryer, the environmental temperature of the moisture regain machine and the environmental humidity of the moisture regain machine;
in the step S3, the expression of the linear regression model I taking the water content of the leaf storage cabinet as the target value is that,
Y 1 =2.19486-0.07020X 1 -0.05841X 2 +1.18721X 3 （2）
wherein:
Y 1 -leaf storage cabinet outlet water content
X 1 -the ambient temperature of the cut-tobacco drier
X 2 -environmental humidity of cut-tobacco drier
X 3 -the moisture content of tobacco shreds at the inlet of the heating and humidifying equipment;
in the step S6, the correlation coefficient r is less than 0.3, no correlation exists, r is more than or equal to 0.3 and less than 0.5, low correlation is achieved, r is more than or equal to 0.5 and less than or equal to 0.8, obvious correlation is achieved, and r is more than or equal to 0.8 and less than or equal to 1.0, high correlation is achieved;
the correlation variable includes: the method comprises the steps of weighing flow rate of a loosening and conditioning inlet, actual opening of a conditioning water adding film valve, opening of a warming and wetting film valve, flow rate of loosening and conditioning water adding, flow rate of loosening and conditioning steam, temperature of loosening and conditioning hot air, temperature of loosening and conditioning fresh air, pressure of a loosening and conditioning discharging cover, temperature of loosening and conditioning return air, ambient temperature of a conditioning machine, ambient humidity of the conditioning machine, weighing flow rate of an inlet of a charging machine and cumulative amount of a slicing conditioning electronic scale;
and (3) carrying out pairwise analysis on all variables in the data by adopting a correlation analysis method, and selecting a correlation variable with a target value of the moisture content of tobacco leaves at the outlet of the damping machine and a correlation coefficient r more than or equal to 0.5 as a variable of the prediction model in the step S7.
2. The control method of the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine according to claim 1, which is characterized in that: in the step S2, the average value of the recent n=15 batches of production process data and environmental data is obtained as fitting data of the linear regression model i, the probability distribution model, and the linear regression model ii.
3. The control method of the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine according to claim 1, which is characterized in that: in the step S5, the expression of the linear regression model II aiming at the initial value of the loose conditioning water adding accumulation amount is that,
Y 2 =-3395.52679+11.93938X 4 -0.65249X 5 +109.58232X 6 +0.12824X 7 （3）
wherein:
Y 2 initial value of accumulated amount of loosening, conditioning and water adding
X 4 -environmental temperature of the conditioning machine
X 5 -moisture regain machine ambient humidity
X 6 -the set value of the moisture content of tobacco leaves at the outlet of the moisture regain machine
X 7 -slice conditioning electronic scale accumulation.
4. The control method of the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine according to claim 1, which is characterized in that: the collected production process data and environmental data are stored in a database.
5. The method for controlling the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine according to claim 4, wherein the method comprises the following steps: and the database updates the data of the database at preset time intervals, and updates the data of the tobacco leaf moisture content prediction model at the outlet of the damping machine by utilizing the updated database.
6. The control method of the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine according to claim 1, which is characterized in that: in step S8, the optimal loose conditioning water adding flow is solved by utilizing the optimization algorithm.
7. The method for controlling the moisture content of tobacco leaves at the outlet of a loosening and conditioning machine according to claim 5, which is characterized in that: in the production process of the step S9, the nodes are divided into four stages of initial stage, first stage, second stage and third stage according to the total yield of tobacco shreds of 0%, 25%, 50%, 75% and 100%,
when the total tobacco shred yield is more than or equal to 0 percent and less than 25 percent, solving the obtained initial value of the loose conditioning water adding accumulation by using a linear regression model II as the water adding flow value in the initial stage;
when the total tobacco shred yield is 25 percent or less and the tobacco shred yield is less than or equal to 50 percent of the total tobacco shred yield, taking a set value of the moisture regain machine outlet tobacco shred water content obtained by solving a probability distribution model as a pushing target value, adopting an optimization algorithm to push the prediction model, and solving a loose conditioning water adding flow pushing optimal value at the moment of 25 percent of the total tobacco shred yield as a water adding flow pushing optimal value in the first stage;
when the total tobacco shred yield is 50 percent or less and the tobacco shred yield is less than 75 percent of the total tobacco shred yield, taking a set value of the moisture regain machine outlet tobacco shred water content obtained by solving a probability distribution model as a pushing target value, adopting an optimization algorithm to push the prediction model, and solving a loose conditioning water adding flow pushing optimal value at the moment of 50 percent of the total tobacco shred yield as a water adding flow pushing optimal value of a second stage;
when the total tobacco shred yield is 75 percent or less and is less than 100 percent, taking a set value of the tobacco shred water content at the outlet of the damping machine, which is obtained by solving the probability distribution model, as a best pushing target value, adopting an optimization algorithm to push the prediction model, and solving to obtain a loose damping water adding flow best pushing value at the moment of 75 percent of the total tobacco shred yield, which is used as a water adding flow best pushing value in the third stage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210973084.1A CN115251445B (en) | 2022-08-15 | 2022-08-15 | Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210973084.1A CN115251445B (en) | 2022-08-15 | 2022-08-15 | Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115251445A CN115251445A (en) | 2022-11-01 |
| CN115251445B true CN115251445B (en) | 2023-05-23 |
Family
ID=83751699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210973084.1A Active CN115251445B (en) | 2022-08-15 | 2022-08-15 | Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115251445B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115844046A (en) * | 2022-11-30 | 2023-03-28 | 红塔烟草(集团)有限责任公司 | Silk making water content control method based on self-adaptive finite impulse response model |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2149897A (en) * | 1983-11-16 | 1985-06-19 | Brown & Williamson Tobacco | A process for drying tobacco |
| CN104781740A (en) * | 2012-07-10 | 2015-07-15 | 马丁丁亚胡·蒂亚诺 | Modular system for real-time evaluation and monitoring of machining production-line overall performances calculated from each given workpiece, tool and machine |
| WO2019072625A1 (en) * | 2017-10-10 | 2019-04-18 | Hauni Maschinenbau Gmbh | Method and system for producing rod-like products in the tobacco-processing industry |
| CN109674080A (en) * | 2019-03-07 | 2019-04-26 | 山东中烟工业有限责任公司 | Tobacco leaf conditioning amount of water prediction technique, storage medium and terminal device |
| CN111184246A (en) * | 2018-11-14 | 2020-05-22 | 厦门邑通软件科技有限公司 | Method and system for controlling moisture content of cut tobacco drying inlet |
| CN111248485A (en) * | 2018-11-30 | 2020-06-09 | 湖南中烟工业有限责任公司 | Method for online adjusting perfuming moisture of mixed silk |
| CN112220095A (en) * | 2020-09-29 | 2021-01-15 | 云南中烟工业有限责任公司 | Cigarette production method based on temperature of moisture regain area and cigarette |
| CN112434869A (en) * | 2020-11-30 | 2021-03-02 | 红云红河烟草(集团)有限责任公司 | Method for establishing quality prediction model of silk-making thread product |
| CN113096747A (en) * | 2021-04-12 | 2021-07-09 | 江苏丰尚智能科技有限公司 | Method and device for predicting moisture content of discharged material of dryer and computer equipment |
| CN113424979A (en) * | 2021-06-30 | 2021-09-24 | 张家口卷烟厂有限责任公司 | Loose moisture regaining incoming material moisture prediction method based on physical characteristics and physical movement characteristics |
| CN113812658A (en) * | 2021-08-26 | 2021-12-21 | 张家口卷烟厂有限责任公司 | Loose moisture regain and water adding control method based on neural network model and double parameter correction |
| CN114027539A (en) * | 2021-11-04 | 2022-02-11 | 湖北中烟工业有限责任公司 | Model prediction control-based loosening and conditioning quantitative water adding control method |
| CN114115393A (en) * | 2021-11-04 | 2022-03-01 | 湖北中烟工业有限责任公司 | Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line |
| CN114326397A (en) * | 2021-12-27 | 2022-04-12 | 首域科技(杭州)有限公司 | Model prediction control method of loosening and conditioning machine based on iterative learning optimization |
| CN114403487A (en) * | 2022-02-18 | 2022-04-29 | 河南中烟工业有限责任公司 | Water adding control method for loosening and moisture regaining |
| CN114532564A (en) * | 2022-03-24 | 2022-05-27 | 河南中烟工业有限责任公司 | Method for controlling moisture content of tobacco shreds at loosening and conditioning outlet based on principal component analysis strategy |
-
2022
- 2022-08-15 CN CN202210973084.1A patent/CN115251445B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2149897A (en) * | 1983-11-16 | 1985-06-19 | Brown & Williamson Tobacco | A process for drying tobacco |
| CN104781740A (en) * | 2012-07-10 | 2015-07-15 | 马丁丁亚胡·蒂亚诺 | Modular system for real-time evaluation and monitoring of machining production-line overall performances calculated from each given workpiece, tool and machine |
| WO2019072625A1 (en) * | 2017-10-10 | 2019-04-18 | Hauni Maschinenbau Gmbh | Method and system for producing rod-like products in the tobacco-processing industry |
| CN111184246A (en) * | 2018-11-14 | 2020-05-22 | 厦门邑通软件科技有限公司 | Method and system for controlling moisture content of cut tobacco drying inlet |
| CN111248485A (en) * | 2018-11-30 | 2020-06-09 | 湖南中烟工业有限责任公司 | Method for online adjusting perfuming moisture of mixed silk |
| CN109674080A (en) * | 2019-03-07 | 2019-04-26 | 山东中烟工业有限责任公司 | Tobacco leaf conditioning amount of water prediction technique, storage medium and terminal device |
| CN112220095A (en) * | 2020-09-29 | 2021-01-15 | 云南中烟工业有限责任公司 | Cigarette production method based on temperature of moisture regain area and cigarette |
| CN112434869A (en) * | 2020-11-30 | 2021-03-02 | 红云红河烟草(集团)有限责任公司 | Method for establishing quality prediction model of silk-making thread product |
| CN113096747A (en) * | 2021-04-12 | 2021-07-09 | 江苏丰尚智能科技有限公司 | Method and device for predicting moisture content of discharged material of dryer and computer equipment |
| CN113424979A (en) * | 2021-06-30 | 2021-09-24 | 张家口卷烟厂有限责任公司 | Loose moisture regaining incoming material moisture prediction method based on physical characteristics and physical movement characteristics |
| CN113812658A (en) * | 2021-08-26 | 2021-12-21 | 张家口卷烟厂有限责任公司 | Loose moisture regain and water adding control method based on neural network model and double parameter correction |
| CN114027539A (en) * | 2021-11-04 | 2022-02-11 | 湖北中烟工业有限责任公司 | Model prediction control-based loosening and conditioning quantitative water adding control method |
| CN114115393A (en) * | 2021-11-04 | 2022-03-01 | 湖北中烟工业有限责任公司 | Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line |
| CN114326397A (en) * | 2021-12-27 | 2022-04-12 | 首域科技(杭州)有限公司 | Model prediction control method of loosening and conditioning machine based on iterative learning optimization |
| CN114403487A (en) * | 2022-02-18 | 2022-04-29 | 河南中烟工业有限责任公司 | Water adding control method for loosening and moisture regaining |
| CN114532564A (en) * | 2022-03-24 | 2022-05-27 | 河南中烟工业有限责任公司 | Method for controlling moisture content of tobacco shreds at loosening and conditioning outlet based on principal component analysis strategy |
Non-Patent Citations (4)
| Title |
|---|
| 基于广义预测控制的松散回潮出***水率控制***;欧阳江子;王东方;戚晓江;李景伟;朱思奇;秦杨;蔡长兵;;计算机测量与控制(第03期);102-107 * |
| 基于梯度提升树的烟草回潮机出料含水率预测;何毅;李斌;普轶;李曾;周霄天;路瑶;金超;;软件(第06期);159-165 * |
| 李彦伟.《烟草制丝设备与工艺》.华中科技大学出版社,2014,8-10. * |
| 随机森林和DE-ELM的烘丝机入***水率预测;金发岗;王雅琳;张鹏程;陈晓方;李善莲;张炜;;控制工程(第03期);126-133 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115251445A (en) | 2022-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113017132A (en) | Cut tobacco quality optimization method based on cut tobacco dryer process parameter prediction | |
| CN115251445B (en) | Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine | |
| CN108652066A (en) | The water feeding method of loosening and gaining moisture process and the device for predicting the process amount of water | |
| CN102147613B (en) | Model predictive control method and system for moisture and temperature of tobacco redryer | |
| CN114027539B (en) | Model prediction control-based loosening and conditioning quantitative water adding control method | |
| CN112021626A (en) | Intelligent control system and method for tobacco shred making link | |
| CN115529987B (en) | Air port regulating and controlling method, device, equipment and storage medium for crop facility | |
| CN105996106A (en) | Cut tobacco quality control method and system based on airflow drying | |
| CN113812658B (en) | Loose moisture regain and water adding control method based on neural network model and double parameter correction | |
| CN114115393A (en) | Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line | |
| CN112273695A (en) | Method, device and equipment for predicting water content of loose moisture regain outlet | |
| CN113303489B (en) | Method for accurately controlling moisture of tobacco leaves in tobacco shred making process | |
| Gouadria et al. | Comparison between self-tuning fuzzy PID and classic PID controllers for greenhouse system | |
| CN114532564A (en) | Method for controlling moisture content of tobacco shreds at loosening and conditioning outlet based on principal component analysis strategy | |
| CN117195747B (en) | Uniform heat distribution optimization method for magnetic material drying | |
| CN112263012B (en) | Moisture content control method of redrying machine based on formula parameter library | |
| CN116611349B (en) | Neural network-based roller wire drying process parameter optimization method and system | |
| CN116268521A (en) | Self-adaptive intelligent control system of charging machine | |
| CN116757354A (en) | Tobacco redrying section key parameter screening method based on multilayer perceptron | |
| CN112800671B (en) | Data processing method and device and electronic equipment | |
| CN112231977A (en) | Construction method of formula parameter library of redrying machine | |
| CN107510089A (en) | A kind of method that accuracy of leaf feeding is improved in tobacco cutting production | |
| CN114668164B (en) | Loose conditioning water-adding amount self-adaptive control system based on incoming material difference | |
| CN117993770A (en) | Quality prediction method and system for machined product of conditioning machine | |
| Sangwan et al. | Comparative Study of Optimization Techniques for Tuning of PI Gains for Greenhouse Climate Control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |