CN110876480B - Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer - Google Patents

Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer Download PDF

Info

Publication number
CN110876480B
CN110876480B CN201910842040.3A CN201910842040A CN110876480B CN 110876480 B CN110876480 B CN 110876480B CN 201910842040 A CN201910842040 A CN 201910842040A CN 110876480 B CN110876480 B CN 110876480B
Authority
CN
China
Prior art keywords
wnn
cut tobacco
model
tobacco
drying
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
Application number
CN201910842040.3A
Other languages
Chinese (zh)
Other versions
CN110876480A (en
Inventor
周锋
朱培栋
谢明华
曹丽华
左毅
陈俊东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Changsha University
Original Assignee
Changsha University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Changsha University filed Critical Changsha University
Priority to CN201910842040.3A priority Critical patent/CN110876480B/en
Publication of CN110876480A publication Critical patent/CN110876480A/en
Application granted granted Critical
Publication of CN110876480B publication Critical patent/CN110876480B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

The invention discloses a modeling method of a cut tobacco drying process tail drying process of a drum-type cut tobacco dryer, which adopts a wavelet neural network to approximate a function type coefficient of a variable coefficient structure (VC) model, designs a WNN-VC model of the cut tobacco drying tail drying process, and can better describe the global nonlinear characteristic of the cut tobacco drying tail drying process. At present, most of parameter optimization methods aiming at variable coefficient models are Structural Nonlinear Parameter Optimization Methods (SNPOM) based on gradients, but the methods have obvious defects that the methods easily fall into local optimization. Considering that the particle filtering algorithm is particularly suitable for processing the problems of parameter estimation and state filtering of a nonlinear and non-Gaussian system, the particle filtering algorithm is adopted to carry out parameter optimization on the established model of the cut-tobacco drying tail process, and the method can optimize the global optimal parameters of the WNN-VC model of the cut-tobacco drying tail process.

Description

Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer
Technical Field
The invention relates to the technical field of cut tobacco drying processing in the tobacco industry, in particular to a modeling method for a tail drying process of cut tobacco drying of a drum-type cut tobacco drying machine.
Background
The drum-type cut tobacco drying machine is the most main processing equipment in a cut tobacco drying process workshop of a cigarette production enterprise. And (4) the cut tobacco transmitted from the previous cutting processing procedure is conveyed into a cut tobacco drying machine through a conveyor belt, and the cut tobacco drying machine is used for drying the cut tobacco, so that the moisture content of the cut tobacco after drying meets the process requirements of cigarette brands. The drum-type cut tobacco dryer mainly uses steam as a heat source to heat the cylinder body, and cut tobacco is fully contacted with the high-temperature cylinder wall in the rotating cylinder body, so that moisture in supplied cut tobacco is evaporated, and the moisture is discharged out of the machine body through a moisture discharge air door of the drum, and the cut tobacco drying process is realized. According to the production process, the whole process of drying the cut tobacco by the drum-type cut tobacco dryer can be divided into three stages: a stem section, a mid-section, and a stem tail section. And when the detected moisture content of the cut tobacco at the outlet is basically stabilized at a set value, the cut tobacco drying process is finished. Then, the process enters a relatively long tobacco drying intermediate process. When the flow of the cut tobacco at the inlet is changed from a normal value to zero, the cut tobacco drying tail process is marked to start until the cut tobacco is discharged out of the machine body, and the whole cut tobacco drying process is finished.
The tobacco shred baking process is an extremely complex nonlinear process and has the characteristics of multivariable, strong coupling and large time lag. Especially the dry-head-dry-tail process, due to the lack of important process variables, namely: the method is lack of detection values of the moisture content of the cut tobacco at the outlet or the inlet in the process of drying the head or the tail, and an accurate physical relation model of the cut tobacco drying process is difficult to establish in practice. At present, the prior art mainly establishes a simplified physical model based on various assumed conditions. Zhu et al (Zhu W, Lin H, Cao Y, et al. Thermal properties measurement of cut tobacco based on TPS method and Thermal conductivity model [ J ]. Journal of Thermal Analysis and calibration, 2014,116(3):1117-1123.) carried out simulation study on the heat transfer process of tobacco shred drying, and establish prediction models of tobacco shred heat conductivity coefficient under different conditions. Gu et al (Gu C, Zhang C, Zhang X, et al. modeling and simulation on flexible tobacco rod composites in the screw dryers [ J ]. Korean Journal of Chemical Engineering,2017,34(1):20-28.) verified the influence of the drum temperature on the heat and mass transfer of the dryer during the tobacco drying process through experiments, and proposed a mathematical model between the drum temperature and the moisture content of the tobacco at the outlet. However, in the above method, due to the limitation of the physical structure of the actual cut-tobacco drier, some key parameters are still difficult to obtain in the simplified physical relationship model, and the generality of the model is poor. At present, the research of obtaining models of tobacco shred drying stages by adopting advanced system identification technology is in a preliminary stage. How to adopt a data-driven modeling technology to obtain an identification model of each stage of tobacco shred drying is a key problem to be solved urgently in the tobacco shred drying industry at present.
Disclosure of Invention
The invention aims to solve the technical problem of providing a modeling method for a dry tail process of a cut tobacco drying procedure of a drum-type cut tobacco dryer aiming at the defects of the prior art,
in order to solve the technical problems, the technical scheme adopted by the invention is as follows: a modeling method for a dry tail process of a cut tobacco drying procedure of a drum-type cut tobacco dryer comprises the following steps:
1) collecting process variables in the tobacco shred drying and tail drying process, and constructing a modeling data set;
2) fitting a functional coefficient by using the modeling data set, and establishing a WNN-VC model structure in the cut tobacco drying tail process;
3) and carrying out global optimization on the WNN-VC model structure, and selecting the final WNN-VC model order m in the tobacco shred drying and tail drying process by using the minimum information criterion for the optimized WNN-VC model structure.
In the step 1), when the flow of the inlet cut tobacco is reduced to 40% from a normal set value, N process variable data are collected within 2 seconds of a sampling period, and the collection process is finished until the water content of the outlet cut tobacco is reduced to 2.5%; the N process variable data sets acquired in the whole process are as follows: data set of inlet cut tobacco flow
Figure BDA0002194019690000021
Data set of water content of inlet cut tobacco
Figure BDA0002194019690000022
Data set of drum rotation frequency
Figure BDA0002194019690000023
Data set of drum temperature
Figure BDA0002194019690000024
Data set of moisture extraction damper opening
Figure BDA0002194019690000025
And data set of water content of outlet cut tobacco { y1...yn}。
In the step 2), the WNN-VC model structure in the cut tobacco drying and tail drying process is as follows:
Figure BDA0002194019690000026
Figure BDA0002194019690000027
wherein: m is the order of the WNN-VC model, ytIs the moisture content of the cut tobacco at the outlet at the sampling moment t,
Figure BDA0002194019690000028
is the inlet tobacco flow at the sampling time t,
Figure BDA0002194019690000029
is the moisture content of the inlet cut tobacco at the sampling moment t,
Figure BDA00021940196900000210
is the drum rotation frequency at the moment of sampling t,
Figure BDA0002194019690000031
is the drum temperature at the sampling time t,
Figure BDA0002194019690000032
opening degree, xi, of the moisture exhausting damper at the time of t samplingtGaussian white noise at the sampling time t;
Figure BDA0002194019690000033
is the state quantity of WNN-VC model; phi is a0(wt-1)，φy,i(wt-1)，
Figure BDA0002194019690000034
And
Figure BDA0002194019690000035
all relate to the state quantity wt-1The wavelet neural network coefficients of (a);
Figure BDA0002194019690000036
in order to be offset in the amount of the offset,
Figure BDA0002194019690000037
is a weight coefficient;
Figure BDA0002194019690000038
are all the basis functions of the wavelet neural network.
The specific structure of the basis function of the wavelet neural network is as follows:
Figure BDA0002194019690000039
wherein the content of the first and second substances,
Figure BDA00021940196900000310
is the scale factor of the wavelet basis function,
Figure BDA00021940196900000311
is a shift factor of the wavelet basis function,
Figure BDA00021940196900000312
representing a 2-norm operation.
In step 3), the specific implementation process of performing global optimization on the WNN-VC model structure comprises the following steps:
1) defining a parameter set theta to be estimated in a WNN-VC model at the time ttIs a state vector, i.e.
Figure BDA00021940196900000313
Figure BDA00021940196900000314
The self-organization state space model structure is constructed as follows:
Figure BDA00021940196900000315
wherein:
Figure BDA00021940196900000316
and is
Figure BDA00021940196900000317
Figure BDA00021940196900000318
Wherein j is 0pAnd when l ═ j-i ≦ 0,
Figure BDA00021940196900000319
Φt+1is system noise, gammatTo observe noise; definition of
Figure BDA0002194019690000041
Then
Figure BDA0002194019690000042
For observing noisetA prediction output sequence of a WNN-VC model in the later cut tobacco drying and tail drying process;
2) initializing the particle state x (0), the noise term Φt+1And upsilontAre gaussian distributions independently of each other, i.e.
Figure BDA0002194019690000043
Φt+1～N(0,Q)，Υt～N(0,R)；
3) At the time t, according to the WNN-VC model and the constructed self-organizing state space model
Figure BDA0002194019690000044
Calculating a multi-step forward prediction error term:
Figure BDA0002194019690000045
the objective function defining the optimization algorithm is:
Figure BDA0002194019690000046
wherein n is the length of the identification data, npM is the order of the WNN-VC model for predicting the time domain;
4) sequentially encoding parameters x (t) to be optimized as particles, and calculating each object pair by adopting a Monte Carlo particle filter algorithmObjective function V (x)*) Selecting the optimal individual with lower adaptive value to be reserved to the next generation and defining the optimal individual as an excellent population, carrying out cross and mutation operation on other particles, and generating a corresponding alternative population; comparing the fitness of the individuals in the alternative population, selecting the optimal individual to be added into the excellent population, and repeating the step 3) until the optimal particle set x is obtained*And let parameter set theta of WNN-VC modelt＝x*。
Minimum information criterion AIC ═ n-m) log (V (x)*))+4(4m+1)。
Compared with the prior art, the invention has the beneficial effects that: aiming at the problem that a mechanism model is difficult to obtain in the tobacco shred drying and tail drying process of a drum-type tobacco shred dryer, the WNN-VC model in the tobacco shred drying and tail drying process is established by adopting a system identification technology. The WNN-VC model integrates the advantages of the WNN model in the aspect of nonlinear function approximation and the VC model in the aspect of industrial process system modeling, the established WNN-VC model has the capability of describing global nonlinear characteristics of the cut tobacco drying tail-drying process, and meanwhile the order of the WNN network in the model is effectively reduced, so that the established WNN-VC model in the tail-drying process is more suitable for the design of a subsequent advanced control algorithm based on the model. The WNN-VC model in the tobacco drying and tail drying process is subjected to parameter optimization by adopting a particle filtering algorithm, and the method can effectively avoid the model from being partially optimal in the parameter optimization process, so that the overall optimal parameters of the WNN-VC model in the tobacco drying and tail drying process are optimized, and the aim of greatly improving the modeling precision of the tobacco drying and tail drying process is fulfilled.
Drawings
FIG. 1 is a schematic view of a process of drying tail of cut tobacco in a drum-type cut tobacco dryer according to the present invention.
Detailed Description
Aiming at the tobacco shred drying tail process of the drum-type shred dryer shown in figure 1, the specific implementation mode of the modeling and parameter optimization method based on the WNN-VC model is as follows:
step 1: after the tobacco shred drying and tail drying process starts, the real-time data of 6 process variables shown in the figure 1 are respectively collected in a sampling period of 2 seconds until the tail drying process is finished. Finally, obtainData set of inlet tobacco flow to modeling for tobacco drying and tail drying process
Figure BDA0002194019690000051
Data set of water content of inlet cut tobacco
Figure BDA0002194019690000052
Data set of drum rotation frequency
Figure BDA0002194019690000053
Data set of drum temperature
Figure BDA0002194019690000054
Data set of moisture extraction damper opening
Figure BDA0002194019690000055
And data set of water content of outlet cut tobacco
Figure BDA0002194019690000056
Step 2: the WNN-VC model structure for the cut tobacco drying and tail drying process shown in FIG. 1 is established as follows:
Figure BDA0002194019690000057
Figure BDA0002194019690000058
wherein: m is the order of the WNN-VC model, ytIs the moisture content of the cut tobacco at the outlet at the sampling moment t,
Figure BDA0002194019690000059
is the inlet tobacco flow at the sampling time t,
Figure BDA00021940196900000510
is the moisture content of the inlet cut tobacco at the sampling moment t,
Figure BDA00021940196900000511
is the drum rotation frequency at the moment of sampling t,
Figure BDA00021940196900000512
is the drum temperature at the sampling time t,
Figure BDA00021940196900000513
opening degree, xi, of the moisture exhausting damper at the time of t samplingtGaussian white noise at the sampling time t;
Figure BDA00021940196900000514
is the state quantity of WNN-VC model; phi is a0(wt-1)，φy,i(wt-1)，
Figure BDA00021940196900000515
And
Figure BDA00021940196900000516
all relate to the state quantity wt-1The wavelet neural network coefficients of (a);
Figure BDA00021940196900000517
in order to be offset in the amount of the offset,
Figure BDA0002194019690000061
are weight coefficients. In formula (2)
Figure BDA0002194019690000062
The structure of the basis function is as follows:
Figure BDA0002194019690000063
wherein the content of the first and second substances,
Figure BDA0002194019690000064
is the scale factor of the wavelet basis function,
Figure BDA0002194019690000065
is a shift factor of the wavelet basis function,
Figure BDA0002194019690000066
representing a 2-norm operation.
And step 3: and globally optimizing the parameters of the WNN-VC model (1) by adopting a particle filtering algorithm.
Step 1: defining a parameter set theta to be estimated in a WNN-VC model (1) at the time ttAs a state vector, i.e.:
Figure BDA0002194019690000067
selecting the prediction output of the model as the observation vector of the filter, i.e.
Figure BDA0002194019690000068
Wherein n ispIs the prediction time domain. In order to carry out parameter estimation on the WNN-VC model (1) by adopting a particle filtering method, the structure of the constructed self-organizing state space model is as follows:
Figure BDA0002194019690000069
wherein: phit+1Is system noise, gammatIn order to observe the noise, it is,
Figure BDA00021940196900000610
and f isj(t) the specific structure is as follows:
Figure BDA00021940196900000611
in the formula (6), when l ═ j-i ≦ 0,
Figure BDA00021940196900000612
and is
Figure BDA00021940196900000613
Step 2: an initialization stage: initializing the particle state x (0), the noise term Φt+1And upsilontAre gaussian distributions independently of each other, i.e.
Figure BDA00021940196900000614
Φt+1～N(0,Q)，Υt～N(0,R)。
Step 3: a prediction stage: at time t, according to the WNN-VC model (1) and the state space equation (5) of the particles, the multi-step forward prediction error is calculated as follows:
Figure BDA0002194019690000071
the objective function defining the optimization algorithm is as follows:
Figure BDA0002194019690000072
wherein n is the length of the identification data, npTo predict the time domain, m is the order of the WNN-VC model.
Step 4: and (3) an updating stage: and (3) coding the parameters x (t) to be optimized as particles in sequence, calculating the adaptive value of each individual to the objective function (8) in Step 3 by adopting a Monte Carlo particle filter algorithm, selecting the optimal individual with a lower adaptive value to be reserved to the next generation and defining the optimal individual as an excellent group, and performing cross and variation operation on other particles to generate a corresponding alternative group. Then, the fitness of the individuals in the alternative population is compared, the optimal individual is selected and added into the excellent population, and the process is repeated until the optimal particle set x is obtained*And a parameter set theta of WNN-VC modelt＝x*。
And 4, step 4: and selecting the optimal order m of the WNN-VC model in the tobacco shred drying and tail drying process.
The formula for defining the minimum information criterion AIC value is as follows:
AIC＝(n-m)log(V(x*))+4(4m+1) (9)
and (3) aiming at different WNN-VC model orders m, repeatedly adopting the method in the step (3) to carry out parameter optimization on the model (1), and searching for an order m which can enable the AIC value to be minimum as the order of the WNN-VC model in the tobacco cut-tobacco drying tail-drying process, wherein the order m of the WNN-VC model (1) in the cut-tobacco drying tail-drying process finally selected in the specific embodiment is 12.

Claims (5)

1. A modeling method for a dry tail process of a cut tobacco drying procedure of a drum-type cut tobacco dryer is characterized by comprising the following steps:
1) collecting process variables in the tobacco shred drying and tail drying process, and constructing a modeling data set;
2) fitting a functional coefficient by using the modeling data set, and establishing a WNN-VC model structure in the cut tobacco drying tail process; the WNN-VC model structure in the cut tobacco drying and tail drying process is as follows:
Figure FDA0003124487160000011
Figure FDA0003124487160000012
wherein: m is the order of the WNN-VC model, ytIs the moisture content of the cut tobacco at the outlet at the sampling moment t,
Figure FDA0003124487160000013
is the inlet tobacco flow at the sampling time t,
Figure FDA0003124487160000014
is the moisture content of the inlet cut tobacco at the sampling moment t,
Figure FDA0003124487160000015
is the drum rotation frequency at the moment of sampling t,
Figure FDA0003124487160000016
roller barrel for sampling time tThe temperature of the mixture is controlled by the temperature,
Figure FDA0003124487160000017
opening degree, xi, of the moisture exhausting damper at the time of t samplingtGaussian white noise at the sampling time t;
Figure FDA0003124487160000018
is the state quantity of WNN-VC model; phi is a0(wt-1)，φy,i(wt-1)，
Figure FDA0003124487160000019
Figure FDA00031244871600000110
And
Figure FDA00031244871600000111
all relate to the state quantity wt-1The wavelet neural network coefficients of (a);
Figure FDA00031244871600000112
in order to be offset in the amount of the offset,
Figure FDA00031244871600000113
is a weight coefficient;
Figure FDA00031244871600000114
all are basis functions of wavelet neural network
3) And carrying out global optimization on the WNN-VC model structure, and selecting the final WNN-VC model order m in the tobacco shred drying and tail drying process by using the minimum information criterion for the optimized WNN-VC model structure.
2. The modeling method for the dry tail process in the cut tobacco drying process of the roller-type cut tobacco dryer according to claim 1, characterized in that in the step 1), when the flow rate of the inlet cut tobacco is reduced to 40% from a normal set value, N process variable data are collected within 2 seconds of a sampling period until the outlet smoke is exhaustedWhen the moisture content of the silk is reduced to 2.5%, the collection process is finished; the N process variable data sets acquired in the whole process are as follows: data set of inlet cut tobacco flow
Figure FDA0003124487160000021
Data set of water content of inlet cut tobacco
Figure FDA0003124487160000022
Data set of drum rotation frequency
Figure FDA0003124487160000023
Data set of drum temperature
Figure FDA0003124487160000024
Data set of moisture extraction damper opening
Figure FDA0003124487160000025
And data set of water content of outlet cut tobacco { y1...yn}。
3. The modeling method for the dry tail process of the cut-tobacco drying process of the roller-type cut-tobacco dryer according to claim 1, characterized in that the specific structure of the basis function of the wavelet neural network is as follows:
Figure FDA0003124487160000026
wherein the content of the first and second substances,
Figure FDA0003124487160000027
is the scale factor of the wavelet basis function,
Figure FDA0003124487160000028
is a translation factor of the wavelet basis function, | | | · | | represents a 2-norm operation.
4. The modeling method for the dry tail process of the cut-tobacco drying process of the roller-type cut-tobacco dryer according to claim 1, wherein in the step 3), the specific implementation process for carrying out global optimization on the WNN-VC model structure comprises the following steps:
1) defining a parameter set theta to be estimated in a WNN-VC model at the time ttIs a state vector, i.e.
Figure FDA0003124487160000029
Figure FDA00031244871600000210
The self-organization state space model structure is constructed as follows:
Figure FDA00031244871600000211
wherein:
Figure FDA00031244871600000212
and is
Figure FDA00031244871600000213
Figure FDA00031244871600000214
Wherein j is 0pAnd when l ═ j-i ≦ 0,
Figure FDA00031244871600000215
Φt+1is system noise, gammatTo observe noise; definition of
Figure FDA0003124487160000031
Then
Figure FDA0003124487160000032
For observing noisetA prediction output sequence of a WNN-VC model in the later cut tobacco drying and tail drying process;
2) initializing the particle state x (0), the noise term Φt+1And upsilontAre independent Gaussian distributions, i.e. x (0) to N (μ0,θ0)，Φt+1～N(0,Q)，Υt～N(0,R)；
3) At the time t, according to the WNN-VC model and the constructed self-organizing state space model
Figure FDA0003124487160000033
Calculating a multi-step forward prediction error term:
Figure FDA0003124487160000034
the objective function defining the optimization algorithm is:
Figure FDA0003124487160000035
wherein n is the length of the identification data, npM is the order of the WNN-VC model for predicting the time domain;
4) sequentially encoding parameters x (t) to be optimized as particles, and calculating each object by adopting a Monte Carlo particle filter algorithm aiming at a target function V (x)*) Selecting the optimal individual with lower adaptive value to be reserved to the next generation and defining the optimal individual as an excellent population, carrying out cross and mutation operation on other particles, and generating a corresponding alternative population; comparing the fitness of the individuals in the alternative population, selecting the optimal individual to be added into the excellent population, and repeating the step 3) until the optimal particle set x is obtained*And let parameter set theta of WNN-VC modelt＝x*。
5. The method of claim 1, wherein the minimum information criterion AIC ═ n-m log (V (x) is selected*))+4(4m+1)。
CN201910842040.3A 2019-09-06 2019-09-06 Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer Active CN110876480B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910842040.3A CN110876480B (en) 2019-09-06 2019-09-06 Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910842040.3A CN110876480B (en) 2019-09-06 2019-09-06 Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer

Publications (2)

Publication Number Publication Date
CN110876480A CN110876480A (en) 2020-03-13
CN110876480B true CN110876480B (en) 2021-08-13

Family

ID=69727814

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910842040.3A Active CN110876480B (en) 2019-09-06 2019-09-06 Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer

Country Status (1)

Country Link
CN (1) CN110876480B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB1486201A (en) * 1973-10-15 1977-09-21 Industrial Nucleonics Corp Tobacco dryer controller
CN102488308A (en) * 2011-12-14 2012-06-13 东华大学 Advanced coordinated control system for moisture in cut tobacco dryer
CN102871214A (en) * 2012-10-08 2013-01-16 秦皇岛烟草机械有限责任公司 Model prediction based cut tobacco dryer outlet moisture control method
CN103610227A (en) * 2013-12-09 2014-03-05 中南大学 Cut tobacco dryer head and tail section process variable optimizing control method
CN104268408A (en) * 2014-09-28 2015-01-07 江南大学 Energy consumption data macro-forecast method based on wavelet coefficient ARMA model

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB1486201A (en) * 1973-10-15 1977-09-21 Industrial Nucleonics Corp Tobacco dryer controller
CN102488308A (en) * 2011-12-14 2012-06-13 东华大学 Advanced coordinated control system for moisture in cut tobacco dryer
CN102871214A (en) * 2012-10-08 2013-01-16 秦皇岛烟草机械有限责任公司 Model prediction based cut tobacco dryer outlet moisture control method
CN103610227A (en) * 2013-12-09 2014-03-05 中南大学 Cut tobacco dryer head and tail section process variable optimizing control method
CN104268408A (en) * 2014-09-28 2015-01-07 江南大学 Energy consumption data macro-forecast method based on wavelet coefficient ARMA model

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
烘丝机动态特性建模及基于模型的出口水分优化控制方法;王小飞等;《中国烟草学会2016年度优秀论文汇编》;20161201;1098-1104页 *

Also Published As

Publication number Publication date
CN110876480A (en) 2020-03-13

Similar Documents

Publication Publication Date Title
CN111275288B (en) XGBoost-based multidimensional data anomaly detection method and device
CN111045326B (en) Tobacco shred drying process moisture prediction control method and system based on recurrent neural network
CN110946313B (en) Method and system for controlling water content of outlet of cut tobacco drying process
CN114115393A (en) Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line
CN116646030B (en) Tobacco tar component identification method and system based on electronic smoke detection
CN110876480B (en) Modeling method for dry tail process of cut tobacco drying process of drum-type cut tobacco dryer
CN105334738B (en) A kind of method of evaluating performance suitable for tobacco processing course pid control circuit
CN111241754B (en) Soft measurement method for key process parameters of paper drying
CN116602435A (en) Machine learning-based method for analyzing moisture change of cut tobacco in cut tobacco making machine
CN109211311B (en) Tobacco shred drying process quality consistency evaluation method based on different production line processing
CN110472321B (en) PSO-GPR (particle swarm optimization-GPR) based method for predicting processing energy consumption of all-metal semi-hard shell solid rocket cabin section
CN116757354A (en) Tobacco redrying section key parameter screening method based on multilayer perceptron
CN110826229A (en) Cut tobacco drying head process modeling and optimal setting control method based on long-term prediction
CN111838744B (en) Continuous real-time prediction method for moisture in tobacco shred production process based on LSTM (localized surface plasmon resonance) environment temperature and humidity
CN110558598B (en) Data-driven FNN-ARX modeling method for tobacco shred drying and head drying process
CN112183642A (en) Method and system for detecting coal consumption of cement firing based on random forest model
CN110580326B (en) Modeling method for tobacco shred drying intermediate process of tobacco shred drying machine
CN110673490B (en) Long-term prediction modeling and optimal setting control method for cut tobacco drying tail process
CN117766051A (en) AE-BiRNN network-based hyperspectral sheet tobacco smoke moisture degree detection and evaluation method
CN116205363A (en) Method for predicting processing strength of dried shreds based on CNN-GRU combined network model
Feng et al. An adaptive dual-population based evolutionary algorithm for industrial cut tobacco drying system
CN114593960B (en) Pure steam sampling method and system under multiple factors
CN116525024A (en) Method for predicting water content of tobacco shreds at outlet of drying device
CN116998747A (en) Method, medium and system for adjusting water content in silk manufacturing process
CN116258080A (en) Method and device for predicting outlet water content of sheet cut-tobacco dryer

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