CN112231977A - Construction method of formula parameter library of redrying machine - Google Patents
Construction method of formula parameter library of redrying machine Download PDFInfo
- Publication number
- CN112231977A CN112231977A CN202011101490.6A CN202011101490A CN112231977A CN 112231977 A CN112231977 A CN 112231977A CN 202011101490 A CN202011101490 A CN 202011101490A CN 112231977 A CN112231977 A CN 112231977A
- Authority
- CN
- China
- Prior art keywords
- redrying
- data
- algorithm
- parameter library
- 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.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 12
- 238000001035 drying Methods 0.000 claims abstract description 42
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 35
- 238000000034 method Methods 0.000 claims abstract description 30
- 238000013528 artificial neural network Methods 0.000 claims abstract description 21
- 241000208125 Nicotiana Species 0.000 claims abstract description 15
- 235000002637 Nicotiana tabacum Nutrition 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 11
- 230000002068 genetic effect Effects 0.000 claims abstract description 8
- 238000004886 process control Methods 0.000 claims abstract description 5
- 238000003062 neural network model Methods 0.000 claims abstract description 3
- 238000012549 training Methods 0.000 claims description 25
- 238000004519 manufacturing process Methods 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000008030 elimination Effects 0.000 claims description 3
- 238000003379 elimination reaction Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 2
- 239000013072 incoming material Substances 0.000 claims description 2
- 210000002364 input neuron Anatomy 0.000 claims description 2
- 238000012567 pattern recognition method Methods 0.000 claims description 2
- 239000013598 vector Substances 0.000 claims description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 abstract description 8
- 230000008878 coupling Effects 0.000 abstract description 3
- 238000010168 coupling process Methods 0.000 abstract description 3
- 238000005859 coupling reaction Methods 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 description 3
- 241000196324 Embryophyta Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000018044 dehydration Effects 0.000 description 1
- 238000006297 dehydration reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- A24B3/10—Roasting or cooling tobacco
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Fuzzy Systems (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Feedback Control In General (AREA)
Abstract
The invention belongs to the related technical field of tobacco processing process control, and relates to a construction method of a formula parameter library of a redrying machine, in particular to a construction method of a formula parameter library aiming at a drying link of tobacco redrying equipment and an implementation method thereof. The method is characterized in that: aiming at the characteristics of numerous control factors, complex mechanism and strong coupling of a drying link of redrying equipment, a formula parameter library based on a neural network model is constructed by adopting an improved BP algorithm combined with a genetic algorithm, a proper drying mode is firstly determined, neural network modeling is carried out on the basis, a momentum term and a self-adaptive learning rate are added into a mixed BP algorithm to improve the convergence speed of the algorithm and the prediction accuracy of the model, the model is further continuously updated and optimized by utilizing a mode identification method and historical data, the formula parameter library suitable for the redrying drying link is obtained, and a foundation is laid for intelligent control of a redrying water content process.
Description
Technical Field
The invention relates to the related technical field of tobacco processing process control, in particular to a construction method and an implementation method of a formula parameter library aiming at a drying link of tobacco redrying equipment.
Background
The moisture content control of the redrying machine is the most complex and key control link in the threshing and redrying process, the control quality of the redrying machine directly influences the processing quality of a subsequent processing link, and the redrying machine is a control link for focusing and monitoring of processing enterprises. The formula parameter library is the basis for realizing the intelligent control of the moisture content of the redrying machine, and the key for realizing the intelligent and accurate control is how to construct the formula parameter library which is suitable for the process control characteristics of the redrying equipment and meets the intelligent control requirements.
The drying link of the redrying equipment has the characteristics of numerous control parameters, complex control mechanism and strong coupling, and can be designed and completed by adopting a neural network modeling method, obtaining a formula parameter library with better prediction capability through massive training of input and output parameter samples and carrying out intelligent control on the basis. The construction method has the following problems in practical application: firstly, the drying mode has no reasonable optimization design, training samples are experience operation data, secondly, neural network training mostly adopts a single BP algorithm, the convergence rate is slow, the training time is long, the training is easy to trap into local minimum, and the practicability and the accuracy of a formula parameter library are influenced.
Disclosure of Invention
The invention aims to improve the defects of the method for constructing the formula parameter library, provides the method for constructing the formula parameter library of the redrying machine, and lays a foundation for intelligent control of the moisture content of redrying.
The invention aims to provide the following technical scheme for realizing the purpose:
a construction method of a redrying machine formula parameter library comprises the steps of firstly analyzing and optimizing a redrying drying mode to obtain a proper drying mode, on the basis, adopting a BP algorithm combined with a genetic algorithm to a production data sample, adding a momentum term and a self-adaptive learning rate into the BP algorithm to improve the convergence speed and accuracy of the algorithm, obtaining a formula database with better prediction and generalization capability, and further optimizing and maintaining the database by adopting a mode identification-based method. The method comprises the following specific steps:
(1) the method specifically comprises the steps of forming different drying modes by adjusting drying temperatures of first to sixth drying zones of the drying zones on the premise of ensuring that the total drying strength and the moisture content of a cold room are basically consistent, inspecting the influence of the different drying modes on the redrying quality, and determining the optimal drying mode.
(2) In the actual production, the tobacco flakes are processed by adopting an optimized drying mode, and input and output parameter variables of a formula parameter library are determined through analysis of the production process, incoming material conditions, grades and the like.
(3) Determining an acquisition rule: including the sampling time and the training data range. According to the characteristics of the tobacco processing process, the sampling time is 5-15 seconds; selecting a data section (removing a stub bar, a stub bar and production data of a fault stage) which stably runs in the production process in the evaluation data range;
(4) data preprocessing: if the data sample meets the training requirement of the neural network, further preprocessing the data, including abnormal value elimination and normalization processing, and if the data sample does not meet the training requirement, increasing the similar adjacent sample data expansion sample range by a pattern recognition method.
(5) Determining a neural network topology: aiming at the characteristics of the redrying process, from the aspects of simplicity, effectiveness and practicability, a BP neural network with a single hidden layer structure, namely a three-layer BP neural network structure, is usually selected for training and construction.
(6) A BP neural network algorithm combined with a genetic algorithm is adopted, and momentum items and a self-adaptive learning rate are added into the BP algorithm to train and model data samples, so that the convergence speed of the algorithm and the model accuracy are improved.
(7) The method has the advantages that a large number of trained neural network models are constructed, model analysis and data calling are facilitated, the models are optimized and maintained continuously through historical production data, generalization capability and prediction accuracy of the models are improved, and accordingly a formula parameter library capable of being updated and optimized continuously is constructed.
And (4) removing the abnormal value in the step (4) by adopting a Leidete criterion to remove coarse error data, namely removing the abnormal value if the residual error of the measured value is more than three times of the standard deviation. The normalization process controls the data to be in the range of [ -1,1], which is expressed as follows:
where j denotes the number of rows of data p and also the number of input neurons, pj minAnd pj maxRespectively representing the minimum and maximum values of the j-th row in p.
In the step (6), momentum terms and self-adaptive learning rate are added into the BP algorithm to improve the convergence speed of the algorithm, and the weight value adjustment formula is as follows:
w(k+1)＝w(k)+(1-a)η(k)d(k)+a(w(k)-w(k-1)) (2)
wherein w (k +1), w (k) and w (k-1) are weight vectors at the time points k +1, k and k-1, respectively, η (k) is a learning rate at the time point k, d (k) is a negative gradient at the time point k, and α is a momentum factor.
In consideration of numerous controlled factors in the redrying process, complex control mechanism and strong coupling, the invention adopts the BP algorithm combined with the genetic algorithm to carry out neural network modeling on input and output parameters, adds momentum items and self-adaptive learning rate in the BP algorithm to improve the convergence speed of the algorithm and the model precision, further increases the number of training samples by a mode recognition method and continuously updates an optimization model by historical production data, increases the practicability of the model and lays a foundation for the intelligent control of the redrying water content.
Drawings
FIG. 1 shows a step of constructing a recipe parameter library of a redrying machine;
FIG. 2 shows different drying mode settings in the redrying and drying process;
FIG. 3 is a formula parameter library neural network topology;
FIG. 4 is an improved neural network training algorithm incorporating a genetic algorithm.
Detailed Description
The invention is further described with reference to the following figures and examples:
the method comprises the following specific implementation steps of constructing a formula parameter library for a drying link by taking redrying equipment (the flow of a front electronic scale is 9000kg/h) of a certain threshing and redrying plant as an implementation object and redrying tobacco leaf production data taking Henan Zhongyan tobacco C3F as a main formula (shown in figure 1):
(1) on the premise of ensuring that the total drying strength and the moisture content of the cold room are basically consistent, the drying temperatures of the first-sixth drying areas are adjusted to form different drying modes (figure 2), and the influence of the different drying modes on the redrying quality is inspected. Through analyzing and comparing the dehydration rate of the batch of the sheet tobacco and the shrinkage rate of the outlet sheet tobacco, the method is suitable for adopting a first drying mode, the drying efficiency is high, the uniformity of the moisture content of the tobacco leaves after cooling and moisture regaining is good, the tobacco sheets are stretched, and the shrinkage rate is small.
(2) Determining 5 parameters of inlet water content, inlet flow, environment temperature and humidity and outlet water content as input variables of a formula parameter library by combining the actual raw material conditions and process characteristics of redrying equipment production; the hot air temperature set values of the first zone to the sixth zone of the drying zone are output variables of the formula parameter library.
(3) After a certain batch of tobacco flakes completely enter redrying (namely a normal production stage), after the system stably runs for 5 minutes, simultaneously sampling 5 parameters of inlet water content, inlet flow, environment temperature and humidity, outlet water content and hot air temperature set values of first to sixth areas of a drying area (the sampling time is 10 seconds), and sampling for 5 minutes before production is finished.
(4) If the requirement of neural network training is met (the sample capacity is larger than 9000), carrying out abnormal value elimination and normalization processing on input and output sample data, and randomly selecting 70% of the input and output sample data as training data and 30% of the input and output sample data as test data. And if the sample training requirement is not met, increasing the number of similar input and output samples through pattern recognition.
(5) According to the modeling reality and the production characteristics of the redrying and drying link, a BP neural network with a single hidden layer structure, namely a three-layer BP neural network structure, is adopted for training and construction (figure 3). f1, f2 and f3 respectively represent an input layer, a hidden layer and an output layer in a network structure, xi, i ═ 1,2,3,4,5 represent input variables of a neural network (corresponding to the instantaneous flow rate of an electronic scale, the feed water content, the ambient temperature, the ambient humidity and the discharge water content), and yi, i ═ 1,2,3,4,5,6 represent the output of the whole system, namely the set value of the hot air temperature corresponding to the drying areas 1-6. The number of cryptic neurons was designed to be 5.
(6) Setting the population size NT to be 20, the cross probability Pc to be 0.4, the variation probability Pm to be 0.05 and the maximum iteration number T to be 100; and adding momentum term and adaptive learning rate (equations (2) - (3)) into BP algorithm, wherein the initial learning rate is 0.24, alpha is 0.8, k is1＝0.7，k21.07, a BP neural network algorithm combined with a genetic algorithm was used for network training (fig. 4).
(7) And after the data training is finished, the construction of a related database is carried out, so that the parameter calling and the data analysis are facilitated, a model training window is opened, and the continuous updating training and optimization of the model training window at the later stage are facilitated.
Compared with a single BP algorithm, the average convergence rate of the method provided by the invention is improved by 30%, and training oscillation does not occur in the convergence process, so that the difficulty in constructing a formula parameter library in the drying link of redrying equipment is well solved, and a key foundation is laid for further intelligent redrying control.
Claims (4)
1. A construction method of a redrying machine formula parameter library is characterized by comprising the following steps: firstly, analyzing and optimizing a redrying drying mode to obtain a proper drying mode, on the basis, adopting a BP algorithm combined with a genetic algorithm to a production data sample, adding a momentum term and a self-adaptive learning rate into the BP algorithm to improve the convergence speed and precision of the algorithm, obtaining a formula database with better prediction and generalization capability, and further optimizing and maintaining the database by adopting a mode recognition-based method.
2. The method for constructing a parameter library of a redrying machine formula according to claim 1, wherein: the method comprises the following specific steps:
(1) analyzing the influence of the drying mode on the redrying quality, specifically forming different drying modes by adjusting the drying temperatures of first to sixth drying zones on the premise of ensuring that the total drying strength is basically consistent with the moisture content of a cold room, inspecting the influence of the different drying modes on the redrying quality, and determining the optimal drying mode;
(2) processing the tobacco flakes by adopting an optimized drying mode in actual production, and determining input and output parameter variables of a formula parameter library through analysis of the production process, incoming material conditions, grades and the like;
(3) determining an acquisition rule: including the sampling time and the training data range. According to the characteristics of the tobacco processing process, the sampling time is 5-15 seconds; selecting a data section which stably runs in the production process in the evaluation data range, namely removing the production data of a stub bar, a stub bar and a fault stage;
(4) data preprocessing: if the data sample meets the training requirement of the neural network, further preprocessing the data, including abnormal value elimination and normalization processing, and if the data sample does not meet the training requirement, increasing the similar adjacent sample data expansion sample range by a pattern recognition method;
(5) determining a neural network topology: aiming at the characteristics of the redrying process, from the aspects of simplicity, effectiveness and practicability, a BP neural network with a single hidden layer structure, namely a three-layer BP neural network structure, is usually selected for training and construction;
(6) a BP neural network algorithm combined with a genetic algorithm is adopted, and momentum items and a self-adaptive learning rate are added into the BP algorithm to train and model data samples, so that the convergence speed of the algorithm and the model accuracy are improved;
(7) the method has the advantages that a large number of trained neural network models are constructed, model analysis and data calling are facilitated, the models are optimized and maintained continuously through historical production data, generalization capability and prediction accuracy of the models are improved, and accordingly a formula parameter library capable of being updated and optimized continuously is constructed.
3. The method for constructing a parameter library of a redrying machine formula according to claim 1, wherein: in the step (4), removing the abnormal value by adopting a Leaiter criterion to remove coarse error data, namely removing the abnormal value if the residual error of the measured value is more than three times of the standard deviation; the normalization process controls the data to be in the range of [ -1,1], which is expressed as follows:
where j denotes the number of rows of data p and also the number of input neurons, pjminAnd pjmaxRespectively representing the minimum and maximum values of the j-th row in p.
4. The method for constructing a parameter library of a redrying machine formula according to claim 1, wherein:
in the step (6), momentum terms and self-adaptive learning rate are added into the BP algorithm to improve the convergence speed of the algorithm, and the weight value adjustment formula is as follows:
w(k+1)＝w(k)+(1-a)η(k)d(k)+a(w(k)-w(k-1)) (2)
wherein w (k +1), w (k) and w (k-1) are weight vectors at the time points k +1, k and k-1, respectively, η (k) is a learning rate at the time point k, d (k) is a negative gradient at the time point k, and α is a momentum factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011101490.6A CN112231977A (en) | 2020-10-15 | 2020-10-15 | Construction method of formula parameter library of redrying machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011101490.6A CN112231977A (en) | 2020-10-15 | 2020-10-15 | Construction method of formula parameter library of redrying machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112231977A true CN112231977A (en) | 2021-01-15 |
Family
ID=74113069
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011101490.6A Pending CN112231977A (en) | 2020-10-15 | 2020-10-15 | Construction method of formula parameter library of redrying machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112231977A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115104753A (en) * | 2022-07-13 | 2022-09-27 | 四川中烟工业有限责任公司 | Cigarette cut-tobacco drier fault detection method and device based on twin neural network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101419209A (en) * | 2008-12-04 | 2009-04-29 | 哈尔滨工程大学 | Cigarette sensing appraise and flume index immune neural net prediction method |
| CN107862970A (en) * | 2017-11-20 | 2018-03-30 | 无锡开放大学 | A kind of teaching quality evaluation model for being used to overturn classroom |
| CN108107716A (en) * | 2017-12-19 | 2018-06-01 | 电子科技大学 | A kind of Parameter Measuring method based on improved BP neural network |
| 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 |
| CN110134710A (en) * | 2019-04-02 | 2019-08-16 | 杭州电子科技大学 | A method of cigarette composition Rule Extraction and composition maintenance based on Bayesian network |
-
2020
- 2020-10-15 CN CN202011101490.6A patent/CN112231977A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101419209A (en) * | 2008-12-04 | 2009-04-29 | 哈尔滨工程大学 | Cigarette sensing appraise and flume index immune neural net prediction method |
| CN107862970A (en) * | 2017-11-20 | 2018-03-30 | 无锡开放大学 | A kind of teaching quality evaluation model for being used to overturn classroom |
| CN108107716A (en) * | 2017-12-19 | 2018-06-01 | 电子科技大学 | A kind of Parameter Measuring method based on improved BP neural network |
| 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 |
| CN110134710A (en) * | 2019-04-02 | 2019-08-16 | 杭州电子科技大学 | A method of cigarette composition Rule Extraction and composition maintenance based on Bayesian network |
Non-Patent Citations (2)
| Title |
|---|
| 张强 等: "《配方模块制丝关键工艺参数的优化设计》", 《中国烟草学报》, vol. 19, no. 3, 30 June 2013 (2013-06-30), pages 28 - 34 * |
| 张强 等: "配方模块制丝关键工艺参数的优化设计", 《中国烟草学报》, vol. 19, no. 3, pages 28 - 34 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115104753A (en) * | 2022-07-13 | 2022-09-27 | 四川中烟工业有限责任公司 | Cigarette cut-tobacco drier fault detection method and device based on twin neural network |
| CN115104753B (en) * | 2022-07-13 | 2023-01-20 | 四川中烟工业有限责任公司 | Cigarette cut-tobacco drier fault detection method and device based on twin neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107730031B (en) | Ultra-short-term peak load prediction method and system | |
| CN111639823B (en) | Building cold and heat load prediction method constructed based on feature set | |
| CN111887460A (en) | Tobacco cut-tobacco drying moisture and temperature control prediction system and method | |
| CN113017132A (en) | Cut tobacco quality optimization method based on cut tobacco dryer process parameter prediction | |
| CN113107626B (en) | Load prediction method of combined cycle generator set based on multivariable LSTM | |
| CN117421618B (en) | Building energy consumption monitoring method and system | |
| CN112263012B (en) | Moisture content control method of redrying machine based on formula parameter library | |
| CN112231977A (en) | Construction method of formula parameter library of redrying machine | |
| CN105334738B (en) | A kind of method of evaluating performance suitable for tobacco processing course pid control circuit | |
| CN113325721A (en) | Model-free adaptive control method and system for industrial system | |
| CN116384700A (en) | Energy efficiency optimization method for central air conditioning system based on data mining | |
| CN114115393A (en) | Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line | |
| 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 | |
| CN114139717A (en) | Cut tobacco drying inlet water content control method based on multivariate time sequence analysis prediction model | |
| CN114429238A (en) | Wind turbine generator fault early warning method based on space-time feature extraction | |
| CN116757354A (en) | Tobacco redrying section key parameter screening method based on multilayer perceptron | |
| CN113642238A (en) | Micrometeorological factor-based radial basis function neural network power transmission line icing prediction method | |
| CN110766234B (en) | Cement cooling process grate down pressure prediction method based on information fusion | |
| CN110991723A (en) | Application method of artificial intelligence in seasonal load prediction | |
| CN114997470A (en) | Short-term power load prediction method based on LSTM neural network | |
| CN114707684A (en) | Improved LSTM-based raw tobacco stack internal temperature prediction algorithm | |
| CN115251445B (en) | Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine | |
| Zhu et al. | Control method of outlet moisture content of tobacco primary processing based on TVA-TCN-MPC | |
| CN114934198B (en) | Method for preparing high-purity indium based on machine learning optimization vacuum distillation | |
| CN117369391A (en) | System and method for optimizing process parameters of end-edge cloud cooperation |
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 |