CN111144667A - Tobacco conditioner discharged material water content prediction method based on gradient lifting tree - Google Patents
Tobacco conditioner discharged material water content prediction method based on gradient lifting tree Download PDFInfo
- Publication number
- CN111144667A CN111144667A CN202010045510.6A CN202010045510A CN111144667A CN 111144667 A CN111144667 A CN 111144667A CN 202010045510 A CN202010045510 A CN 202010045510A CN 111144667 A CN111144667 A CN 111144667A
- Authority
- CN
- China
- Prior art keywords
- model
- data
- prediction
- water content
- tobacco
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
Abstract
The invention discloses a method for predicting the discharged material moisture content of a tobacco conditioner based on a gradient lifting tree, which belongs to the field of tobacco. The model can complete the prediction of the water content of the discharged material without manual experience intervention under the working environment of the actual moisture regaining process, and has high prediction result precision and high calculation speed. The prediction method can automatically calculate the discharged water content and the required water adding amount of the damping machine, and further more effectively control the temperature and the humidity of the tobacco leaves so as to improve the quality of the tobacco shreds.
Description
Technical Field
The invention belongs to the field of tobacco, and particularly relates to a method for predicting moisture content of discharged material of a tobacco conditioner based on a gradient lifting tree.
Background
The loosening and moisture regaining are important components of the tobacco production process, and play a key role in the quality and taste of cigarettes. With the improvement of automation and intellectualization level in the tobacco industry, the requirement on the discharge temperature and humidity control of the loosening and conditioning machine is further improved. However, in actual operation, due to the large number of variables and complex interrelation, it is difficult to establish the moisture content of the discharged material that is effectively conditioned by the discharged material moisture content control method, depending on manual adjustment by an operator using experience or a simple prediction model, such as grouping products based on experience, and using a prediction method based on linear regression or a neural network. The water adding strategy of Anhui Zhongyan is improved by grouping tobacco leaves, and Fujian Zhongyan and Henan Zhongyan use the environmental temperature and humidity and partial process parameters respectively to establish a multiple regression model to predict the water adding amount of loosening and conditioning. An engineer of the cigarette in Guizhou predicts the moisture content of the discharged material of the damping machine based on the initial moisture content and the water adding amount of the material by using a two-parameter non-exponential empirical model equation provided by Peleg. An engineer of Fujian Longyan tobacco industry Limited liability company uses an Elman neural network, the characteristics of the Elman neural network and the excellent stability are fully utilized, and tobacco in Shandong inputs real-time data loose moisture regaining discharging water content and tobacco moistening charging discharging water content of a tobacco making section into a model by utilizing a neural network model based on a radial basis function, so that the cut tobacco raw silk moisture content value of a cut tobacco dryer is predicted. According to practical results, the method of linear regression and neural network is used for predicting and guiding the control of the water adding amount, so that the method has a good effect, and the quality of products can be effectively improved in actual production.
However, the method based on linear regression fails to fully consider and integrate the complexity of the actual production process of tobacco moisture regain, uses a traditional process combination improvement or a simpler prediction method such as linear regression and the like, has insufficient processing capacity on the complexity of a system, and fails to fully consider the common influence of environmental parameters and a production process on the moisture content of discharged materials. The calculated result has limited precision and cannot be reliably used in the prediction of the moisture content of the discharged material of the damping machine.
In the early research of the dampening machine, the method is limited by hardware conditions and data acquisition capacity, the types, data quantity and granularity of parameters for model calculation and fitting are insufficient, and the method is not beneficial to establishing a complex and reliable dampening model. Meanwhile, the maturity of the mathematical theory, particularly the machine learning theory, and the research level facing the industrial production process are low, and a solution combining a complex machine learning model and a moisture regaining process is lacked.
Disclosure of Invention
The method uses the complete process parameters and environment parameters of the damping machine as independent variables, and utilizes the gradient lifting tree algorithm to establish a damping prediction model, fully considers the influence of different influence factors on the discharged material moisture content of the tobacco damping machine, and simultaneously enhances the prediction and interpretation capability of the prediction model on the nonlinear characteristics of the damping process. The model can complete the prediction of the water content of the discharged material without manual experience intervention under the working environment of the actual moisture regaining process, and has high prediction result precision and high calculation speed. The prediction method can automatically calculate the discharged water content and the required water adding amount of the damping machine, and further more effectively control the temperature and the humidity of the tobacco leaves so as to improve the quality of the tobacco shreds.
In order to realize the purpose, the invention is realized by adopting the following technical scheme: the method for predicting the discharged water content of the tobacco conditioner based on the gradient lifting tree is realized by adopting the following steps:
step 1, acquiring data of a production process and environmental parameters;
step 2, preprocessing the data, deleting abnormal values and aligning time points of different parameters;
step 3, dividing the data into a training set and a test set;
step 4, training different discharging water content prediction models by using the training set, and generating the prediction models by using four algorithms of regression, a support vector machine, a neural network and a gradient lifting tree respectively;
and 5, verifying and comparing the difference between the prediction result and the true value of the model on the test set, and selecting the optimal prediction algorithm according to the comparison between the service requirement and the statistical index.
Preferably, the method for predicting the discharged water content of the tobacco conditioner based on the gradient lifting tree comprises a data acquisition unit, a data preprocessing unit, a discharged water content model training unit, a discharged water content model evaluation unit and a discharged water content model performance reporting unit.
Preferably, the step 1 is a data acquisition link, important parameters of the damping machine on each production line are acquired, and the acquired data cover production processes and environmental parameters, including the tobacco leaf inlet water content, the silk material flow, the sheet material flow, the hot air temperature, the water adding flow, the outlet humidity, the outlet temperature, the time point, the ambient temperature and the ambient humidity around the machine, and the brand of the produced product.
Preferably, the step 2 is a data preprocessing step, and for missing data points of each process parameter, firstly, an interpolation mode is used for completion; then screening and removing data points which obviously deviate from the normal operation range based on the production requirements and industry experience of the damping machine; the data is then resampled to obtain data with a fixed time interval.
Preferably, the step 4 is a model training and tuning link, and the sub-data meeting the data condition is input into the model training link, and the characteristics as independent variables include all key parameters of the dampening process: the method comprises the steps of tobacco leaf inlet water content, wire material flow, sheet material flow, hot air temperature, water adding flow, ambient temperature and ambient humidity around a machine, and the dependent variable serving as a prediction target is the discharged water content of the outlet of a damping process machine.
Preferably, the algorithm adopted in the step 4 comprises a plurality of regression and machine learning algorithms such as gradient lifting tree, support vector machine, neural network and linear regression; in the selection and optimization of the model, a prediction error threshold value determined by the process is set, iterative calculation is carried out based on algorithm development logic, and convergence is carried out after the process requirement and the statistical error index are met; in the algorithm optimization process, a plurality of candidate prediction models are calculated at the same time, the Root Mean Square Error (RMSE), the average absolute percentage error (MAPE), the goodness-of-fit (R-square) and the percentage deviation of each model are calculated, the model with the best prediction effect is selected after comparison, namely, the model with the smaller percentage deviation and error but the larger goodness-of-fit is selected as the optimal model, and the gradient lifting tree algorithm is determined as the optimal algorithm for generating the prediction model through comparison; the resulting model is stored in a memory location for constant recall.
The invention has the beneficial effects that:
1, establishing a set of prediction models based on the operation characteristics of the damping machine and a gradient lifting tree algorithm to describe the relation of different variables in a complex system formed by outlet control variables, materials and environmental parameters of the damping machine without controlled experimental environment measurement; and 2, the accuracy and the reliability of the prediction result of the production and operation process of the damping machine are improved, the water content of the discharged material can be predicted and calculated on line in real time, and the prediction accuracy and the production process level can be evaluated.
Drawings
FIG. 1 is a system block diagram;
FIG. 2 data processing and prediction flow.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
As shown in the figure 1-2, the method aims at the defects of accurate prediction and control capability of the discharged water content in the tobacco moisture regaining process. The method comprises the steps of firstly, acquiring data of production process technology and environmental parameters from a factory production management system; then preprocessing the data, deleting abnormal values and aligning time points of different parameters; dividing the data into a training set and a testing set, wherein the training set is used for training different discharging water content prediction models, and the prediction models are generated by four algorithms of regression, a support vector machine, a neural network and a gradient lifting tree respectively; and finally, verifying and comparing the difference between the prediction result and the true value of the model on the test set, and selecting the optimal prediction algorithm according to the comparison between the service requirement and the statistical index.
A data acquisition unit: and acquiring process parameters and environmental parameters of the damping machine on each production line, wherein the process parameters and the environmental parameters comprise the water content of inlet tobacco leaves, the flow of wire materials, the flow of sheets, the temperature of hot air, the flow of water, the humidity of an outlet, the temperature of the outlet, data acquisition time points, the environmental temperature and the environmental humidity around the machine, the brand of a produced product and the like. Compared with the prior art that only process parameters or environment parameters are used and the working mechanism of the damping machine cannot be effectively considered, the data acquisition unit has the advantages that the parameter types of the data acquired by the data acquisition unit are greatly improved, the parameters related in the working process of the damping machine are effectively integrated into a model, and the mechanism of the damping process is better mastered.
A data preprocessing unit: according to the understanding of the mechanism of the production process, data points deviating from the normal temperature and humidity range are screened and removed, and meanwhile time points of the temperature and humidity and process parameters are aligned, so that the time intervals of the data are kept consistent, and the time points with temperature and humidity value loss are supplemented in an interpolation mode. From the purposes and results of data preprocessing, the time interval of data processed by the scheme of the invention is shorter, so that the data volume is larger, the information which can be contained is more, and the model is favorable for covering more production information; the abnormal data is removed according to the production process, and the real situation of the moisture regaining process is better reflected.
And judging whether the data scale meets the requirements or not for different production lines, and then respectively modeling.
The data requirements comprise that the data volume generated by the moisture regaining process is not lower than the modeling requirement in the process duration period of the collected data, namely the requirements of the data on time span and time length are met.
Discharged water content model training unit: the core of the method is that a gradient lifting tree algorithm is used for fitting a characteristic variable and a target, and a prediction model of the moisture content of the discharged material is established. Firstly, taking production process and environmental parameters which play a key role in the moisture regain process as independent variables, wherein the independent variables comprise the water content of a tobacco leaf inlet, the flow of wires, the flow of sheets, the temperature of hot air, the flow of added water, the ambient temperature and the ambient humidity around a machine; and then, taking the discharged water content at the outlet of the conditioning process machine as a prediction target, and training a prediction model. In the training process of the model, after a prediction error threshold value based on the process is manually set, iterative calculation is carried out by using an algorithm until the error requirement is met, and then convergence is carried out. In the model training process, four different prediction algorithms are used to generate the model, including support vector machine, neural network, linear regression and gradient lifting tree algorithm. And then, the existing technology is used as a reference model for comparison, and the advantages of the gradient lifting model as an optimal model are evaluated through precision evaluation of the four models. And then storing the obtained model in a storage unit, and calling the model and using the model for predicting the discharged water content of other moisture regaining data. Compared with the prior art, the prediction algorithm used by the invention better captures the working mechanism of the tobacco conditioner, integrates the comprehensive influence of the process parameters and the environmental parameters, has higher complexity and stronger capability of processing nonlinear problems. The method comprises comparison and verification of various prediction algorithms, selects an optimal method, uses various indexes including production process requirements and statistical indexes in the comparison process, is more comprehensive than the evaluation indexes in the prior art, and can give consideration to both the production process requirements and the machine learning evaluation method.
The discharged water content model evaluation unit: firstly, randomly splitting data according to a set proportion, and automatically splitting a data set into a training set and a testing set. Then, based on the data of the training set, respectively training a prediction model by using different algorithms, then applying the model to the working conditions and parameter conditions of the test set, calculating a predicted value of the discharged material water content of the test set, and then calculating the difference between the predicted value and the true value. In the model evaluation process, the Root Mean Square Error (RMSE), the Mean Absolute Percent Error (MAPE), the goodness of fit (R-square) and the percentage deviation of each model are respectively compared, and an algorithm with small percentage deviation and error but large goodness of fit is taken as an optimal algorithm.
And a discharged water content model performance reporting unit: and respectively generating an evaluation report according to the calculation result of the training set and the verification result of the test set. The report includes, but is not limited to, the following two basic contents:
the evaluation indexes of the prediction algorithm are applied to the mutual comparison of different batches of products in the same dampening process flow production line, and the product with poor indexes shows that the prediction result of the discharged water content is poor, so that the difficulty of accurate control is improved, and the control accuracy is reduced along with the reduction of the prediction accuracy. And listing the calculation results of different batches of products of the same moisture regain technological process production line at the current evaluation time.
The evaluation index of the prediction algorithm can be applied to different batches of products in the same dampening process flow production line, and the comparison on the change along with the time is analyzed, if the index presents an obvious downward trend, the moisture regulation capability of the water heating system of the dampening machine is shown to be degraded. And (4) listing the calculation results of all batches of products of all the moisture regaining technological process production lines.
A data acquisition step: the link acquires important parameters of the damping machine on each production line, and the acquired data covers the production process and environmental parameters, including the tobacco leaf inlet water content, the silk material flow, the sheet flow, the hot air temperature, the water adding flow, the outlet humidity, the outlet temperature, the time point, the ambient temperature and the ambient humidity around the machine, and the brand of the produced product. After the data acquisition is finished, the data can be stored as an Excel file or other data files which can be rapidly read, and can also be directly imported into a data preprocessing link.
A data preprocessing link: for missing data points of each process parameter, firstly, completing the data points in an interpolation mode; then screening and removing data points which obviously deviate from the normal operation range based on the production requirements and industry experience of the damping machine; the data is then resampled to obtain data with a fixed time interval. And aligning the time points of the temperature and the humidity and the process parameters, adjusting the time interval of the data of the environmental temperature and the humidity to be consistent with the process parameters, and supplementing the time points of the missing values of the temperature and the humidity through interpolation measures. After the data preprocessing is finished, the data for modeling should meet the following requirements of time span and data length of the data generated by the conditioning process within the process duration period of the collected data: the time span should have a data size of not less than 1 month, or the fixed data length should reach or exceed 100000 to cover a wider range of ambient temperature and humidity and process flow data complexity.
Model training and tuning links: inputting the subdata meeting the data conditions into a model training link, wherein the characteristics serving as independent variables comprise all key parameters of the dampening process: the method comprises the steps of tobacco leaf inlet water content, wire material flow, sheet material flow, hot air temperature, water adding flow, ambient temperature and ambient humidity around a machine, and the dependent variable serving as a prediction target is the discharged water content of the outlet of a damping process machine. The used algorithms comprise various regression and machine learning algorithms such as gradient lifting trees, support vector machines, neural networks, linear regression and the like. In the selection and optimization of the model, a prediction error threshold value determined by the process is set, iterative calculation is carried out based on algorithm development logic, and convergence is carried out after process requirements and statistical error indexes are met. In the algorithm optimization process, calculation is simultaneously carried out on a plurality of candidate prediction models, and the Root Mean Square Error (RMSE), the Mean Absolute Percentage Error (MAPE), the goodness-of-fit (R-square) and the percentage deviation of each model are calculated. And selecting the model with the best prediction effect after comparison, namely selecting the model with smaller percentage deviation and error but larger fitting goodness as the optimal model, and determining the gradient lifting tree algorithm as the optimal algorithm for generating the prediction model through comparison. The resulting model is stored in a memory location for constant recall. The percentage deviation, goodness of fit, root mean square error, mean absolute percentage error, and time-consuming comparisons of the four models are shown in the table below.
According to the setting, automatically splitting the acquired data set into a training set and a testing set, and calculating the discharging water content predicted by the model under the condition of the working condition and parameter change of the testing set and the difference between the predicted value and the true value based on the model obtained in the training process.
And respectively generating an evaluation report according to the calculation result of the training set and the verification result of the test set. The report includes, but is not limited to, the following two basic contents.
1) The index can be applied to the mutual comparison of different batches of products on the same moisture regain process flow production line, and the product batch with poor index shows that the prediction result of the discharged water content is poor, so that the difficulty of accurate control is improved, the control accuracy is reduced, and the quality of tobacco leaves is reduced. And listing the calculation results and corresponding visual information of different batches of products in the same moisture regain process flow production line at the current evaluation time.
2) The index can be applied to different batches of products in the same dampening process flow production line, so that the comparison of the index along with the change of time is analyzed, and if the index presents an obvious downward trend, the process adjusting capability of the water adding and heating system of the dampening machine is degraded. And listing the calculation results and corresponding visual information of all batches of products of all the moisture regaining technological process production lines.
The main characteristics of the invention include: 1, establishing an integral method for predicting the discharged water content of a tobacco conditioner, and establishing a discharged water content prediction model of an outlet of the conditioner by using historical data of a tobacco loosening conditioner to describe the physical logics among factors such as the water content of a tobacco inlet, the flow of wires, the flow of sheets, the temperature of hot air, the flow of added water, the ambient temperature and the ambient humidity around the machine and the discharged water content; 2, simultaneously using the process parameters and the environmental parameters as input characteristics required by prediction, observing the change of the changed discharged material moisture content of different parameters under the environment represented by the parameters, and using the change as the representation of the tobacco leaf quality and the working capacity of the moisture regaining machine; 3, establishing a dampening prediction model based on a gradient lifting tree algorithm, wherein the performance of the dampening prediction model is superior to that of a traditional dampening prediction method, and the nonlinear relation and the complex relation among all parameters in the running process of the dampening machine can be better captured; and 4, real-time training, real-time verification and evaluation, dynamic construction of the outlet state of the damping machine, the relation between the environment and a complex system formed by materials, and more real reflection of the working capacity of the damping machine.
While the present invention has been described in detail with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, and various changes and modifications can be made without departing from the spirit and scope of the present invention as those skilled in the art will understand.
Claims (6)
1. A tobacco conditioner discharged material moisture content prediction method based on a gradient lifting tree is characterized by comprising the following steps: the method for predicting the discharged water content of the tobacco conditioner based on the gradient lifting tree is realized by adopting the following steps:
step 1, acquiring data of a production process and environmental parameters;
step 2, preprocessing the data, deleting abnormal values and aligning time points of different parameters;
step 3, dividing the data into a training set and a test set;
step 4, training different discharging water content prediction models by using the training set, and generating the prediction models by using four algorithms of regression, a support vector machine, a neural network and a gradient lifting tree respectively;
and 5, verifying and comparing the difference between the prediction result and the true value of the model on the test set, and selecting the optimal prediction algorithm according to the comparison between the service requirement and the statistical index.
2. The method for predicting the discharged water content of the tobacco conditioner based on the gradient lifting tree is characterized by comprising a data acquisition unit, a data preprocessing unit, a discharged water content model training unit, a discharged water content model evaluation unit and a discharged water content model performance reporting unit.
3. The method for predicting the moisture content of the discharged material of the tobacco conditioner based on the gradient lifting tree as claimed in claim 1, wherein the method comprises the following steps: the step 1 is a data acquisition link, important parameters of the moisture regaining machine on each production line are acquired, and the acquired data cover production processes and environmental parameters, including the tobacco leaf inlet water content, the silk material flow, the sheet flow, the hot air temperature, the water adding flow, the outlet humidity, the outlet temperature, the time point, the environmental temperature and the environmental humidity around the machine, and the brand of the produced product.
4. The method for predicting the moisture content of the discharged material of the tobacco conditioner based on the gradient lifting tree as claimed in claim 1 or 3, wherein the method comprises the following steps: the step 2 is a data preprocessing link, and for missing data points of each process parameter, firstly, an interpolation mode is used for completion; then screening and removing data points which obviously deviate from the normal operation range based on the production requirements and industry experience of the damping machine; the data is then resampled to obtain data with a fixed time interval.
5. The method for predicting the moisture content of the discharged material of the tobacco conditioner based on the gradient lifting tree as claimed in claim 4, wherein the method comprises the following steps: and 4, a model training and tuning link, wherein the subdata meeting the data conditions is input into the model training link and is used as the independent variable characteristic including all key parameters of the dampening process: the method comprises the steps of tobacco leaf inlet water content, wire material flow, sheet material flow, hot air temperature, water adding flow, ambient temperature and ambient humidity around a machine, and the dependent variable serving as a prediction target is the discharged water content of the outlet of a damping process machine.
6. The method for predicting the moisture content of the discharged material of the tobacco conditioner based on the gradient lifting tree as claimed in claim 5, wherein the method comprises the following steps: the algorithm adopted in the step 4 comprises a plurality of regression and machine learning algorithms such as a gradient lifting tree, a support vector machine, a neural network, a linear regression and the like; in the selection and optimization of the model, a prediction error threshold value determined by the process is set, iterative calculation is carried out based on algorithm development logic, and convergence is carried out after the process requirement and the statistical error index are met; in the algorithm optimization process, a plurality of candidate prediction models are calculated at the same time, the Root Mean Square Error (RMSE), the average absolute percentage error (MAPE), the goodness-of-fit (R-square) and the percentage deviation of each model are calculated, the model with the best prediction effect is selected after comparison, namely, the model with the smaller percentage deviation and error but the larger goodness-of-fit is selected as the optimal model, and the gradient lifting tree algorithm is determined as the optimal algorithm for generating the prediction model through comparison; the resulting model is stored in a memory location for constant recall.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010045510.6A CN111144667A (en) | 2020-01-16 | 2020-01-16 | Tobacco conditioner discharged material water content prediction method based on gradient lifting tree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010045510.6A CN111144667A (en) | 2020-01-16 | 2020-01-16 | Tobacco conditioner discharged material water content prediction method based on gradient lifting tree |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111144667A true CN111144667A (en) | 2020-05-12 |
Family
ID=70525302
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010045510.6A Pending CN111144667A (en) | 2020-01-16 | 2020-01-16 | Tobacco conditioner discharged material water content prediction method based on gradient lifting tree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111144667A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111950795A (en) * | 2020-08-18 | 2020-11-17 | 安徽中烟工业有限责任公司 | Random forest based method for predicting water adding proportion of loosening and conditioning |
| CN112021626A (en) * | 2020-07-10 | 2020-12-04 | 张家口卷烟厂有限责任公司 | Intelligent control system and method for tobacco shred making link |
| CN112800671A (en) * | 2021-01-26 | 2021-05-14 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
| CN112884215A (en) * | 2021-02-02 | 2021-06-01 | 国网甘肃省电力公司信息通信公司 | Parameter optimization method based on gradient enhancement tree population prediction model |
| CN112947342A (en) * | 2021-02-26 | 2021-06-11 | 四川中烟工业有限责任公司 | Data-driven tobacco raw silk moisture control system and control method |
| CN113076309A (en) * | 2021-03-26 | 2021-07-06 | 四川中烟工业有限责任公司 | System and method for predicting water adding amount of raw tobacco shred |
| CN113080499A (en) * | 2021-02-26 | 2021-07-09 | 红云红河烟草(集团)有限责任公司 | Method for controlling temperature of loose moisture regaining hot air by Q-Learning algorithm based on strategy |
| CN113934185A (en) * | 2021-10-09 | 2022-01-14 | 北京远舢智能科技有限公司 | Control method and device for improving process quality and electronic equipment |
| CN114035629A (en) * | 2021-10-20 | 2022-02-11 | 科大讯飞股份有限公司 | Control method of drying equipment and related equipment |
| CN114668164A (en) * | 2022-04-01 | 2022-06-28 | 河南中烟工业有限责任公司 | Loose moisture regain water volume adaptive control system based on supplied material difference |
| CN111950795B (en) * | 2020-08-18 | 2024-04-26 | 安徽中烟工业有限责任公司 | Random forest-based prediction method for loosening and conditioning water adding proportion |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108652066A (en) * | 2018-05-31 | 2018-10-16 | 福建中烟工业有限责任公司 | The water feeding method of loosening and gaining moisture process and the device for predicting the process amount of water |
| 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 |
| CN110245802A (en) * | 2019-06-20 | 2019-09-17 | 杭州安脉盛智能技术有限公司 | Based on the cigarette void-end rate prediction technique and system for improving gradient promotion decision tree |
-
2020
- 2020-01-16 CN CN202010045510.6A patent/CN111144667A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108652066A (en) * | 2018-05-31 | 2018-10-16 | 福建中烟工业有限责任公司 | The water feeding method of loosening and gaining moisture process and the device for predicting the process amount of water |
| 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 |
| CN110245802A (en) * | 2019-06-20 | 2019-09-17 | 杭州安脉盛智能技术有限公司 | Based on the cigarette void-end rate prediction technique and system for improving gradient promotion decision tree |
Non-Patent Citations (2)
| Title |
|---|
| 崔升: "烟叶松散率影响因素研究", 《广西科学》 * |
| 陈晓杜 等: "基于Elman神经网络的卷烟制丝松散回潮出口含水率控制方法", 《安徽农学通报》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112021626A (en) * | 2020-07-10 | 2020-12-04 | 张家口卷烟厂有限责任公司 | Intelligent control system and method for tobacco shred making link |
| CN112021626B (en) * | 2020-07-10 | 2021-08-17 | 张家口卷烟厂有限责任公司 | Intelligent control system and method for tobacco shred making link |
| CN111950795A (en) * | 2020-08-18 | 2020-11-17 | 安徽中烟工业有限责任公司 | Random forest based method for predicting water adding proportion of loosening and conditioning |
| CN111950795B (en) * | 2020-08-18 | 2024-04-26 | 安徽中烟工业有限责任公司 | Random forest-based prediction method for loosening and conditioning water adding proportion |
| CN112800671A (en) * | 2021-01-26 | 2021-05-14 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
| CN112884215A (en) * | 2021-02-02 | 2021-06-01 | 国网甘肃省电力公司信息通信公司 | Parameter optimization method based on gradient enhancement tree population prediction model |
| CN113080499A (en) * | 2021-02-26 | 2021-07-09 | 红云红河烟草(集团)有限责任公司 | Method for controlling temperature of loose moisture regaining hot air by Q-Learning algorithm based on strategy |
| CN112947342B (en) * | 2021-02-26 | 2024-03-12 | 四川中烟工业有限责任公司 | Data-driven tobacco raw silk moisture control system and control method |
| CN112947342A (en) * | 2021-02-26 | 2021-06-11 | 四川中烟工业有限责任公司 | Data-driven tobacco raw silk moisture control system and control method |
| CN113076309A (en) * | 2021-03-26 | 2021-07-06 | 四川中烟工业有限责任公司 | System and method for predicting water adding amount of raw tobacco shred |
| CN113934185A (en) * | 2021-10-09 | 2022-01-14 | 北京远舢智能科技有限公司 | Control method and device for improving process quality and electronic equipment |
| CN114035629A (en) * | 2021-10-20 | 2022-02-11 | 科大讯飞股份有限公司 | Control method of drying equipment and related equipment |
| CN114668164A (en) * | 2022-04-01 | 2022-06-28 | 河南中烟工业有限责任公司 | Loose moisture regain water volume adaptive control system based on supplied material difference |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111144667A (en) | Tobacco conditioner discharged material water content prediction method based on gradient lifting tree | |
| CN109222208B (en) | Cut tobacco making process analysis optimization method and system oriented to cigarette production index control | |
| CN109674080B (en) | Tobacco leaf conditioning water adding amount prediction method, storage medium and terminal equipment | |
| CN112021626B (en) | Intelligent control system and method for tobacco shred making link | |
| CN111461555A (en) | Production line quality monitoring method, device and system | |
| CN112384924A (en) | Method and device for establishing product performance prediction model, computer equipment, computer readable storage medium, product performance prediction method and prediction system | |
| CN113017132A (en) | Cut tobacco quality optimization method based on cut tobacco dryer process parameter prediction | |
| CN108133391A (en) | Method for Sales Forecast method and server | |
| CN112273696B (en) | Method, device and equipment for controlling moisture after shredding | |
| CN108652066A (en) | The water feeding method of loosening and gaining moisture process and the device for predicting the process amount of water | |
| CN111103854A (en) | System and method for improving production stability of tobacco cut-tobacco drier | |
| CN112273695A (en) | Method, device and equipment for predicting water content of loose moisture regain outlet | |
| CN114115393A (en) | Method for controlling moisture and temperature at outlet of cut tobacco dryer for sheet cut tobacco making line | |
| CN115936262B (en) | Yield prediction method, system and medium based on big data environment interference | |
| CN117008557B (en) | Production control method and system for blending type interpenetrating network thermoplastic elastomer | |
| CN110876481B (en) | Control method and device for tobacco shred drying parameters | |
| CN112132316A (en) | System and method for monitoring abnormality of on-line equipment in silk making link | |
| CN111428329A (en) | Model-based machine learning system | |
| CN116602435A (en) | Machine learning-based method for analyzing moisture change of cut tobacco in cut tobacco making machine | |
| CN115169737A (en) | Process quality prediction method based on CNN-LSTM hybrid neural network model | |
| CN111849544B (en) | Hydrocracking product quality automatic control method, device and storage | |
| CN116757354A (en) | Tobacco redrying section key parameter screening method based on multilayer perceptron | |
| CN111165866A (en) | Quality control method and system based on airflow type cut stem drying | |
| CN116090349A (en) | Optical film production process optimization method, equipment and storage medium | |
| US20230221687A1 (en) | A system and method for evaluation of sand compactibility |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200512 |