Cigarette making machine parameter deep learning autonomous optimization system and method based on multiple factors
By building a deep learning autonomous optimization system for cigarette making machine parameters based on multiple factors, the cigarette making machine process parameters can be adjusted in real time, solving the problem of quality instability of cigarette making machines in dynamic environments and improving production efficiency and cigarette quality.
Patent Information
- Application Number
- CN202511012657.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-22
- Publication Date
- 2025-09-16
AI Technical Summary
Traditional cigarette-making machines lack the ability to adaptively adjust to changes in the physical properties of cigarette paper and tipping paper, tobacco characteristics, and ambient temperature and humidity, resulting in unstable cigarette quality and low production efficiency.
A multi-factor-based deep learning autonomous optimization system for cigarette making machine parameters is adopted. The data acquisition module collects cigarette materials, tobacco characteristics and environmental data in real time, builds a long-short-term memory network model, and automatically adjusts the cigarette making machine process parameters.
It achieves the stability of cigarette quality and the improvement of production efficiency, reduces manual intervention, and ensures the efficient production of cigarette making machines in a dynamic environment.
Smart Images
Figure CN120654738A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of cigarette production, and more specifically, to a system and method for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors. Background Art
[0002] During cigarette production, the physical properties of cigarette paper and tipping paper, such as tensile strength, air permeability, and basis weight, have a critical impact on the final shape and quality of cigarettes. Furthermore, tobacco moisture content, whole tobacco fraction, broken tobacco fraction, and filler content also significantly influence cigarette quality. Furthermore, fluctuations in temperature and humidity within the production environment can also affect the stability of cigarette quality.
[0003] Traditional cigarette-making machines lack effective adaptive adjustment capabilities when faced with the dynamic changes of these factors, and usually require manual intervention to adjust process parameters. This is not only inefficient, but also makes it difficult to ensure the consistency and stability of cigarette quality.
[0004] Therefore, there is an urgent need for a system and method for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors. Summary of the Invention
[0005] The purpose of the present invention is to provide a system and method for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors to solve the problems in the above-mentioned prior art. It can automatically adjust the process parameters according to the physical properties of the cigarette materials, the characteristics of the tobacco, and the changes in the ambient temperature and humidity to ensure the quality of cigarettes.
[0006] The present invention provides a system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors, which includes:
[0007] Data acquisition module, used to collect physical property data of cigarette materials, tobacco characteristics data, ambient temperature and humidity data and corresponding cigarette making machine process parameters;
[0008] a deep learning model construction module for constructing a cigarette making machine parameter deep learning model based on a long short-term memory network based on the data collected by the data collection module, wherein the physical property data of the cigarette material, the tobacco characteristic data, and the ambient temperature and humidity data serve as inputs to the cigarette making machine parameter deep learning model, and the corresponding cigarette making machine process parameters serve as outputs of the cigarette making machine parameter deep learning model;
[0009] The cigarette making machine process parameter adjustment module is used to input the physical performance data, tobacco characteristic data and environmental temperature and humidity data of the current batch of cigarette materials collected in real time by the data acquisition module into the cigarette making machine parameter deep learning model to obtain the corresponding cigarette making machine process parameters, and adjust the cigarette making machine process parameters according to the output results of the cigarette making machine parameter deep learning model.
[0010] In the system for autonomous optimization of cigarette machine parameters based on deep learning of multiple factors, preferably, the data acquisition module includes:
[0011] Temperature and humidity collection unit, used to collect ambient temperature and humidity data in the production workshop;
[0012] The tobacco moisture meter is installed at the tobacco feed port and is used to detect the moisture content of each batch of tobacco online;
[0013] Tobacco structure detection device, used to detect the whole tobacco rate, broken tobacco rate, medium tobacco rate, long tobacco rate and short tobacco rate of each batch;
[0014] The filling value detector is installed in the quality inspection room and is used to detect the filling value data of each batch of tobacco and store the filling value data of tobacco in the database with batch as label;
[0015] The cigarette paper and tipping paper quality index detection unit is used to detect the quality indexes of cigarette paper and tipping paper.
[0016] In the system for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors as described above, preferably, the temperature and humidity acquisition unit includes a plurality of temperature and humidity sensors distributed in the production workshop.
[0017] In the above-mentioned system for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors, preferably, the quality indicators of the cigarette paper and tipping paper include at least one of tensile strength, air permeability, basis weight, and material manufacturer.
[0018] The cigarette paper and tipping paper quality index detection unit uses the barcodes carried by the cigarette paper and tipping paper as labels and stores them in a database.
[0019] The system for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors as described above, wherein preferably, the system for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors further comprises:
[0020] A data preprocessing module is used to preprocess the data collected by the data collection module. The data preprocessing module includes at least one of a data cleaning unit, a standardization processing unit and a missing value filling unit.
[0021] In the above-mentioned system for autonomous optimization of cigarette machine parameters based on deep learning of multiple factors, preferably, the deep learning model construction module is specifically used to divide the data collected by the data collection module into a training set, a validation set, and a test set according to a preset ratio, wherein:
[0022] The training set is used to train the update parameters of the model. The training set is trained using the mean square error loss function and the Adam optimization algorithm is used to iteratively update the model weights;
[0023] The validation set is used to evaluate the performance of the model during the training process;
[0024] The test set is used to finally test the performance of the model.
[0025] As described above, the system for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors, wherein preferably, the cigarette making machine process parameters include: cigarette making machine speed, wire return amount, wire reel speed, air chamber pressure, fan pressure, weight control rejection parameters, empty end rejection parameters, end density rejection parameters and circumference rejection parameters.
[0026] As described above, the system for autonomous optimization of cigarette machine parameters based on deep learning of multiple factors, wherein preferably, the cigarette machine process parameter adjustment module is specifically used to: deploy the trained deep learning model into the intelligent control system of the cigarette machine, so as to receive the data collected by the data acquisition module in real time through the intelligent control system, predict and quickly output process parameter adjustment instructions.
[0027] The present invention also provides a method for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors using the above system, comprising the following steps:
[0028] Collect physical property data of cigarette materials, tobacco characteristics data, ambient temperature and humidity data and corresponding cigarette making machine process parameters;
[0029] Based on the collected data, a long short-term memory network-based deep learning model for cigarette making machine parameters is constructed, wherein the physical property data of the cigarette material, the tobacco characteristics data, and the ambient temperature and humidity data serve as inputs to the deep learning model for cigarette making machine parameters, and the corresponding cigarette making machine process parameters serve as outputs of the deep learning model for cigarette making machine parameters;
[0030] The physical property data, tobacco characteristic data and ambient temperature and humidity data of the current batch of cigarette materials collected in real time are input into the cigarette making machine parameter deep learning model to obtain the corresponding cigarette making machine process parameters, and the cigarette making machine process parameters are adjusted according to the output results of the cigarette making machine parameter deep learning model.
[0031] The present invention provides a system and method for autonomous optimization of cigarette-making machine parameters based on deep learning of multiple factors, enabling the cigarette-making machine to automatically adjust process parameters according to changes in external environmental factors and through a deep learning algorithm, thereby ensuring the stability of cigarette quality, reducing manual intervention, and improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below with reference to the accompanying drawings, in which:
[0033] Figure 1 A structural block diagram of an embodiment of a system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors provided by the present invention;
[0034] Figure 2 This is a flow chart of an embodiment of the method for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors provided by the present invention. DETAILED DESCRIPTION
[0035] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The description of the exemplary embodiments is merely illustrative and is in no way intended to limit the present disclosure, its application, or use. The present disclosure can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to make the present disclosure thorough and complete and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that unless otherwise specifically stated, the relative arrangement of parts and steps, the composition of materials, numerical expressions, and numerical values set forth in these embodiments should be interpreted as being merely exemplary and not as limiting.
[0036] The terms "first," "second," and similar terms used in this disclosure do not indicate any order, quantity, or importance, but are simply used to distinguish different parts. Terms such as "include" or "comprising" mean that the elements preceding the term include the elements listed after the term, and do not exclude the possibility of also including other elements. Terms such as "upper," "lower," and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0037] In the present disclosure, when a specific component is described as being located between a first component and a second component, there may or may not be an intervening component between the specific component and the first component or the second component. When a specific component is described as being connected to another component, the specific component may be directly connected to the other component without an intervening component, or may not be directly connected to the other component but have an intervening component.
[0038] All terms (including technical or scientific terms) used in this disclosure have the same meaning as those understood by one of ordinary skill in the art to which this disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in, for example, general dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or highly formal sense, unless explicitly defined herein.
[0039] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.
[0040] like Figure 1 As shown, the system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors provided in this embodiment includes:
[0041] Data acquisition module 1, used to collect physical property data of cigarette materials, tobacco characteristics data, environmental temperature and humidity data and corresponding cigarette making machine process parameters;
[0042] A deep learning model construction module 2 is used to construct a cigarette making machine parameter deep learning model based on a long short-term memory network (LSTM) based on the data collected by the data collection module, wherein the physical property data of the cigarette material, the tobacco characteristic data, and the ambient temperature and humidity data serve as the input of the cigarette making machine parameter deep learning model, and the corresponding cigarette making machine process parameters serve as the output of the cigarette making machine parameter deep learning model;
[0043] The cigarette making machine process parameter adjustment module 3 is used to input the physical property data, tobacco characteristic data and environmental temperature and humidity data of the current batch of cigarette materials collected in real time by the data acquisition module into the cigarette making machine parameter deep learning model to obtain the corresponding cigarette making machine process parameters, and adjust the cigarette making machine process parameters according to the output results of the cigarette making machine parameter deep learning model.
[0044] The cigarette making machine process parameters include: cigarette making machine speed, wire return amount, wire reel speed, air chamber pressure, fan pressure, weight control rejection parameters, empty end rejection parameters, end density rejection parameters and circumference rejection parameters.
[0045] Furthermore, the data acquisition module includes:
[0046] A temperature and humidity acquisition unit, configured to acquire ambient temperature and humidity data within the production workshop. In one embodiment of the present invention, the temperature and humidity acquisition unit comprises a plurality of temperature and humidity sensors distributed within the production workshop.
[0047] A tobacco moisture meter is installed at the tobacco feed port to detect the moisture content of each batch of tobacco online. For example, the tobacco moisture meter uses near-infrared spectroscopy technology to detect the moisture content of tobacco;
[0048] A tobacco structure detection device is used to detect the whole tobacco rate, broken tobacco rate, medium tobacco rate, long tobacco rate, and short tobacco rate of each batch of tobacco. In one embodiment of the present invention, the tobacco structure detection device uses image recognition technology to achieve tobacco structure detection;
[0049] The filling value detector is installed in the quality inspection room and is used to detect the filling value data of each batch of tobacco. The filling value data of the tobacco is stored in the database with the batch as the label. In the specific implementation, when the cigarette making machine starts to use the current batch of tobacco, the filling value of the corresponding batch is directly extracted from the database;
[0050] The cigarette paper and tipping paper quality index detection unit is used to detect the quality indexes of cigarette paper and tipping paper.
[0051] Specifically, the quality indicators of the cigarette paper and tipping paper include at least one of tensile strength, air permeability, basis weight, and material manufacturer. The cigarette paper and tipping paper quality indicator detection unit uses the barcodes carried by the cigarette paper and tipping paper as tags and stores these indicators in a database. In a specific implementation, before using cigarette materials, the machine scans the barcodes carried by the cigarette paper and tipping paper using a barcode scanner, extracting basic information such as the corresponding material quality indicators and manufacturer. When the machine scans a new material, it simultaneously indicates that the previous material has been used up and the new material has begun to be used, and the extracted data will overwrite the previous material's data.
[0052] Furthermore, in some embodiments of the present invention, the data acquisition module is also used to collect quality evaluation results of cigarettes produced using the process parameters set at the time, including cigarette weight, circumference, draw resistance, etc.
[0053] Furthermore, in some embodiments of the present invention, the system for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors further includes:
[0054] A data preprocessing module is used to preprocess the data collected by the data collection module. The data preprocessing module includes at least one of a data cleaning unit, a standardization processing unit and a missing value filling unit.
[0055] The data cleaning unit is used to remove data points with obvious errors or serious omissions, and to screen out invalid data by setting a reasonable threshold range. In one embodiment of the present invention, the data cleaning unit can screen out the physical property data of cigarette materials, tobacco characteristics data, environmental temperature and humidity data, and cigarette making machine process parameters corresponding to cigarettes with unqualified quality evaluation results;
[0056] The normalization processing unit uses the Z-score normalization method to convert the data into a distribution with a mean of 0 and a standard deviation of 1, so that all features are on the same scale and avoid features with large values dominating the model training;
[0057] For a small number of missing values, the missing value filling unit uses mean, median interpolation or linear interpolation based on adjacent data points to fill in the missing values.
[0058] Furthermore, the deep learning model construction module 2 is specifically used to divide the data collected by the data acquisition module into a training set (for example, 70%), a validation set (for example, 15%), and a test set (for example, 15%) according to a preset ratio, wherein:
[0059] The training set is used to train the update parameters of the model. The training set adopts the mean square error (MSE) loss function for training, and uses the Adam optimization algorithm to iteratively update the model weights to continuously reduce the loss function value;
[0060] The validation set is used to evaluate the performance of the model during training to prevent overfitting;
[0061] The test set is used to finally test the performance of the model.
[0062] Among them, in the training process of the deep learning model, the number of training rounds is set, and by observing the changes in the loss values on the training set and the validation set, it is judged whether the model has converged or overfitted; and the data in the test set is used to evaluate the model performance; when it is found that the model effect is not good, the hyperparameters are adjusted, such as increasing the number of hidden units, changing the number of LSTM layers, adjusting the learning rate, etc., and the model is retrained until a satisfactory prediction effect is achieved.
[0063] Furthermore, the cigarette making machine process parameter adjustment module 3 is specifically configured to deploy the trained deep learning model into the cigarette making machine's intelligent control system, so that the intelligent control system receives data collected by the data acquisition module in real time, predicts, and rapidly outputs process parameter adjustment instructions. In one embodiment of the present invention, when a batch of shredded tobacco, cigarette paper, and tipping paper is used in a cigarette making machine, the data collected in real time by the data acquisition module is input into the cigarette making machine parameter deep learning model. The cigarette making machine parameter deep learning model rapidly matches and outputs corresponding cigarette making machine process parameters (including cigarette making machine speed, amount of returned tobacco, reel speed, air chamber pressure, fan pressure, weight control rejection parameter, short end rejection parameter, end end density rejection parameter, and circumference rejection parameter), and issues adjustment instructions to the cigarette making machine. In a specific implementation, the deep learning model for cigarette making machine parameters receives new inputs when changing cigarette paper, tipping paper, or tobacco batches, or when the ambient temperature and humidity change. Based on these changes, the model re-matches the new process parameters. The system also stores these data results for self-learning and optimization. This invention seamlessly integrates the deep learning model with the existing automated control and sensor systems of cigarette making machines.
[0064] The system for autonomous optimization of cigarette-making machine parameters based on deep learning of multiple factors provided by the embodiment of the present invention enables the cigarette-making machine to automatically adjust process parameters according to changes in external environmental factors and through deep learning algorithms, thereby ensuring the stability of cigarette quality, reducing manual intervention, and improving production efficiency and product quality.
[0065] like Figure 2 As shown, the method for deep learning autonomous optimization of cigarette making machine parameters based on multiple factors provided in this embodiment includes the following steps during actual execution:
[0066] Step S1: Collecting physical property data of cigarette materials, tobacco characteristics data, ambient temperature and humidity data, and corresponding cigarette making machine process parameters.
[0067] Step S2: Based on the collected data, a deep learning model of cigarette making machine parameters based on a long short-term memory network is constructed, wherein the physical property data of the cigarette material, the tobacco characteristic data and the ambient temperature and humidity data serve as the input of the deep learning model of cigarette making machine parameters, and the corresponding cigarette making machine process parameters serve as the output of the deep learning model of cigarette making machine parameters.
[0068] Step S3: input the physical property data, tobacco characteristic data and ambient temperature and humidity data of the current batch of cigarette materials collected in real time into the cigarette making machine parameter deep learning model to obtain the corresponding cigarette making machine process parameters, and adjust the cigarette making machine process parameters according to the output results of the cigarette making machine parameter deep learning model.
[0069] The method for autonomous optimization of cigarette-making machine parameters based on deep learning of multiple factors provided in an embodiment of the present invention enables the cigarette-making machine to automatically adjust process parameters according to changes in external environmental factors and through a deep learning algorithm, thereby ensuring the stability of cigarette quality, reducing manual intervention, and improving production efficiency and product quality.
[0070] Thus far, various embodiments of the present disclosure have been described in detail. To avoid obscuring the concept of the present disclosure, some details known in the art have not been described. Based on the above description, those skilled in the art can fully understand how to implement the technical solutions disclosed herein.
[0071] Although some specific embodiments of the present disclosure have been described in detail through examples, those skilled in the art will understand that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. Those skilled in the art will understand that the above embodiments may be modified or some technical features may be replaced with equivalents without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.
Claims
1. A system for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors, characterized by: include: Data acquisition module, used to collect physical property data of cigarette materials, tobacco characteristics data, ambient temperature and humidity data and corresponding cigarette making machine process parameters; a deep learning model construction module for constructing a cigarette making machine parameter deep learning model based on a long short-term memory network based on the data collected by the data collection module, wherein the physical property data of the cigarette material, the tobacco characteristic data, and the ambient temperature and humidity data serve as inputs to the cigarette making machine parameter deep learning model, and the corresponding cigarette making machine process parameters serve as outputs of the cigarette making machine parameter deep learning model; The cigarette making machine process parameter adjustment module is used to input the physical performance data, tobacco characteristic data and environmental temperature and humidity data of the current batch of cigarette materials collected in real time by the data acquisition module into the cigarette making machine parameter deep learning model to obtain the corresponding cigarette making machine process parameters, and adjust the cigarette making machine process parameters according to the output results of the cigarette making machine parameter deep learning model.
2. The system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors according to claim 1 is characterized in that: The data acquisition module includes: Temperature and humidity collection unit, used to collect ambient temperature and humidity data in the production workshop; The tobacco moisture meter is installed at the tobacco feed port and is used to detect the moisture content of each batch of tobacco online; Tobacco structure detection device, used to detect the whole tobacco rate, broken tobacco rate, medium tobacco rate, long tobacco rate and short tobacco rate of each batch; The filling value detector is installed in the quality inspection room and is used to detect the filling value data of each batch of tobacco and store the filling value data of tobacco in the database with batch as label; The cigarette paper and tipping paper quality index detection unit is used to detect the quality indexes of cigarette paper and tipping paper.
3. The system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors according to claim 2 is characterized in that: The temperature and humidity acquisition unit includes a plurality of temperature and humidity sensors distributed in the production workshop.
4. The system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors according to claim 2 is characterized in that: The quality indicators of the cigarette paper and tipping paper include at least one of tensile strength, air permeability, basis weight and material manufacturer. The cigarette paper and tipping paper quality index detection unit uses the barcodes carried by the cigarette paper and tipping paper as labels and stores them in a database.
5. The system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors according to claim 1 is characterized in that: The system for autonomous optimization of cigarette machine parameters based on deep learning of multiple factors also includes: A data preprocessing module is used to preprocess the data collected by the data collection module. The data preprocessing module includes at least one of a data cleaning unit, a standardization processing unit and a missing value filling unit.
6. The system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors according to claim 1 is characterized in that: The deep learning model construction module is specifically used to divide the data collected by the data collection module into a training set, a validation set and a test set according to a preset ratio, wherein: The training set is used to train the update parameters of the model. The training set is trained using the mean square error loss function and the Adam optimization algorithm is used to iteratively update the model weights; The validation set is used to evaluate the performance of the model during the training process; The test set is used to finally test the performance of the model.
7. The system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors according to claim 1 is characterized in that: The cigarette making machine process parameters include: cigarette making machine speed, wire return amount, wire reel speed, air chamber pressure, fan pressure, weight control rejection parameter, empty end rejection parameter, end density rejection parameter and circumference rejection parameter.
8. The system for deep learning and autonomous optimization of cigarette making machine parameters based on multiple factors according to claim 1 is characterized in that: The cigarette making machine process parameter adjustment module is specifically used to: deploy the trained deep learning model into the intelligent control system of the cigarette making machine, so as to receive the data collected by the data acquisition module in real time through the intelligent control system, predict and quickly output process parameter adjustment instructions.
9. A method for autonomous optimization of cigarette making machine parameters based on deep learning of multiple factors using the system according to any one of claims 1 to 8, characterized in that: include: Collect physical property data of cigarette materials, tobacco characteristics data, ambient temperature and humidity data and corresponding cigarette making machine process parameters; Based on the collected data, a long short-term memory network-based deep learning model for cigarette making machine parameters is constructed, wherein the physical property data of the cigarette material, the tobacco characteristics data, and the ambient temperature and humidity data serve as inputs to the deep learning model for cigarette making machine parameters, and the corresponding cigarette making machine process parameters serve as outputs of the deep learning model for cigarette making machine parameters; The physical property data, tobacco characteristic data and ambient temperature and humidity data of the current batch of cigarette materials collected in real time are input into the cigarette making machine parameter deep learning model to obtain the corresponding cigarette making machine process parameters, and the cigarette making machine process parameters are adjusted according to the output results of the cigarette making machine parameter deep learning model.