CN112180832A - Artificial intelligent wine picking device and wine picking method thereof - Google Patents
Artificial intelligent wine picking device and wine picking method thereof Download PDFInfo
- Publication number
- CN112180832A CN112180832A CN202011218685.9A CN202011218685A CN112180832A CN 112180832 A CN112180832 A CN 112180832A CN 202011218685 A CN202011218685 A CN 202011218685A CN 112180832 A CN112180832 A CN 112180832A
- Authority
- CN
- China
- Prior art keywords
- wine
- data
- layer
- picking
- artificial intelligence
- 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
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/05—Programmable logic controllers, e.g. simulating logic interconnections of signals according to ladder diagrams or function charts
- G05B19/054—Input/output
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/10—Plc systems
- G05B2219/11—Plc I-O input output
- G05B2219/1103—Special, intelligent I-O processor, also plc can only access via processor
Abstract
The invention discloses an artificial intelligent wine picking device, which belongs to the technical field of wine picking devices and comprises a wine retort, wherein the lower end of the wine retort is provided with a pressure sensor, the lower end of the side edge of the wine retort is provided with a pipeline, a wine picking bowl is arranged below the pipeline, an online component analysis device is arranged inside the wine picking bowl, and a conduit at the lower end of the side edge of the wine picking bowl is provided with an electromagnetic valve. Improve the wine picking efficiency and accuracy and avoid human errors.
Description
Technical Field
The invention belongs to the technical field of wine picking devices, and particularly relates to an artificial intelligence wine picking device and a wine picking method thereof.
Background
At present, the traditional liquor picking mode in the liquor industry adopts a method of picking liquor by observing flowers. The operation mode of picking the wine by watching the flowers is that the alcohol content of the wine flowing at the moment is judged by observing the size, the shape and the duration of the hops after the white wine flows out by an operator with eyes, and three sections of the head, the body and the tail are distinguished.
The prior art has the following problems: 1. the method for watching flowers and picking wine completely depends on the experience of operators, is greatly influenced by the subjectivity of the operators, cannot form a unified assessment method and standard, and is difficult to inherit and short in talents because the inheritance method completely depends on the fact that teachers carry brothers;
2. the hops are influenced by temperature and pressure, and the higher the alcohol content is, the faster the hops are dissipated, so that the hops are easily influenced by ambient environmental factors and the detection speed of operators, the wine picking efficiency and accuracy are not high, and the disadvantage of human errors is inevitable;
3. the judgment of the wine cannot be accurately finished through the single factor of the hop, and other factors such as alcohol concentration, fermentation time, chemical components, container pressure and the like all affect the quality of the wine;
in conclusion, the traditional liquor picking process is operated manually to watch flowers and pick liquor at present, the quality of the picked liquor cannot be guaranteed, the production period is long, the efficiency is low, and the dependence on liquor pickers is strong.
Disclosure of Invention
To solve the problems set forth in the background art described above. The invention provides an artificial intelligent wine picking device which has the characteristics of simple structure, convenience in wine picking and high efficiency.
The invention also aims to provide a method for picking wine by means of the artificial intelligence wine picking device.
In order to achieve the purpose, the invention provides the following technical scheme: the utility model provides an artificial intelligence plucks wine device, includes the wine rice steamer, the lower extreme of wine rice steamer is provided with pressure sensor, the side lower extreme of wine rice steamer is provided with the pipeline, the below of pipeline is provided with plucks the wine bowl, the inside of plucking the wine bowl is provided with online composition analytical equipment, the side lower extreme pipe of plucking the wine bowl is provided with the solenoid valve, the side of pipeline is provided with the industrial computer, the side of industrial computer and pipeline lower extreme position are provided with online thermal imager, the lower extreme of online thermal imager is provided with the camera.
Furthermore, the side of industrial computer is provided with the PLC controller.
Furthermore, the pressure sensor, the online thermal imager, the camera, the online component analysis device, the electromagnetic valve and the industrial personal computer are all electrically connected with the PLC.
Furthermore, a plurality of raw wine collectors are arranged below the electromagnetic valve.
Furthermore, the camera uses a laser camera to automatically emit laser without a light source for lighting.
Furthermore, a timer used for timing is arranged inside the industrial personal computer.
Further, the wine picking method of the artificial intelligent wine picking device comprises the following steps:
firstly, data acquisition, namely selecting a plurality of different wine retorts, dynamically acquiring a hop photo, wine retort pressure, fermentation time, temperature data, composition data (including acetaldehyde, ethyl acetate, methanol, fusel oil, ethyl linoleate, ethyl oleate, ethyl palmitate and other concentrations) and corresponding raw wine grades in various states in the whole wine picking process;
classifying the data into a neural network learning part and a testing part;
thirdly, model training: the collected hop state, temperature, components and grade are imported and trained through deep learning, a model is established, the deep learning has a plurality of hidden neural networks, the process of characteristic extraction is integrated, the characteristics of a data set can be automatically learned,
a) fuzzy set data mining through artificial intelligence deep learning of hop status, allowing definition of fuzzy thresholds or boundaries, fuzzy logic using truth values of 0.0-1.0 to indicate a particular value is a member-specified degree, rather than class or set-exact truncation, abstracting the data, similar to experience, without a certain quantization standard,
b) learning the temperature, the pressure, the fermentation time, the ingredients, the excavated hop state data and the manually judged raw wine grade data through a forward propagation algorithm and a backward propagation algorithm of a neural network, wherein the neural network comprises an input layer, an output layer and a hidden layer, the excavated hop state data, the temperature, the pressure, the fermentation time and the ingredients are used as nodes of the input layer,
firstly, forward propagation learning is carried out, neurons of a hidden layer are used as nodes to be connected with nodes of a previous layer, parameters of the nodes of the previous layer are linearly combined through a certain weight proportion to obtain a new number, then a ReLU activation function is applied to the new number to obtain an output value, a plurality of neurons run in parallel, weight formulas of linear combination in the neurons of the same layer are different, the obtained result is used as a new variable to be transmitted to the next layer, multi-layer abstraction is carried out on multiple nonlinear transformation of combined low-layer features to form more abstract high-layer representation attribute categories or features so as to find data features and finally obtain the output value,
after the weight and the deviation are obtained by forward propagation learning, the backward propagation learning is carried out according to the raw wine grade data judged manually and the original node data of the input layer, the change condition of the error along with the output can be calculated, the weight and the deviation are corrected,
in the deep learning process, a discardable regularization algorithm is also used, the algorithm can effectively reduce the dependency among the neurons, so that the neural network can learn more robust characteristics with other neurons, overfitting can be reduced, the accuracy can be improved,
adding a discarding probability adjusting function to the discarding function in the discarding regularization process, adjusting the neuron with the largest weight value, wherein the first layer and the last layer are not added,
each neuron expression hi ═ Φ (∑ x)iw1i+bi)
Coefficient determinant is taken for each hidden layer
Calculating tripIn the column wniMaximum value, selecting corresponding hi to adjust the discarding probability,
x drop probability:
the rest is stretched correspondingly, which enhances the strength of the regularization,
in the whole process, the hop state (fuzzy judgment standard), the wine retort pressure, the fermentation time, the temperature data, the component data and other accurate data are combined, so that the grade of the original wine can be more effectively and accurately determined, the inaccuracy caused by individual difference in the traditional mode is avoided, the test data is used for off-line test, and if the deviation exists, the correction is carried out;
fourthly, verification: verifying the obtained model, selecting a wine retort whole process for verification, comparing a machine inspection result with a wine picking engineer evaluation result, and correcting if a deviation is large;
fifthly, data accumulation: data are continuously accumulated and expanded in the using process, the more the data contain more information, the more accurate the result is given after the neural network is learned, and the mining can help to improve the production process and improve the production efficiency through more data collection;
sixthly, the pressure sensor sends data to the PLC, the PLC is processed and then sent to the industrial personal computer, the data enter the input layer, other data directly enter the model input layer through the industrial personal computer, the grade of the unblended wine is obtained after processing, the industrial personal computer feeds the data back to the PLC, the PLC opens the corresponding electromagnetic valve and closes the electromagnetic valve which is not needed, and the unblended wine flows into the unblended wine collector with the corresponding grade.
Compared with the prior art, the invention has the beneficial effects that:
the method has the advantages that the factors influencing the quality of the liquor, such as alcohol concentration, fermentation time, chemical components, container pressure and the like, are monitored and analyzed, the traditional liquor picking process of picking the liquor by manual operation is avoided, the quality of the picked liquor cannot be guaranteed, the dependence on liquor pickers is strong, the data is deeply mined by taking the data as the basis, the data is learned by a neural network to find out the implicit rule and the relevance, the grade of the original liquor is judged according to the current input data, the liquor picking efficiency and the accuracy are improved, and the artificial error is avoided.
Drawings
FIG. 1 is a schematic structural view of the present invention;
FIG. 2 is a neural network learning diagram of the present invention;
in the figure: 1. carrying out wine retort; 2. a pressure sensor; 3. a pipeline; 4. picking wine bowls; 5. an online thermal imager; 6. a camera; 7. an online component analysis device; 8. an electromagnetic valve; 9. an industrial personal computer; 10. a PLC controller; 11. a wine base collector.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, 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 invention.
Referring to fig. 1-2, the present invention provides the following technical solutions: the utility model provides an artificial intelligence plucks wine device, including wine rice steamer 1, the lower extreme of wine rice steamer 1 is provided with pressure sensor 2, the side lower extreme of wine rice steamer 1 is provided with pipeline 3, the below of pipeline 3 is provided with plucks wine bowl 4, the inside of plucking wine bowl 4 is provided with online composition analytical equipment 7, the side lower extreme pipe of plucking wine bowl 4 is provided with solenoid valve 8, the side of pipeline 3 is provided with industrial computer 9, the side of industrial computer 9 is provided with online thermal imager 5 with 3 lower extreme positions in pipeline, the lower extreme of online thermal imager 5 is provided with camera 6.
Further, a PLC (programmable logic controller) 10 is arranged on the side edge of the industrial personal computer 9.
Further, the pressure sensor 2, the online thermal imager 5, the camera 6, the online component analysis device 7, the electromagnetic valve 8 and the industrial personal computer 9 are electrically connected with the PLC 10.
Further, a plurality of wine base collectors 11 are arranged below the electromagnetic valve 8.
Further, the camera 6 uses a laser camera to automatically emit laser without a light source for lighting.
Further, a timer for timing is arranged in the industrial personal computer 9.
Further, the wine picking method of the artificial intelligent wine picking device comprises the following steps:
firstly, data acquisition, namely selecting a plurality of different wine retorts 1, dynamically acquiring hop photos, wine retort pressure, fermentation time, temperature data, composition data (including acetaldehyde, ethyl acetate, methanol, fusel oil, ethyl linoleate, ethyl oleate, ethyl palmitate and other concentrations) and corresponding raw wine grades in various states in the whole wine picking process;
classifying the data into a neural network learning part and a testing part;
thirdly, model training: the collected hop state, temperature, components and grade are imported and trained through deep learning, a model is established, the deep learning has a plurality of hidden neural networks, the process of characteristic extraction is integrated, the characteristics of a data set can be automatically learned,
a) fuzzy set data mining through artificial intelligence deep learning of hop status, allowing definition of fuzzy thresholds or boundaries, fuzzy logic using truth values of 0.0-1.0 to indicate a particular value is a member-specified degree, rather than class or set-exact truncation, abstracting the data, similar to experience, without a certain quantization standard,
b) learning the temperature, the pressure, the fermentation time, the ingredients, the excavated hop state data and the manually judged raw wine grade data through a forward propagation algorithm and a backward propagation algorithm of a neural network, wherein the neural network comprises an input layer, an output layer and a hidden layer, the excavated hop state data, the temperature, the pressure, the fermentation time and the ingredients are used as nodes of the input layer,
firstly, forward propagation learning is carried out, neurons of a hidden layer are used as nodes to be connected with nodes of a previous layer, parameters of the nodes of the previous layer are linearly combined through a certain weight proportion to obtain a new number, then a ReLU activation function is applied to the new number to obtain an output value, a plurality of neurons run in parallel, weight formulas of linear combination in the neurons of the same layer are different, the obtained result is used as a new variable to be transmitted to the next layer, multi-layer abstraction is carried out on multiple nonlinear transformation of combined low-layer features to form more abstract high-layer representation attribute categories or features so as to find data features and finally obtain the output value,
after the weight and the deviation are obtained by forward propagation learning, the backward propagation learning is carried out according to the raw wine grade data judged manually and the original node data of the input layer, the change condition of the error along with the output can be calculated, the weight and the deviation are corrected,
in the deep learning process, a discardable regularization algorithm is also used, the algorithm can effectively reduce the dependency among the neurons, so that the neural network can learn more robust characteristics with other neurons, overfitting can be reduced, the accuracy can be improved,
adding a discarding probability adjusting function to the discarding function in the discarding regularization process, adjusting the neuron with the largest weight value, wherein the first layer and the last layer are not added,
each neuron expression hi ═ Φ (∑ x)iw1i+bi)
Coefficient determinant is taken for each hidden layer
Calculating w in determinantniMaximum value, selecting corresponding hi to adjust the discarding probability,
x drop probability:
the rest is stretched correspondingly, which enhances the strength of the regularization,
in the whole process, the fuzzy judgment standard of the hop state and accurate data such as the pressure, fermentation time, temperature data, composition data and the like of a wine retort are provided, and the combination of the fuzzy judgment standard of the hop state and the accurate data can more effectively and accurately determine the grade of the original wine, so that the inaccuracy caused by individual difference in the traditional mode is avoided, the offline test is carried out by using the test data, and if the deviation exists, the correction is carried out;
fourthly, verification: verifying the obtained model, selecting a wine retort whole process for verification, comparing a machine inspection result with a wine picking engineer evaluation result, and correcting if a deviation is large;
fifthly, data accumulation: data are continuously accumulated and expanded in the using process, the more the data contain more information, the more accurate the result is given after the neural network is learned, and the mining can help to improve the production process and improve the production efficiency through more data collection;
sixthly, the pressure sensor 2 sends data to the PLC controller 10, the PLC controller 10 sends the processed data to the industrial personal computer 9, the data enters the input layer, other data directly enter the model input layer through the processing of the industrial personal computer 9, the processed data obtain the grade of the original wine, the industrial personal computer feeds the data back to the PLC controller 10, the PLC controller 10 opens the corresponding electromagnetic valve 8, the electromagnetic valve 8 which is not needed is closed, and the original wine flows into the original wine collector 11 with the corresponding grade.
The working principle and the using process of the invention are as follows: when the invention is used:
firstly, data acquisition, namely selecting a plurality of different wine retorts 1, dynamically acquiring hop photos, wine retort pressure, fermentation time, temperature data, composition data (including acetaldehyde, ethyl acetate, methanol, fusel oil, ethyl linoleate, ethyl oleate, ethyl palmitate and other concentrations) and corresponding raw wine grades in various states in the whole wine picking process;
classifying the data into a neural network learning part and a testing part;
thirdly, model training: the collected hop state, temperature, components and grade are imported and trained through deep learning, a model is established, the deep learning has a plurality of hidden neural networks, the process of characteristic extraction is integrated, the characteristics of a data set can be automatically learned,
a) fuzzy set data mining through artificial intelligence deep learning of hop status, allowing definition of fuzzy thresholds or boundaries, fuzzy logic using truth values of 0.0-1.0 to indicate a particular value is a member-specified degree, rather than class or set-exact truncation, abstracting the data, similar to experience, without a certain quantization standard,
b) learning the temperature, the pressure, the fermentation time, the ingredients, the excavated hop state data and the manually judged raw wine grade data through a forward propagation algorithm and a backward propagation algorithm of a neural network, wherein the neural network comprises an input layer, an output layer and a hidden layer, the excavated hop state data, the temperature, the pressure, the fermentation time and the ingredients are used as nodes of the input layer,
firstly, forward propagation learning is carried out, neurons of a hidden layer are used as nodes to be connected with nodes of a previous layer, parameters of the nodes of the previous layer are linearly combined through a certain weight proportion to obtain a new number, then a ReLU activation function is applied to the new number to obtain an output value, a plurality of neurons run in parallel, weight formulas of linear combination in the neurons of the same layer are different, the obtained result is used as a new variable to be transmitted to the next layer, multi-layer abstraction is carried out on multiple nonlinear transformation of combined low-layer features to form more abstract high-layer representation attribute categories or features so as to find data features and finally obtain the output value,
after the weight and the deviation are obtained by forward propagation learning, the backward propagation learning is carried out according to the raw wine grade data judged manually and the original node data of the input layer, the change condition of the error along with the output can be calculated, the weight and the deviation are corrected,
in the deep learning process, a discardable regularization algorithm is also used, the algorithm can effectively reduce the dependency among the neurons, so that the neural network can learn more robust characteristics with other neurons, overfitting can be reduced, the accuracy can be improved,
adding a discarding probability adjusting function to the discarding function in the discarding regularization process, adjusting the neuron with the largest weight value, wherein the first layer and the last layer are not added,
each neuron expression hi ═ Φ (∑ x)iw1i+bi)
Coefficient determinant is taken for each hidden layer
Calculating w in determinantniMaximum value, selecting corresponding hi to adjust the discarding probability,
x drop probability:
the rest is stretched correspondingly, which enhances the strength of the regularization,
in the whole process, the fuzzy judgment standard of the hop state and accurate data such as the pressure, fermentation time, temperature data, composition data and the like of a wine retort are provided, and the combination of the fuzzy judgment standard of the hop state and the accurate data can more effectively and accurately determine the grade of the original wine, so that the inaccuracy caused by individual difference in the traditional mode is avoided, the offline test is carried out by using the test data, and if the deviation exists, the correction is carried out;
fourthly, verification: verifying the obtained model, selecting a wine retort whole process for verification, comparing a machine inspection result with a wine picking engineer evaluation result, and correcting if a deviation is large;
fifthly, data accumulation: data are continuously accumulated and expanded in the using process, the more the data contain more information, the more accurate the result is given after the neural network is learned, and the mining can help to improve the production process and improve the production efficiency through more data collection;
sixthly, the pressure sensor 2 sends data to the PLC controller 10, the PLC controller 10 sends the processed data to the industrial personal computer 9, the data enters the input layer, other data directly enter the model input layer through the processing of the industrial personal computer 9, the processed data obtain the grade of the original wine, the industrial personal computer feeds the data back to the PLC controller 10, the PLC controller 10 opens the corresponding electromagnetic valve 8, the electromagnetic valve 8 which is not needed is closed, and the original wine flows into the original wine collector 11 with the corresponding grade.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. The utility model provides an artificial intelligence plucks wine device, includes wine rice steamer (1), its characterized in that: the wine retort is characterized in that a pressure sensor (2) is arranged at the lower end of the wine retort (1), a pipeline (3) is arranged at the lower end of the side edge of the wine retort (1), a wine picking bowl (4) is arranged below the pipeline (3), an online component analysis device (7) is arranged inside the wine picking bowl (4), an electromagnetic valve (8) is arranged on a conduit at the lower end of the side edge of the wine picking bowl (4), an industrial personal computer (9) is arranged at the side edge of the pipeline (3), an online thermal imager (5) is arranged at the position of the side edge of the industrial personal computer (9) and the lower end of the pipeline (3), and a camera (6) is arranged at the lower end of the online thermal imager (5).
2. The artificial intelligence wine extraction device of claim 1, wherein: and a PLC (programmable logic controller) (10) is arranged on the side edge of the industrial personal computer (9).
3. The artificial intelligence wine extraction device of claim 1, wherein: the pressure sensor (2), the online thermal imager (5), the camera (6), the online component analysis device (7), the electromagnetic valve (8) and the industrial personal computer (9) are all electrically connected with the PLC controller (10).
4. The artificial intelligence wine extraction device of claim 1, wherein: a plurality of raw wine collectors (11) are arranged below the electromagnetic valve (8).
5. The artificial intelligence wine extraction device of claim 1, wherein: the camera (6) uses a laser camera to automatically emit laser without a light source for lighting.
6. The artificial intelligence wine extraction device of claim 1, wherein: and a timer for timing is arranged in the industrial personal computer (9).
7. The artificial intelligence wine extraction device of claim 1, wherein: the wine picking method of the artificial intelligent wine picking device comprises the following steps:
firstly, data acquisition, namely selecting a plurality of different wine retorts (1), dynamically acquiring hop photos, wine retort pressure, fermentation time, temperature data, composition data (including acetaldehyde, ethyl acetate, methanol, fusel oil, ethyl linoleate, ethyl oleate, ethyl palmitate and other concentrations) and corresponding raw wine grades in various states in the whole wine picking process;
classifying the data into a neural network learning part and a testing part;
thirdly, model training: the collected hop state, temperature, components and grade are imported and trained through deep learning, a model is established, the deep learning has a plurality of hidden neural networks, the process of characteristic extraction is integrated, the characteristics of a data set can be automatically learned,
a) fuzzy set data mining through artificial intelligence deep learning of hop status, allowing definition of fuzzy thresholds or boundaries, fuzzy logic using truth values of 0.0-1.0 to indicate a particular value is a member-specified degree, rather than class or set-exact truncation, abstracting the data, similar to experience, without a certain quantization standard,
b) learning the temperature, the pressure, the fermentation time, the ingredients, the excavated hop state data and the manually judged raw wine grade data through a forward propagation algorithm and a backward propagation algorithm of a neural network, wherein the neural network comprises an input layer, an output layer and a hidden layer, the excavated hop state data, the temperature, the pressure, the fermentation time and the ingredients are used as nodes of the input layer,
firstly, forward propagation learning is carried out, neurons of a hidden layer are used as nodes to be connected with nodes of a previous layer, parameters of the nodes of the previous layer are linearly combined through a certain weight proportion to obtain a new number, then a ReLU activation function is applied to the new number to obtain an output value, a plurality of neurons run in parallel, weight formulas of linear combination in the neurons of the same layer are different, the obtained result is used as a new variable to be transmitted to the next layer, multi-layer abstraction is carried out on multiple nonlinear transformation of combined low-layer features to form more abstract high-layer representation attribute categories or features so as to find data features and finally obtain the output value,
after the weight and the deviation are obtained by forward propagation learning, the backward propagation learning is carried out according to the raw wine grade data judged manually and the original node data of the input layer, the change condition of the error along with the output can be calculated, the weight and the deviation are corrected,
in the deep learning process, a discardable regularization algorithm is also used, the algorithm can effectively reduce the dependency among the neurons, so that the neural network can learn more robust characteristics with other neurons, overfitting can be reduced, the accuracy can be improved,
adding a discarding probability adjusting function to the discarding function in the discarding regularization process, adjusting the neuron with the largest weight value, wherein the first layer and the last layer are not added,
each neuron expression hi ═ Φ (∑ x)iw1i+bi)
Coefficient determinant is taken for each hidden layer
Calculating w in determinantniMaximum value, selecting corresponding hi to adjust the discarding probability,
x drop probability:
the rest is stretched correspondingly, which enhances the strength of the regularization,
in the whole process, the hop state (fuzzy judgment standard), the wine retort pressure, the fermentation time, the temperature data, the component data and other accurate data are combined, so that the grade of the original wine can be more effectively and accurately determined, the inaccuracy caused by individual difference in the traditional mode is avoided, the test data is used for off-line test, and if the deviation exists, the correction is carried out;
fourthly, verification: verifying the obtained model, selecting a wine retort whole process for verification, comparing a machine inspection result with a wine picking engineer evaluation result, and correcting if a deviation is large;
fifthly, data accumulation: data are continuously accumulated and expanded in the using process, the more the data contain more information, the more accurate the result is given after the neural network is learned, and the mining can help to improve the production process and improve the production efficiency through more data collection;
sixthly, pressure sensor (2) give PLC controller (10) with data transmission, PLC controller (10) are handled the back and are sent industrial computer (9), data entering input layer, other data are direct to be handled through industrial computer (9) and are got into the model input layer, reachd the wine base grade after handling, the industrial computer feeds back data to PLC controller (10), PLC controller (10) open corresponding solenoid valve (8), close unnecessary solenoid valve (8), the wine base flows in the wine base collector (11) of corresponding grade.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011218685.9A CN112180832A (en) | 2020-11-04 | 2020-11-04 | Artificial intelligent wine picking device and wine picking method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011218685.9A CN112180832A (en) | 2020-11-04 | 2020-11-04 | Artificial intelligent wine picking device and wine picking method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112180832A true CN112180832A (en) | 2021-01-05 |
Family
ID=73917033
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011218685.9A Pending CN112180832A (en) | 2020-11-04 | 2020-11-04 | Artificial intelligent wine picking device and wine picking method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112180832A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112699968A (en) * | 2021-01-18 | 2021-04-23 | 河北工业大学 | Pluck wine equipment based on white spirit hops image |
| CN112766806A (en) * | 2021-02-03 | 2021-05-07 | 河北工业大学 | Raw wine grade evaluation method and online wine picking and raw wine grade evaluation device |
| CN112765900A (en) * | 2021-02-03 | 2021-05-07 | 河北工业大学 | Alcohol content online detection method and system for segmented liquor picking |
| CN113621470A (en) * | 2021-08-10 | 2021-11-09 | 河北凤来仪酒业有限公司 | Method for segmented liquor picking by utilizing conductivity of raw liquor |
| CN114276891A (en) * | 2022-01-11 | 2022-04-05 | 安徽金种子酒业股份有限公司 | Artifical supplementary hierarchical wine device that connects |
| CN117709914A (en) * | 2024-02-05 | 2024-03-15 | 天津徙木科技有限公司 | Post matching method and system |
| CN117709914B (en) * | 2024-02-05 | 2024-05-10 | 台州徙木数字服务有限公司 | Post matching method and system |
-
2020
- 2020-11-04 CN CN202011218685.9A patent/CN112180832A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112699968A (en) * | 2021-01-18 | 2021-04-23 | 河北工业大学 | Pluck wine equipment based on white spirit hops image |
| CN112699968B (en) * | 2021-01-18 | 2022-05-27 | 河北工业大学 | Liquor picking equipment and method based on liquor hops image |
| CN112766806A (en) * | 2021-02-03 | 2021-05-07 | 河北工业大学 | Raw wine grade evaluation method and online wine picking and raw wine grade evaluation device |
| CN112765900A (en) * | 2021-02-03 | 2021-05-07 | 河北工业大学 | Alcohol content online detection method and system for segmented liquor picking |
| CN113621470A (en) * | 2021-08-10 | 2021-11-09 | 河北凤来仪酒业有限公司 | Method for segmented liquor picking by utilizing conductivity of raw liquor |
| CN114276891A (en) * | 2022-01-11 | 2022-04-05 | 安徽金种子酒业股份有限公司 | Artifical supplementary hierarchical wine device that connects |
| CN114276891B (en) * | 2022-01-11 | 2023-07-11 | 安徽金种子酒业股份有限公司 | Manual auxiliary grading wine receiving device |
| CN117709914A (en) * | 2024-02-05 | 2024-03-15 | 天津徙木科技有限公司 | Post matching method and system |
| CN117709914B (en) * | 2024-02-05 | 2024-05-10 | 台州徙木数字服务有限公司 | Post matching method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112180832A (en) | Artificial intelligent wine picking device and wine picking method thereof | |
| CN109063728A (en) | A kind of fire image deep learning mode identification method | |
| CN109858516A (en) | A kind of fire and smog prediction technique, system and medium based on transfer learning | |
| CN111127423B (en) | Rice pest and disease identification method based on CNN-BP neural network algorithm | |
| CN112547807B (en) | Decision tree algorithm-based hot rolled strip quality accurate judgment method | |
| CN115570228B (en) | Intelligent feedback control method and system for welding pipeline gas supply | |
| CN110287540A (en) | A kind of photovoltaic cell parameter identification method based on elite masses' differential evolution algorithm | |
| CN107290481B (en) | Cowshed environment ammonia intelligent detection system based on wireless sensor network | |
| CN109194498A (en) | A kind of network flow prediction method based on LSTM | |
| CN213690307U (en) | Artificial intelligence wine picking device | |
| CN110309609A (en) | A kind of architecture indoor air quality evaluation method based on rough set and wavelet neural network | |
| CN114049305A (en) | Distribution line pin defect detection method based on improved ALI and fast-RCNN | |
| CN114282443A (en) | Residual service life prediction method based on MLP-LSTM supervised joint model | |
| CN106779346A (en) | A kind of Forecasting Methodology of monthly power consumption | |
| CN114564513A (en) | Sea fog prediction method, device, equipment and storage medium based on neural network | |
| Liu et al. | Research on the strategy of locating abnormal data in IOT management platform based on improved modified particle swarm optimization convolutional neural network algorithm | |
| CN108665001B (en) | Cross-tested idle state detection method based on deep belief network | |
| CN117574776A (en) | Task planning-oriented model self-learning optimization method | |
| Ma et al. | Integration agricultural knowledge and Internet of Things for multi-agent deficit irrigation control | |
| CN117439069A (en) | Electric quantity prediction method based on neural network | |
| CN111275025A (en) | Parking space detection method based on deep learning | |
| CN110738307A (en) | Application of BP neural network model in prediction of water quality of peach forest mouth reservoir | |
| CN116244596A (en) | Industrial time sequence data anomaly detection method based on TCN and attention mechanism | |
| CN110009620A (en) | The method and system of power grid stockbridge damper intelligent recognition are carried out based on transfer learning algorithm | |
| CN115115880A (en) | Mushroom quality grading method based on deep learning |
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 |