Automatic control system for optimizing the temperature of the circulating air in the feeding process

Abstract

The present invention discloses an automatic control system for optimizing the circulating air temperature of feeding, which includes a factor screening module, a model establishment module, a prediction and control module, a debugging module and a model optimization module. The present invention is applied to the blade feeder in the blade processing process, aiming to realize the intelligent control of the blade feeder, improve the stability of the control of the circulating air temperature index, optimize the control method, change the existing control method to intelligent precise control, and finally realize functions such as improving product quality, reducing product differences between batches, and intelligent warning.

Description

[0001] This application is a divisional application of the application with the application number 202210748901.3 and the invention title of "Automatic Control Method for Optimizing the Circulating Air Temperature of Feeding Based on Response Surface Method". Technical Field

[0002] The present invention is applied to the field of cigarette making and silk reeling, and specifically relates to an automatic control system for optimizing the circulating air temperature of feeding based on the response surface method. Background Art

[0003] The silk reeling line is a fully automatic production line centrally controlled by the central control room. The central control room is also known as the "heart" of the silk reeling workshop. The parameters and indicators of each process point on site are centrally controlled by the central control room. At present, 4 employees in the central control room control the production line in sections. In the leaf pretreatment process section, the circulating air temperature CPK of leaf feeding is the key point of process index assessment.

[0004] According to the batch weight distribution table, when CPK≥2, the standard grade of this batch can be listed as A+. The circulating air temperature is in manual control mode. The central control driver adjusts according to the displayed value of the circulating air temperature based on operation experience, and at the same time has to take into account the outlet temperature and outlet moisture indicators. There will be phenomena of adjustment lag and inaccurate adjustment during the adjustment process; at the same time, there is no early warning system, and it depends entirely on manual monitoring.

[0005] Taking Diamond (Hard Special Alcohol) as an example, the CPK data of the circulating air temperature of leaf feeding in three teams in the workshop from January to May 2022 were statistically analyzed, and the average value of CPK of the circulating air temperature was analyzed. Although some cigarettes in Team A could reach the A+ grade, the average value of CPK was 1.89≤2, and the overall control level only reached the qualified requirement. By analyzing the stability of the quality control level of the central control driver in each shift through the operation chart, it can be seen that the control of the circulating air temperature fluctuates greatly and there is a large room for improvement.

[0006] There is an urgent need for a system that can accurately control and accurately give early warnings to realize the intelligence and automation of the production control of the circulating air temperature. Summary of the Invention

[0007] To solve the above problems, the present application provides an automatic control system for optimizing the circulating air temperature of feeding based on the response surface method, which is applied to the leaf feeder in the leaf treatment process, aiming to realize the intelligent control of the leaf feeder, improve the stability of the control of the circulating air temperature index, optimize the control method, and finally realize functions such as improving product quality, reducing product differences between batches, and intelligent early warning.

[0008] The technical solution adopted by the present invention to solve its technical problems is as follows:

[0009] An automatic control method for optimizing the circulating air temperature of feeding based on the response surface method, including the following steps:

[0010] (1) Factor screening

[0011] 1. Factor search

[0012] Search for factors related to the circulating air temperature and its CPK in the leaf processing operation:

[0013] Moisture at the leaf feeding outlet, moisture at the leaf feeding inlet, temperature at the leaf feeding outlet, opening of the compensating steam for leaf feeding, circulating air temperature for leaf feeding, CPK of the circulating air temperature for leaf feeding, opening of the moisture exhaust for leaf feeding, material flow rate of leaf feeding, temperature of the liquid for leaf feeding, ambient humidity;

[0014] 2. Factor screening

[0015] Classify the identified relevant factors:

[0016] Process parameters: Moisture at the leaf feeding outlet, moisture at the leaf feeding inlet, temperature at the leaf feeding outlet, circulating air temperature for leaf feeding, CPK of the circulating air temperature for leaf feeding, temperature of the liquid for leaf feeding;

[0017] Production parameters: Opening of the compensating steam for leaf feeding, opening of the moisture exhaust for leaf feeding;

[0018] Environmental parameters: Ambient humidity;

[0019] 3. Key factor screening

[0020] Screen out the key factors affecting the circulating air temperature and its CPK through the C&E matrix and FMEA factor analysis method:

[0021] Opening of the compensating steam for leaf feeding, opening of the moisture exhaust for leaf feeding, material flow rate of leaf feeding;

[0022] (2) Model establishment

[0023] 1. Prediction model I

[0024] The prediction model I adopts a response surface design model, which takes the key factors as input values, the circulating air temperature for leaf feeding as the output value, the CPK of the circulating air temperature for leaf feeding as the evaluation item, and the moisture at the leaf feeding outlet and the temperature at the leaf feeding outlet as reference items;

[0025] Prediction process of prediction model I:

[0026] ① 3-factor 2-level experiment

[0027] Take the opening of the compensating steam for leaf feeding as Z1, Z2, Z3, the opening of the moisture exhaust for leaf feeding as P1, P2, P3, and the material flow rate of leaf feeding as W1, W2, W3 respectively, add a central point experimental plan, and conduct M experiments;

[0028] ② Dual response surface analysis

[0029] Based on the response surface method, under the condition of meeting the process standard requirements, the optimal value range of each parameter at the above-mentioned values is obtained through the response optimizer, that is, the predicted value I;

[0030] 2. Prediction model II

[0031] The prediction model II adopts a neural network model, which takes production parameters and process parameters as modeling factors, takes the air temperature of the leaf feeding circulation as the output value, and takes other factors except the air temperature of the leaf feeding circulation and the CPK of the air temperature of the leaf feeding circulation as input values;

[0032] (3) Model prediction and control

[0033] Based on the prediction model I, the predicted value I of the air temperature of the leaf feeding circulation is obtained;

[0034] Based on the prediction model II, the predicted value II of the air temperature of the leaf feeding circulation is obtained;

[0035] The predicted value I and the predicted value II are calculated and compared:

[0036] When the deviation between the predicted values of the two models ≤ 2 degrees, normal production is carried out and the predicted value I is used as the control parameter to intelligently control the air temperature of the leaf feeding circulation;

[0037] When the deviation between the predicted values of the two models > 2 degrees, a warning is prompted;

[0038] In the process of intelligently controlling the air temperature of the leaf feeding circulation with the predicted value I as the control parameter:

[0039] Combined with the weight of the scale, the production process is divided into three stages for intelligent control respectively:

[0040] At the beginning of production, the control parameter is the predicted value I plus 10%;

[0041] When the material production reaches 500KG, the control parameter is the predicted value I minus 5%;

[0042] When the material production reaches 700KG, the control parameter is the predicted value I;

[0043] Because the air temperature of the leaf feeding circulation decreases faster than it increases during the production process, the above-mentioned intelligent control based on model prediction can stabilize the air temperature of the leaf feeding circulation at the standard midline;

[0044] (4) Model optimization

[0045] An optimized adaptive model, which adopts a neural network model, takes process parameters, production parameters, and environmental parameters as modeling factors, takes the circulating air temperature of leaf feeding as the output value, and is established with the moisture content at the outlet of leaf feeding, the moisture content at the inlet of leaf feeding, the temperature at the outlet of leaf feeding, the opening degree of the compensation steam for leaf feeding, the opening degree of the moisture exhaust for leaf feeding, the material flow rate of leaf feeding, the liquid temperature of the leaf feeding material, and the environmental humidity as input values;

[0046] After the system issues a warning prompt, optimize the prediction model II through the optimized adaptive model, and return to step (III);

[0047] If the system still issues a warning prompt, stop production and wait for maintenance;

[0048] Utilize the adaptive function of the neural network model to continuously calculate and optimize the model to eliminate the errors caused by the influence of environmental temperature and humidity.

[0049] The purpose of the present invention also lies in providing an automatic control system for optimizing the circulating air temperature of feeding based on the response surface method.

[0050] The automatic control system for optimizing the circulating air temperature of feeding based on the response surface method includes a factor screening module, a model establishment module, a prediction and control module, a debugging module, and a model optimization module;

[0051] The factor screening module is used to search for factors related to the circulating air temperature and the circulating air temperature CPK in the leaf processing process, classify the found related factors, and screen out the key factors affecting the circulating air temperature and the circulating air temperature CPK through the C&E matrix and the FMEA factor analysis method:

[0052] The opening degree of the compensation steam for leaf feeding, the opening degree of the moisture exhaust for leaf feeding, the material flow rate of leaf feeding;

[0053] The model establishment module is used to construct a prediction model, and the prediction model includes a prediction model I and a prediction model II;

[0054] The prediction model I adopts a response surface design model, which takes the key factors as input values, takes the circulating air temperature of leaf feeding as the output value, takes the circulating air temperature CPK of leaf feeding as an evaluation item, and is established with the moisture content at the outlet of leaf feeding and the temperature at the outlet of leaf feeding as reference items;

[0055] The prediction model II adopts a neural network model, which takes production parameters and process parameters as modeling factors, takes the circulating air temperature of leaf feeding as the output value, and is established with other factors except the circulating air temperature of leaf feeding and the circulating air temperature CPK of leaf feeding as input values;

[0056] A prediction and control module, which is used to obtain the predicted value I of the circulating air temperature of the blade feeder based on the prediction model I and the predicted value II of the circulating air temperature of the blade feeder based on the prediction model II, and perform arithmetic comparison between the predicted value I and the predicted value II:

[0057] When the deviation between the predicted values of the two models ≤ 2 degrees, normal production is carried out and the predicted value I is used as the control parameter to intelligently control the circulating air temperature of the blade feeder;

[0058] When the deviation between the predicted values of the two models > 2 degrees, a warning is prompted;

[0059] During the process of intelligently controlling the circulating air temperature of the blade feeder with the predicted value I as the control parameter:

[0060] Combined with the weight of the scale, the production process is divided into three stages for intelligent control respectively:

[0061] At the beginning of production, the control parameter is the predicted value I plus 10%;

[0062] When the material production reaches 500 KG, the control parameter is the predicted value I minus 5%;

[0063] When the material production reaches 700 KG, the control parameter is the predicted value I;

[0064] A debugging module, which is used to connect this automatic control system with the current production operation system wincc for debugging;

[0065] A model optimization module, which is used to optimize the prediction model II based on process parameters, production parameters and environmental parameters after the system issues a warning prompt, and return to the prediction and control module; if the system still issues a warning prompt, stop production and wait for maintenance.

[0066] The beneficial effects brought by the present invention are:

[0067] By optimizing the control method of the circulating air temperature of the blade feeder, the present invention changes the existing control method to intelligent precise control, realizes the intelligent control of the blade feeder, can improve the stability of the control of the circulating air temperature index, and significantly improves the CPK of the circulating air temperature. At the same time, it has functions such as intelligent warning and model optimization and adaptation;

[0068] Based on the dual models, the prediction and intelligent control of the control parameters of the circulating air temperature are realized, which is convenient for realizing unified production control, thereby effectively reducing the difference in product quality stability between different production teams, improving the quality stability, realizing homogenized production, reducing the product difference between batches, and improving the quality guarantee ability of wire making. Description of the Drawings

[0069] The present invention will be further described below in conjunction with the drawings and specific embodiments.

[0070] Figure 1 It is the system block diagram of the automatic control system for optimizing the feeding circulating air temperature based on the response surface method;

[0071] Figure 2 It is the construction structure diagram of Prediction Model I;

[0072] Figure 3 It is the residual diagram of the circulating air temperature of Prediction Model I in Example 1

[0073] Figure 4 It is the CPK residual diagram of the circulating air temperature of Prediction Model I in Example 1;

[0074] Figure 5 It is the optimization diagram of the response optimizer of Prediction Model I in Example 1;

[0075] Figure 6 It is the surface diagram of the CPK of the circulating air temperature of Prediction Model I in Example 1, the opening of the compensating steam, and the opening of the moisture exhaust;

[0076] Figure 7 It is the construction structure diagram of Prediction Model II. Specific implementation manners

[0077] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0078] Example 1

[0079] An automatic control method for optimizing the feeding circulating air temperature based on the response surface method includes the following steps:

[0080] (1) Factor screening

[0081] 1. Factor search

[0082] Search for factors related to the circulating air temperature and the CPK of the circulating air temperature in the blade processing process:

[0083] The moisture at the blade feeding outlet, the moisture at the blade feeding inlet, the temperature at the blade feeding outlet, the opening of the compensating steam for blade feeding, the circulating air temperature for blade feeding, the CPK of the circulating air temperature for blade feeding, the opening of the moisture exhaust for blade feeding, the material flow rate of blade feeding, the temperature of the feed liquid for blade feeding, the environmental humidity;

[0084] 2. Factor screening

[0085] Classify the identified relevant factors:

[0086] Process parameters: moisture content at the outlet of leaf feeding, moisture content at the inlet of leaf feeding, temperature at the outlet of leaf feeding, temperature of the circulating air for leaf feeding, CPK of the temperature of the circulating air for leaf feeding, temperature of the liquid material for leaf feeding;

[0087] Production parameters: opening degree of the compensating steam for leaf feeding, opening degree of the moisture exhaust for leaf feeding;

[0088] Environmental parameters: environmental humidity;

[0089] 3. Screening of key factors

[0090] The key factors affecting the temperature of the circulating air and the CPK of the temperature of the circulating air are screened out through the C&E matrix and the FMEA factor analysis method:

[0091] Opening degree of the compensating steam for leaf feeding, opening degree of the moisture exhaust for leaf feeding, material flow rate of leaf feeding;

[0092] (2) Model establishment

[0093] 1. Prediction model I

[0094] Refer to Figure 2 , the prediction model I adopts the response surface design model, which takes the key factors as input values, the temperature of the circulating air for leaf feeding as the output value, the CPK of the temperature of the circulating air for leaf feeding as the evaluation item, and the moisture content at the outlet of leaf feeding and the temperature at the outlet of leaf feeding as reference items;

[0095] Prediction process of prediction model I: [[ID=3A]]

[0096] ① 3-factor 2-level experiment

[0097] The opening degree of the compensating steam for leaf feeding is respectively taken as 42%, 43%, 44%, the opening degree of the moisture exhaust for leaf feeding is taken as 30%, 40%, 50%, and the material flow rate of leaf feeding is taken as 9500 kg / h, 9600 kg / h, 9700 kg / h. The central point experimental scheme is added, and 20 times (the value of M is determined according to the number of factors and the levels of factors) of experiments are carried out. The results are shown in Table 1;

[0098] Table 1

[0099] C1 C2 C3 C4 C5 C6 C7 C8 C9 Standard order Running order Point type Block Compensation steam opening Exhaust opening Material flow rate Recirculating air temperature Recirculating air temperature CPK 19 1 0 1 43.0000 40.0000 9600.00 52.26 2.01 13 2 -1 1 43.0000 40.0000 9431.82 51.40 1.29 4 3 1 1 44.0000 50.0000 9500.00 52.54 2.55 6 4 1 1 44.0000 30.0000 9700.00 49.93 2.26 3 5 1 1 42.0000 50.0000 9500.00 52.23 3.16 15 6 0 1 43.0000 40.0000 9600.00 50.27 3.24 11 7 -1 1 43.0000 23.1821 9600.00 50.27 2.75 10 8 -1 1 44.6818 40.0000 9600.00 53.08 1.82 1 9 -1 1 42.0000 30.0000 9500.00 53.35 1.95 14 10 -1 1 43.0000 40.0000 9768.18 53.35 2.13 18 11 0 1 43.0000 40.0000 9600.00 52.54 3.10 8 12 1 1 44.0000 50.0000 9700.00 52.54 2.34 17 13 0 1 43.0000 40.0000 9600.00 52.09 3.68 7 14 1 1 42.0000 50.0000 9700.00 51.99 2.68 12 15 -1 1 43.0000 56.8179 9600.00 49.77 1.29 5 16 1 1 42.0000 30.0000 9700.00 53.08 1.92 16 17 0 1 43.0000 40.0000 9600.00 53.54 3.10 2 18 1 1 44.0000 30.0000 9500.00 53.13 1.89 9 19 -1 1 41.3182 40.0000 9600.00 53.13 2.42 20 20 0 1 43.0000 40.0000 9600.00 52.62 2.98

[0100] ② Dual response surface analysis

[0101] Figure 3 Shows the residual plot of the temperature of the circulating air; Figure 4 Shows the residual plot of the CPK of the temperature of the circulating air;

[0102] Figure 5 Shows the optimization diagram of the response optimizer; Figure 6 Shows the surface diagram of the CPK of the temperature of the circulating air and the opening degrees of the compensating steam and the moisture exhaust; Note: In the translation, "3A" in the original text might be a mislabeling. It should probably be "31" as per the sequential numbering. This has been noted in the translation for reference.

[0103] Based on the response surface method, under the condition of meeting the process standard requirements, the optimal value range of each parameter at the above-mentioned values is obtained through the response optimizer, that is, the predicted value I.

[0104] Optimal value range: When the compensation steam opening is 43%, the exhaust opening is 40%, and the material flow rate is 9611 kg / h, the circulating air temperature CPK > 2.

[0105] 2. Prediction model II

[0106] Refer to Figure 7 , the prediction model II adopts a neural network model, which takes production parameters and process parameters as modeling factors, takes the circulating air temperature of the vane feeding as the output value, and takes other factors except the circulating air temperature of the vane feeding and the CPK of the circulating air temperature of the vane feeding as input values to establish.

[0107] This neural network model includes three layers: an input layer, a hidden layer, and an output layer, with 4 neurons, a set training target of 0.05, a training speed of 0.01, and a maximum number of steps of 100.

[0108] (III) Model prediction and control

[0109] Based on the prediction model I, the predicted value I of the circulating air temperature of the vane feeding is obtained;

[0110] Based on the prediction model II, the predicted value II of the circulating air temperature of the vane feeding is obtained;

[0111] The predicted value I and the predicted value II are calculated and compared:

[0112] When the deviation between the predicted values of the two models ≤ 2 degrees, normal production is carried out and the predicted value I is used as the control parameter to intelligently control the circulating air temperature of the vane feeding;

[0113] When the deviation between the predicted values of the two models > 2 degrees, a warning is prompted; the preset value of the deviation warning is adjustable;

[0114] [[ID=V37]]During the process of intelligently controlling the circulating air temperature of the vane feeding with the predicted value I as the control parameter:

[0115] Combined with the weight of the scale (i.e., the material flow rate of the vane feeding), the production process is divided into three stages for intelligent control respectively:

[0116] At the beginning of production, the control parameter is the predicted value I plus 10%;

[0117] When the material production reaches 500 KG, the control parameter is the predicted value I minus 5%;

[0118] When the material production reaches 700 KG, the control parameter is the predicted value I;

[0119] Since the temperature of the circulating air for leaf feeding decreases faster than it increases during the production process, the intelligent control based on model prediction can stabilize the temperature of the circulating air for leaf feeding at the standard midline.

[0120] (V) Model Optimization

[0121] Optimize the adaptive model, which uses a neural network model. The process parameters, production parameters, and environmental parameters are used as modeling factors, the temperature of the circulating air for leaf feeding is used as the output value, and the moisture content at the outlet of leaf feeding, the moisture content at the inlet of leaf feeding, the temperature at the outlet of leaf feeding, the opening degree of the compensation steam for leaf feeding, the opening degree of the moisture exhaust for leaf feeding, the material flow rate of leaf feeding, the temperature of the liquid material for leaf feeding, and the environmental humidity are used as input values to establish the model.

[0122] After the system issues a warning prompt, optimize the prediction model II through the optimized adaptive model and return to step (III).

[0123] If the system still issues a warning prompt, stop production and wait for maintenance.

[0124] Utilize the adaptive function of the neural network model to continuously calculate and optimize the model to eliminate the error caused by the influence of environmental temperature and humidity.

[0125] Example 2

[0126] Refer to Figure 1 , an automatic control system for optimizing the temperature of the circulating air for feeding based on the response surface method, including a factor screening module, a model establishment module, a prediction and control module, a debugging module, and a model optimization module;

[0127] The factor screening module is used to find the factors related to the circulating air temperature and the CPK of the circulating air temperature in the leaf processing process, classify the found related factors, and screen out the key factors affecting the circulating air temperature and the CPK of the circulating air temperature through the C&E matrix and the FMEA factor analysis method:

[0128] The opening degree of the compensation steam for leaf feeding, the opening degree of the moisture exhaust for leaf feeding, the material flow rate of leaf feeding;

[0129] The model establishment module is used to construct the prediction model, and the prediction model includes prediction model I and prediction model II;

[0130] Refer to Figure 2 , prediction model I adopts a response surface design model, which takes the key factors as input values, the temperature of the circulating air for leaf feeding as the output value, the CPK of the temperature of the circulating air for leaf feeding as the evaluation item, and the moisture content at the outlet of leaf feeding and the temperature at the outlet of leaf feeding as reference items to establish;

[0131] Refer to Figure 7, the prediction model II adopts a neural network model, which takes production parameters and process parameters as modeling factors, takes the circulating air temperature of the blade feeding cycle as the output value, and takes other factors except the circulating air temperature of the blade feeding cycle and the CPK of the circulating air temperature of the blade feeding cycle as the input values;

[0132] A prediction and control module, which is used to obtain the predicted value I of the circulating air temperature of the blade feeding cycle based on the prediction model I and the predicted value II of the circulating air temperature of the blade feeding cycle based on the prediction model II, and perform arithmetic comparison on the predicted value I and the predicted value II:

[0133] When the deviation between the predicted values of the two models ≤ 2 degrees, normal production is carried out and the predicted value I is used as the control parameter to perform intelligent control on the circulating air temperature of the blade feeding cycle;

[0134] When the deviation between the predicted values of the two models > 2 degrees, a warning is prompted;

[0135] In the process of performing intelligent control on the circulating air temperature of the blade feeding cycle with the predicted value I as the control parameter:

[0136] Combined with the weight of the scale, the production process is divided into three stages for intelligent control respectively:

[0137] At the beginning of production, the control parameter is the predicted value I plus 10%;

[0138] When the material production reaches 500 KG, the control parameter is the predicted value I minus 5%;

[0139] When the material production reaches 700 KG, the control parameter is the predicted value I;

[0140] A debugging module, which is used to dock this automatic control system with the current production operation system wincc for debugging;

[0141] A model optimization module, referring to Embodiment 1, which is used to optimize the prediction model II based on process parameters, production parameters and environmental parameters after the system issues a warning prompt, and return to the prediction and control module; if the system still issues a warning prompt, stop production and wait for maintenance.

[0142] This system realizes the prediction and intelligent control of the control parameters of the circulating air temperature based on a dual model, which is convenient for realizing unified production control, thereby effectively narrowing the difference in product quality stability between different production teams, reducing the product difference between batches, and improving the quality stability.

[0143] Embodiment 3

[0144] Apply the automatic control method for optimizing the feeding circulating air temperature based on the response surface method in Example 1 and the automatic control system for optimizing the feeding circulating air temperature based on the response surface method in Example 2 to the leaf treatment process of the cigarette making line to realize the intelligent control of the leaf feeder; collect the tobacco process assessment indicators after intelligent control, and the results are shown in Table 2;

[0145] Table 2

[0146] Batch Recirculating air temperature qualification rate Recirculating air temperature CPK 1 100％ 2.6 2 100％ 3.4 3 100％ 2.8 4 100％ 2.5 5 100％ 2.3 6 100％ 2.4 7 100％ 2.2 8 100％ 2.9 9 100％ 2.4 10 100％ 2.5

[0147] As can be seen from Table 2, the automatic control method and system in Example 1 and Example 2 have remarkable effects. After application, all process indicators are qualified, and the CPK of the circulating air temperature has been significantly improved.

[0148] It should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An automatic control system for optimizing the temperature of the charging circulating air, characterized in that: It includes a factor screening module, a model establishment module, and a prediction and control module; The factor screening module is used to search for factors related to the circulating air temperature and the circulating air temperature CPK in the leaf processing process, classify the found related factors, and screen out the key factors affecting the circulating air temperature and the circulating air temperature CPK; the related factors include the moisture content at the leaf feeding outlet, the moisture content at the leaf feeding inlet, the temperature at the leaf feeding outlet, the opening of the compensating steam for leaf feeding, the circulating air temperature for leaf feeding, the circulating air temperature CPK for leaf feeding, the opening of the moisture exhaust for leaf feeding, the material flow rate for leaf feeding, the temperature of the liquid for leaf feeding, and the environmental humidity; They are classified as: Process parameters: the moisture content at the leaf feeding outlet, the moisture content at the leaf feeding inlet, the temperature at the leaf feeding outlet, the circulating air temperature for leaf feeding, the circulating air temperature CPK for leaf feeding, the material flow rate for leaf feeding, the temperature of the liquid for leaf feeding; Production parameters: the opening of the compensating steam for leaf feeding, the opening of the moisture exhaust for leaf feeding; Environmental parameters: environmental humidity; The key factors include the opening of the compensating steam for leaf feeding, the opening of the moisture exhaust for leaf feeding, and the material flow rate for leaf feeding; The model establishment module is used to construct a prediction model, and the prediction model includes Prediction Model I and Prediction Model II; The Prediction Model I adopts a response surface design model, which takes the key factors as input values, takes the circulating air temperature for leaf feeding as the output value, takes the circulating air temperature CPK for leaf feeding as the evaluation item, and takes the moisture content at the leaf feeding outlet and the temperature at the leaf feeding outlet as reference items; The Prediction Model II adopts a neural network model, which takes the production parameters and process parameters as modeling factors, takes the circulating air temperature for leaf feeding as the output value, and takes the other factors except the circulating air temperature for leaf feeding and the circulating air temperature CPK for leaf feeding as input values; The prediction and control module is used to obtain the predicted value I of the circulating air temperature for leaf feeding based on the Prediction Model I and the predicted value II of the circulating air temperature for leaf feeding based on the Prediction Model II, and perform arithmetic comparison on the predicted value I and the predicted value II: When the deviation between the predicted values of the two models ≤ 2 degrees, normal production is carried out and the predicted value I is used as the control parameter to intelligently control the circulating air temperature for leaf feeding; When the deviation between the predicted values of the two models > 2 degrees, a warning is prompted; The model optimization module is used to optimize the Prediction Model II based on the process parameters, production parameters, and environmental parameters after the system issues a warning prompt, and return to the prediction and control module; if the system still issues a warning prompt, stop production and wait for maintenance.

2. The automatic control system for optimizing the circulating air temperature for feeding according to claim 1, wherein: The automatic control system further includes A debugging module, which is used to connect this automatic control system with the current production operation system wincc for debugging.

3. The automatic control system for optimizing the circulating air temperature for feeding according to claim 2, wherein: The model optimization module uses a neural network model, taking process parameters, production parameters, and environmental parameters as modeling factors, taking the circulating air temperature of leaf feeding as the output value, and taking the moisture content at the outlet of leaf feeding, the moisture content at the inlet of leaf feeding, the temperature at the outlet of leaf feeding, the opening degree of the compensation steam for leaf feeding, the opening degree of the moisture exhaust for leaf feeding, the material flow rate of leaf feeding, the liquid temperature of the leaf feeding material, and the environmental humidity as input values to establish the model; After the system issues a warning prompt, the prediction model II is optimized through the optimization adaptive model and returned to the prediction and control module; If the system still issues a warning prompt, stop production and wait for maintenance.

4. The automatic control system for optimizing the circulating air temperature of leaf feeding according to claim 1, wherein: The factor screening module screens out the key factors affecting the circulating air temperature and the CPK of the circulating air temperature through the C&E matrix and the FMEA factor analysis method.

5. The automatic control system for optimizing the circulating air temperature of leaf feeding according to claim 1, wherein: The prediction process of the prediction model I: ① 3-factor 2-level experiment The opening degree of the compensation steam for leaf feeding is respectively taken as Z1, Z2, Z3, the opening degree of the moisture exhaust for leaf feeding is taken as P1, P2, P3, and the material flow rate of leaf feeding is taken as W1, W2, W3. The central point experimental scheme is added, and M experiments are carried out; ② Dual response surface analysis Based on the response surface method, the optimal value range of each parameter under the above-mentioned values is obtained through the response optimizer under the condition of meeting the process standard requirements, that is, the prediction value I.

6. The automatic control system for optimizing the circulating air temperature of leaf feeding according to claim 1, wherein: In the process of the prediction and control module performing intelligent control on the circulating air temperature of leaf feeding with the prediction value I as the control parameter: Combined with the weight of the scale, the production process is divided into three stages for intelligent control respectively: At the beginning of production, the control parameter is the prediction value I plus 10%; When the material production reaches 500 KG, the control parameter is the prediction value I minus 5%; When the material production reaches 700 KG, the control parameter is the prediction value I.

Images