Method and system for estimating influence factors of heating and humidifying processes on moisture at cut tobacco drying inlet

Abstract

The invention provides a method and a system for estimating influence factors of heating and humidifying processes on moisture at a cut tobacco drying inlet. The method comprises the following steps: acquiring historical data of a heating and humidifying machine in a target area; constructing a cut tobacco dryer inlet moisture estimation model based on the historical data of the heating and humidifying machine; performing real-time analysis on measured data of a heating and humidifying machine based on the cut-tobacco dryer inlet moisture estimation model to obtain a cut-tobacco dryer inlet moisture predicted value, and feeding back the cut-tobacco dryer inlet moisture predicted value to control processing equipment; acquiring a real-time moisture value at the inlet of the cut-tobacco drier from the control processing equipment, and comparing the real-time moisture value with the predicted moisture value at the inlet of the cut-tobacco drier; and verifying and optimizing the cut tobacco dryer inlet moisture estimation model to complete automatic adjustment of the cut tobacco dryer outlet moisture in the production stage. The method can effectively improve the accuracy of the temperature calculation of the cylinder wall of the cut-tobacco dryer, and provides data support for reducing the temperature difference of the cylinder wall of the two-way cut-tobacco dryer.

Description

Technical Field

[0001] The invention belongs to the technical field of tobacco processing, and relates to an estimation method, in particular to an estimation method and system for an influence factor of a heating and humidifying process on moisture in a drying cut tobacco inlet. Background Art

[0002] The tobacco leaf drying equipment is one of the important equipment in cigarette production and is the key equipment that determines the internal quality of cigarettes. At present, in the tobacco leaf production process in the domestic tobacco industry, the roller-type thin-plate tobacco leaf drying machine is the most commonly used tobacco leaf drying equipment. Its working principle is to use the combined drying method of conduction and convection to dry and dehumidify the tobacco leaf. The working states of the thin-plate tobacco leaf drying machine include preheating, startup, production and tailing.

[0003] When the tobacco drying machine is in the startup state, the tobacco has not yet reached the outlet of the drying drum, so the dryer cannot implement feedback control of moisture. At this stage, the drum wall temperature is mainly determined by feedforward control, which depends on the understanding of the equipment model and the accuracy of parameter setting.

[0004] After the tobacco is steamed in the heating and humidifying machine, the moisture content of the tobacco increases significantly. During the commissioning and acceptance of the tobacco production line, the supplier's engineers solidified the technical parameters of the tobacco drying machine equipment, among which the moisture increase at the inlet of the tobacco drying machine was solidified as a constant parameter in the equipment parameters. However, the actual situation is that due to the changes in working conditions and production conditions during the operation of the heating and humidifying machine equipment, the moisture increase should be a real-time variable. Therefore, in order to address the uncertainty of the moisture content at the inlet of the tobacco drying machine, intelligent estimation is needed to improve the accuracy of the moisture increase parameter setting of the heating and humidifying machine.

[0005] After a period of production and use, it was found that the existing fixed parameter mode of moisture increase at the inlet of the tofu drying machine can no longer adapt to the current production mode of the tofu drying line. This is mainly reflected in:

[0006] (1) Affects the accuracy of calculation of the set value of the barrel wall temperature in the head stage of the silk drying machine.

[0007] (2) It affects the accuracy of automatic adjustment of outlet moisture during the production process of the silk drying machine.

[0008] Under existing technical conditions, due to the overflow of steam from the outlet of the heating and humidifying machine and the excessively high temperature of the tobacco, if an infrared moisture meter is installed at this workstation, the measurement accuracy cannot meet the requirements.

[0009] The results of the moisture oven method test in the early stage showed that the deviation from the set value set in the parameters was > ±0.2%, accounting for 95% of the total production batches. In extreme cases, the difference between the two was ±0.5%. Under the same equipment parameters, the deviation of moisture in the inlet of the wire drying machine was 0.1%, which affected the set value of the wall temperature of the wire drying machine by 0.5℃.

[0010] Therefore, in the operation mode of the fixed parameter mode of the moisture increase at the inlet of the wire drying machine, due to the uncertainty of the actual moisture, the impact on the set value of the wall temperature of the wire drying machine is about 5°C.

[0011] Therefore, in the prior art, during the commissioning and acceptance of the tobacco production line, the supplier's engineers solidified the technical parameters of the tobacco drying machine equipment. However, in actual situations, due to changes in operating conditions and production conditions during the operation of the heating and humidifying equipment, the increase in moisture should be a real-time changing variable, resulting in uncertainty in the moisture at the inlet of the tobacco drying machine, which in turn affects the accuracy of the calculation of the set value of the barrel wall temperature in the head stage of the tobacco drying machine, as well as potential safety hazards. Summary of the invention

[0012] In view of the shortcomings of the prior art mentioned above, the purpose of the present application is to provide a method and system for estimating the factors affecting the moisture content at the inlet of the tobacco cutter during the heating and humidification process, so as to solve the problem that in the prior art, during the commissioning and acceptance of the tobacco cutter production line, the supplier's engineers solidify the technical parameters of the tobacco cutter equipment. However, in actual situations, due to the changes in working conditions and production conditions during the operation of the heating and humidification equipment, the increase in moisture should be a real-time changing variable, resulting in uncertainty in the moisture content at the inlet of the tobacco cutter, which in turn affects the accuracy of the calculation of the set value of the barrel wall temperature at the head stage of the tobacco cutter, as well as the existence of safety hazards.

[0013] To achieve the above-mentioned purpose and other related purposes, in the first aspect, the present application provides a method for estimating the influencing factor of the heating and humidifying process on the moisture content at the wire drying inlet, comprising the following steps: obtaining historical data of the heating and humidifying machine in the target area; the historical data of the heating and humidifying machine comprises: small point data of the heating and humidifying machine equipment parameters and historical data of the moisture content at the wire drying machine inlet; constructing a wire drying machine inlet moisture estimation model based on the historical data of the heating and humidifying machine; performing real-time analysis on the measured data of the heating and humidifying machine based on the wire drying machine inlet moisture estimation model to obtain a predicted value of the moisture content at the wire drying machine inlet, and feeding the predicted value of the moisture content at the wire drying machine inlet back to a control and processing device; obtaining the real-time value of the moisture content at the wire drying machine inlet from the control and processing device, and comparing it with the predicted value of the moisture content at the wire drying machine inlet to obtain a comparison result; verifying and optimizing the wire drying machine inlet moisture estimation model based on the comparison result to complete the automatic adjustment of the moisture content at the wire drying machine outlet in the production stage.

[0014] In an implementation of the first aspect, obtaining historical data of a heating and humidifying machine in a target area includes the following steps: obtaining small point data of equipment parameters of the heating and humidifying machine through a digital acquisition system; wherein, the small point data of equipment parameters of the heating and humidifying machine include: HT steam starting flow rate, pressure before a blade expansion steam valve, pressure after a blade expansion steam valve, CV value of a blade expansion steam valve, volume flow rate of blade expansion steam, flow rate of blade expansion steam, moisture content of blade expansion inlet, temperature of blade expansion steam, temperature of blade expansion, linear speed of a blade expansion electronic scale, load code value of a blade expansion electronic scale, material flow rate of blade expansion, cumulative amount of a blade expansion electronic scale, and any one or more combinations of brand types; obtaining historical moisture data at the inlet of a wire drying machine through manual sampling.

[0015] In an implementation of the first aspect, constructing a wire drying machine inlet moisture estimation model based on the historical data of the heating and humidifying machine includes the following steps: merging the historical data of the heating and humidifying machine according to preset rules to obtain a wide data table of the heating and humidifying machine equipment; dividing the wide data table of the heating and humidifying machine equipment into a wide data table training set and a wide data table test set; performing feature screening on the wide data table training set to obtain input feature variables; constructing a wire drying machine inlet moisture estimation model based on the input feature variables and training the model; performing performance evaluation and verification on the wire drying machine inlet moisture estimation model to obtain an optimized wire drying machine inlet moisture estimation model.

[0016] In an implementation of the first aspect, feature screening is performed on the data wide table training set to obtain input feature variables, including the following steps: data preprocessing is performed on the data wide table training set to obtain a processed data wide table training set; an Xgboost model is created based on the processed data wide table training set; wherein the small point data of the heating and humidifying machine equipment parameters in the data wide table training set is used as an independent variable, and the historical data of the moisture content of the tofu drying machine inlet is used as a dependent variable; and feature screening is performed through the Xgboost model to obtain input feature variables.

[0017] In an implementation of the first aspect, constructing a moisture estimation model at the inlet of a wire drying machine based on the input mold feature variables comprises the following steps: using Xgboost as a regressor to establish a regression model between the device parameters after feature screening and the moisture at the inlet of the wire drying machine; creating an Xgboost regressor object based on the input mold feature variables; inputting the historical data of the heating and humidifying machine into the Xgboost regressor object for training to obtain the moisture estimation model at the inlet of the wire drying machine; and adjusting the parameters of the moisture estimation model at the inlet of the wire drying machine by a random search method to obtain the optimal parameter combination.

[0018] In an implementation of the first aspect, the measured data of the heating and humidifying machine is analyzed in real time based on the wire dryer inlet moisture estimation model to obtain the predicted value of the wire dryer inlet moisture, including the following steps: configuring the production line data acquisition point information and loading it into the wire dryer inlet moisture estimation model; acquiring the real-time data of the mold entry characteristics from the control processing device; inputting the real-time data of the mold entry characteristics into the wire dryer inlet moisture estimation model for analysis and calculation to obtain the predicted value of the wire dryer inlet moisture, and writing the predicted value of the wire dryer inlet moisture into the control processing device.

[0019] In an implementation of the first aspect, the real-time value of moisture at the inlet of the wire dryer is obtained from the control processing device, and compared with the predicted value of moisture at the inlet of the wire dryer, and the comparison result is obtained by the following steps: according to a preset interval time, the real-time data of the mold entry characteristics at different times are obtained; the real-time data of the mold entry characteristics at different times are input into the wire dryer inlet moisture estimation model, and the predicted values ​​of moisture at the inlet of the wire dryer at different times are analyzed; the predicted values ​​of moisture at the inlet of the wire dryer at different times are written into the control processing device through the OPCUA protocol; the real-time value of moisture at the inlet of the wire dryer is obtained from the control processing device, and the real-time value of moisture at the inlet of the wire dryer is collected by the moisture oven method; the predicted value of moisture at the inlet of the wire dryer at a certain time is compared with the real-time value of moisture at the inlet of the wire dryer to obtain a comparison result.

[0020] In an implementation of the first aspect, the wire dryer inlet moisture estimation model is verified and optimized based on the comparison result to complete the automatic adjustment of the wire dryer outlet moisture during the production stage, including the following steps: based on the comparison result, the wire dryer inlet moisture estimation model is verified online; based on the online verification result, the parameters of the wire dryer inlet moisture estimation model are adjusted and optimized; based on the optimized wire dryer inlet moisture estimation model, a new round of data collection and model estimation is performed to obtain a new round of wire dryer inlet moisture real-time value and predicted value; the wire dryer inlet moisture real-time value and predicted value obtained in each round are subtracted to obtain the wire dryer inlet moisture increase; and the wire dryer inlet moisture increase is input into the wire dryer moisture control system to enable the control processing equipment to monitor the wire dryer inlet moisture in real time and complete the automatic adjustment of the wire dryer outlet moisture during the production stage.

[0021] In the second aspect, the present application provides an estimation system for the influencing factors of the heating and humidifying process on the moisture content at the wire drying inlet, including: an acquisition module, used to acquire historical data of the heating and humidifying machine in the target area; the historical data of the heating and humidifying machine includes: small point data of the heating and humidifying machine equipment parameters and historical data of the moisture content at the wire drying machine inlet; a model construction module, used to construct a wire drying machine inlet moisture estimation model based on the heating and humidifying machine historical data; an analysis module, used to perform real-time analysis on the measured data of the heating and humidifying machine based on the wire drying machine inlet moisture estimation model, obtain a predicted value of the moisture content at the wire drying machine inlet, and feed the predicted value of the moisture content at the wire drying machine inlet back to the control and processing equipment; a comparison module, used to acquire the real-time value of the moisture content at the wire drying machine inlet from the control and processing equipment, and compare it with the predicted value of the moisture content at the wire drying machine inlet to obtain a comparison result; a verification and optimization module, used to verify and optimize the wire drying machine inlet moisture estimation model based on the comparison result, so as to complete the automatic adjustment of the moisture content at the wire drying machine outlet in the production stage.

[0022] In the last aspect, the present application provides an estimation device for the influence factor of the heating and humidification process on the moisture content at the inlet of the wire drying, comprising: a processor and a memory. The memory is used to store a computer program; the processor is connected to the memory and is used to execute the computer program stored in the memory, so that the estimation device for the influence factor of the heating and humidification process on the moisture content at the inlet of the wire drying executes the estimation method for the influence factor of the heating and humidification process on the moisture content at the inlet of the wire drying.

[0023] As described above, the method and system for estimating the influencing factor of moisture at the drying inlet of the heating and humidifying process of the present invention have the following beneficial effects:

[0024] This application provides a method for estimating the influencing factor of moisture content at the inlet of the wire drying machine during the heating and humidification process. Through the adjustment and optimization of the system software, it is deployed on three wire drying production lines in the workshop for use; it can effectively improve the accuracy of the moisture factor at the inlet of the wire drying machine; by real-time estimation of the corrected moisture increase after heating and humidification, by comparing the estimated moisture value with the oven method test, the difference between the two is >0.2%, accounting for less than 5% of the production batch, thereby effectively improving the accuracy of the calculation of the wall temperature of the wire drying machine. This application provides data support for reducing the difference in the wall temperature of the dual-path wire drying machine, with significant results and easy to promote. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1A It is a flow chart of an embodiment of a method for estimating the influence factor of the heating and humidifying process on the moisture content of the drying wire inlet of the present invention.

[0026] Figure 1B Shown is a schematic diagram of the control principle of the moisture increase of the original heating and humidifying machine before improvement.

[0027] Figure 1CIt is a schematic diagram of the improved process of the present invention for realizing intelligent estimation of moisture content at the inlet of the tofu drying machine by using machine learning technology.

[0028] Figure 2 It shows a schematic diagram of the model construction process in the method for estimating the influence factor of the heating and humidification process on the moisture content of the drying wire inlet of the present invention.

[0029] Figure 3A It shows a flow chart of S11 in the method for estimating the influence factor of the heating and humidifying process on the moisture content at the drying wire inlet of the present invention.

[0030] Figure 3B It shows a schematic diagram of sampling between the heating and humidifying machine and the tobacco drying machine on the tobacco production line.

[0031] Figure 4 It shows a flow chart of S12 in the method for estimating the influence factor of the moisture content at the drying wire inlet of the heating and humidifying process of the present invention.

[0032] Figure 5A It shows a schematic diagram of the real-time analysis process of the silk drying machine moisture estimation model in the production process of the method for estimating the influence factor of the heating and humidification process on the moisture at the silk drying inlet of the present invention.

[0033] Figure 5B It shows a flow chart of S13 in the method for estimating the influence factor of the heating and humidifying process on the moisture content at the drying wire inlet of the present invention.

[0034] Figure 6 It is a schematic flow chart of S14 in the method for estimating the influence factor of the heating and humidifying process on the moisture content at the drying wire inlet of the present invention.

[0035] Figure 7 It shows a flow chart of S15 in the method for estimating the influence factor of the moisture content at the drying wire inlet of the heating and humidifying process of the present invention.

[0036] Figure 8 It is a schematic diagram of the principle structure of a system for estimating the influence factor of the heating and humidifying process on the moisture content at the drying wire inlet in one embodiment of the present invention.

[0037] Fig. 9 It is a schematic diagram showing the principle structure of an apparatus for estimating the influence factor of the heating and humidifying process on the moisture content at the drying wire inlet in one embodiment of the present invention.

[0038] Component number description

[0039] 81 Get Module

[0040] 82 Model building modules

[0041] 83 Analysis Module

[0042] 84 Comparison module

[0043] 85 Verification and Optimization Module

[0044] 91 Processor

[0045] 92 Memory

[0046] Steps S11 to S15 DETAILED DESCRIPTION

[0047] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0048] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0049] The following will describe in detail the estimation method and system of the influencing factors of moisture in the wire drying inlet of the heating and humidification process provided in the embodiment of the present application in conjunction with the drawings. The present application changes the setting method of the parameters of the moisture increase at the inlet of the wire drying machine, replacing the original method of inputting the fixed parameter value into the moisture control system of the wire drying machine on the human-machine interface of the wire drying process with the method of calculating the estimated moisture at the HT outlet through an algorithm based on the real-time data of the heating and humidification machine process, and then subtracting it from the moisture at the inlet of the heating and humidification machine to obtain the moisture increase at the inlet of the wire drying machine and input it into the moisture control system of the wire drying machine. The present application is used to solve the problem that in the prior art, when the tobacco production line is debugged and accepted, the supplier's engineers solidify the technical parameters of the wire drying machine equipment, but the actual situation is that due to the changes in the working conditions and production conditions during the operation of the heating and humidification machine equipment, the moisture increase should be a real-time variable, thereby solving the problem of uncertainty in the moisture at the inlet of the wire drying machine.

[0050] This application uses machine learning technology to realize intelligent estimation of moisture at the inlet of the wire drying machine. First, based on the historical small point data of the humidifier related parameters in SPCD and the moisture at the inlet of the wire drying machine (i.e.: moisture at the inlet of the heating and humidifying machine + moisture increase of the heating and humidifying machine) data at some moments of manual sampling, a regression model between the moisture at the inlet of the wire drying machine and the equipment parameters of the heating and humidifying machine is established, and the gradient boosting tree Xgboost model is used to characterize the relationship between the two to obtain the moisture estimation model at the inlet of the wire drying machine. Secondly, when the moisture estimation model at the inlet of the wire drying machine trained based on historical data is applied to the actual production line, the moisture estimation model at the inlet of the wire drying machine can estimate the parameter data of the heating and humidifying machine equipment obtained from the production line PLC equipment in real time, and feed back the estimation results to the production line PLC equipment in a timely manner. Finally, the corrected moisture increase of heating and humidifying is calculated in real time to accurately control the setting of the barrel wall temperature in the head stage and the automatic PID adjustment of the moisture at the outlet of the wire drying barrel in the production stage.

[0051] See also Figure 1A , Figure 1B and Figure 1C , respectively showing a flow chart of the method for estimating the influence factor of the heating and humidifying process on the moisture content of the dried silk inlet in one embodiment of the present invention, a schematic diagram of the control principle of the moisture increase amount of the original heating and humidifying machine before improvement, and a flow chart of the present invention for realizing intelligent estimation of the moisture content of the dried silk inlet by using machine learning technology. Figure 1A , Figure 1B and Figure 1C As shown, this embodiment provides a method for estimating the influence factor of the heating and humidification process on the moisture content at the drying wire inlet.

[0052] The method for estimating the influence factor of the temperature and humidity increasing process on the moisture content of the drying wire inlet specifically comprises the following steps:

[0053] S11, obtain the historical data of the heating and humidification machine in the target area. Figure 2 and Figure 3A , respectively showing a schematic diagram of the model construction flow in the method for estimating the influence factor of the moisture content in the drying wire entrance of the heating and humidifying step of the present invention and a schematic diagram of the flow of S11 in the method for estimating the influence factor of the moisture content in the drying wire entrance of the heating and humidifying step of the present invention. Figure 2 and Figure 3A As shown, the S11 comprises the following steps:

[0054] S111, obtain the small point data of the heating and humidifying machine equipment parameters through the data acquisition system.

[0055] In this embodiment, the parameters of the field equipment at various relevant positions of the production line (such as heating and humidifying equipment, etc.) are collected through the SPCD data acquisition system. The small data of the heating and humidifying equipment parameters include but are not limited to: HT (heating and humidifying machine) steam starting flow, blade expansion steam valve front pressure, blade expansion steam valve back pressure, blade expansion steam valve CV value (Circulation Volume, flow coefficient), blade expansion steam volume flow, blade expansion steam flow, blade expansion inlet moisture content, blade expansion steam temperature, blade expansion temperature, blade expansion electronic scale linear speed, blade expansion electronic scale load code value, blade expansion material flow, blade expansion electronic scale cumulative amount, brand type and other related parameters.

[0056] Specifically, for example: the HT steam starting flow, the blade expansion steam volume flow, the blade expansion steam flow, etc. are collected through a flow sensor at the sampling point at the steam inlet; the blade expansion temperature, the blade expansion steam temperature, etc. are measured and collected by a temperature sensor; and the pressure sensor is used to measure the pressure before the blade expansion steam valve, the pressure after the blade expansion steam valve and other related parameters.

[0057] It should be noted that on-site data acquisition equipment is not limited to various types of sensors, and any equipment that can realize the data acquisition function can be used; the corresponding data acquisition systems can also use: SPCD, DCS, SIS, PLC, etc.

[0058] S112, obtaining historical data of moisture at the inlet of the tofu drying machine through manual sampling.

[0059] See also Figure 3B , showing a schematic diagram of sampling between the heating and humidifying machine and the tobacco drying machine on the tobacco production line.

[0060] In this embodiment, since the material between the heating and humidifying machine and the tobacco drying machine on the tobacco production line is high-temperature steam, it is impossible to directly collect data. Therefore, manual sampling is used to obtain data at this position.

[0061] The moisture content of the material at the inlet of the tofu drying machine includes parameters such as moisture content, temperature, material flow rate, etc. These data can be measured and recorded by sensors and measuring instruments.

[0062] S12: construct a moisture estimation model for the inlet of the tofu drying machine based on the historical data of the heating and humidifying machine. Figure 4 , which is a schematic flow chart of S12 in the method for estimating the influence factor of the moisture content in the drying wire inlet of the heating and humidifying process of the present invention. Figure 4 As shown, the S12 comprises the following steps:

[0063] S121, merging the historical data of the heating and humidifying machine according to preset rules to obtain a wide table of heating and humidifying machine equipment data.

[0064] In this embodiment, the collected parameters of the heating and humidifying machine equipment and the historical data of moisture at the inlet of the tofu drying machine obtained by manual sampling are combined according to "batch number", "sampling time", "route type", etc. to form a data wide table. The heating and humidifying machine equipment data wide table is then divided into: a data wide table training set and a data wide table test set.

[0065] Specifically, the data wide table is divided into a data wide table training set and a data wide table test set in a ratio of 6:4. For example, after the collected heating and humidifying machine equipment parameters and the historical data of the moisture content at the inlet of the tofu drying machine are combined, there are 1,000 samples in the data wide table; among them, there are 600 samples in the data wide table training set and 400 samples in the data wide table test set.

[0066] S122, performing feature screening on the data wide table training set to obtain input model feature variables.

[0067] Feature screening is an important data preprocessing step that can help remove irrelevant or redundant features and improve the training efficiency and prediction performance of the model.

[0068] In this embodiment, data preprocessing is performed on the data wide table training set to obtain a processed data wide table training set; an Xgboost model is created based on the processed data wide table training set; wherein, the small point data of the heating and humidifying machine equipment parameters in the data wide table training set are used as independent variables, and the historical data of the moisture content of the tofu drying machine inlet is used as the dependent variable; feature screening is performed through the Xgboost model to obtain the input feature variables.

[0069] Specifically, 600 sample data in the data wide table training set are preprocessed, and the preprocessed data are converted into a format acceptable to the Xgboost model, for example, including scaling or standardizing continuous variables, processing categorical variables, etc., and finally the processed data wide table training set samples are obtained.

[0070] Based on the processed data wide table training set samples, the Xgboost model is created using Python's Xgboost library. The small point data of the heating and humidifying machine equipment parameters in the data wide table training set are used as independent variables, and the historical data of the moisture content at the inlet of the tofu drying machine is used as the dependent variable. This can be achieved by setting the parameters in the Xgboost library, such as the learning rate, maximum depth, objective function, etc.

[0071] Then, feature screening is performed through correlation analysis. The correlation between the features of each sample and the target variable (such as the historical data of moisture at the inlet of the tofu drying machine) is calculated, and then the scoring method is used to sort and score them according to their usefulness to determine which features are useful. Then, based on the ranking of the features, the features with high rankings are selected as the features to be entered into the model; the number of features to be entered into the model can be determined by the cross-validation method.

[0072] S123, constructing a tofu drying machine inlet moisture estimation model based on the mold entry characteristic variables, and training the model.

[0073] In this embodiment, Xgboost is used as a regressor to establish a regression model between the device parameters after feature screening and the moisture at the inlet of the wire drying machine; an Xgboost regressor object is created based on the input feature variables; the historical data of the heating and humidifying machine is input into the Xgboost regressor object for training to obtain a moisture estimation model for the inlet of the wire drying machine; and the parameters of the moisture estimation model for the inlet of the wire drying machine are adjusted by random search to obtain the optimal parameter combination.

[0074] Specifically, Xgboost is used as a regressor to establish a regression model between the equipment parameters after feature screening and the moisture content at the inlet of the torrefaction machine. The regression model is: Y = F(X), where: Y represents the moisture content at the inlet of the torrefaction machine, X = (x1, x2, x3, x4, x5, ..., xn) represents the various equipment parameters mentioned in Step 1, such as: CV value of the blade expansion steam valve, blade expansion steam flow, blade expansion inlet moisture content, blade expansion steam temperature, etc. The extreme gradient boosting tree Xgboost is used as a regressor training model to obtain the regression function F(X) that uses the equipment parameter X to characterize the moisture content at the inlet of the torrefaction machine.

[0075] After creating an Xgboost regressor object based on the input feature variables, the historical data of the heating and humidifying machine is input into the Xgboost regressor object for training to obtain the inlet moisture estimation model of the tofu drying machine; this is achieved through the RandomizedSearchCV function in the sklearn library, which can randomly select model parameters within the specified parameter range to find the best parameter combination. Then, the training data set is input into the Xgboost regressor to let it learn the mapping relationship between the equipment parameters and the inlet moisture of the tofu drying machine.

[0076] S124, performing performance evaluation and verification on the inlet moisture estimation model of the tofu dryer to obtain an optimized inlet moisture estimation model of the tofu dryer.

[0077] In this embodiment, the trained tow-dryer inlet moisture estimation model is evaluated using samples in the data wide table test set, and its R 2, MSE, MAE, MAPE and other evaluation indicators can be used to verify the stability of the tofu drying machine inlet moisture estimation model and whether the model is overfitting.

[0078] Then optimize the moisture estimation model at the inlet of the tofu drying machine based on the evaluation results. This can be achieved by adjusting the parameters of Xgboost, such as learning rate, maximum depth, objective function, etc. Different values ​​of each parameter may affect the performance of the model, so techniques such as grid search or random search can be used to find the best parameter combination. In this step, you can also use, for example, the GridSearchCV or RandomizedSearchCV function in the sklearn library to automatically find the optimal parameter combination. After traversing all the samples in turn, find the optimal hyperparameter combination, so as to obtain the optimized moisture estimation model and mold entry features of the tofu drying machine, and save them. Among them, the mold entry feature corresponds to the equipment parameter X, also known as the independent variable of the model.

[0079] S13, based on the inlet moisture estimation model of the wire drying machine, the measured data of the heating and humidifying machine are analyzed in real time to obtain the inlet moisture prediction value of the wire drying machine, and the inlet moisture prediction value of the wire drying machine is fed back to the control processing device. Figure 5A and Figure 5B , respectively showing a real-time analysis flow diagram of the moisture estimation model of the wire drying machine in the production process in the method for estimating the moisture influencing factor of the heating and humidifying process on the wire drying inlet of the present invention and a flow diagram of S13 in the method for estimating the moisture influencing factor of the heating and humidifying process on the wire drying inlet of the present invention. Figure 5A and Figure 5B As shown, the S13 comprises the following steps:

[0080] S131, configuring and loading the production line data acquisition point information into the tow dryer inlet moisture estimation model.

[0081] In this embodiment, the moisture estimation model at the inlet of the tofu drying machine and the data acquisition point information of the production line are configured and loaded.

[0082] S132, acquiring real-time data of mold input characteristics from the control processing device.

[0083] In this embodiment, the control processing device is preferably a PLC device, and real-time device parameters (input model variables) are obtained from the PLC device.

[0084] Specifically, based on Python language, the OPC UA protocol is used to read the digital acquisition points related to the mold-in features in the PLC device, and the real-time data of the mold-in features.

[0085] For example: First, you need to install the opcua library and install it through pip, that is, pip install opcua. Then, create a new OPCUA client and connect to the server of the PLC device to obtain the data acquisition points in the PLC device. However, you need to know the node ID of the data acquisition point, which can be obtained by querying the device information. Then, you can obtain the real-time data of the data acquisition point through the node ID. Finally, after completing all operations, remember to disconnect from the PLC device.

[0086] S133, inputting the real-time data of the mold entry characteristics into the wire drying machine inlet moisture estimation model for analysis and calculation to obtain a wire drying machine inlet moisture prediction value, and writing the wire drying machine inlet moisture prediction value into the control processing device.

[0087] In this embodiment, after the model is obtained by training with historical data, the regression function F(X) between the heating and humidifying machine equipment parameter X and the inlet moisture of the tofu drying machine is known. Since the function is obtained by training and fitting with a machine learning model, it cannot be expressed as a calculation formula, but is a model file. However, when a set of X values ​​is input into the model, the inlet moisture value of the tofu drying machine can be predicted, that is, the inlet moisture value can be predicted in real time based on the equipment parameters updated in real time. Finally, the predicted value of the inlet moisture of the tofu drying machine is written into the control processing device.

[0088] S14, obtaining the real-time moisture value at the inlet of the torrefaction machine from the control processing device, and comparing it with the predicted moisture value at the inlet of the torrefaction machine to obtain a comparison result. Figure 6 , which is a schematic flow chart of S14 in the method for estimating the influence factor of the heating and humidifying process on the moisture content of the drying wire inlet of the present invention. Figure 6 As shown, the S14 comprises the following steps:

[0089] S141, acquiring real-time data of mold input features at different times according to preset intervals.

[0090] In this embodiment, the preset time interval is preferably 5 seconds, that is, every 5 seconds, the real-time data of the mold entry characteristics at the corresponding moment is obtained, and the data is input into the inlet moisture estimation model of the tofu drying machine, and the inlet moisture prediction value of the tofu drying machine is analyzed and calculated, and written into the PLC device.

[0091] S142, inputting the real-time data of the mold input characteristics at different times into the tow dryer inlet moisture estimation model, and analyzing the tow dryer inlet moisture prediction values ​​at different times.

[0092] In this embodiment, the real-time data of the mold entry characteristics at different times are analyzed according to the above steps to obtain the predicted values ​​of moisture at the inlet of the tofu drying machine at different times.

[0093] S143, writing the predicted values ​​of moisture in the tow dryer inlet at different times into the control processing device through the OPC UA protocol.

[0094] S144, obtaining the real-time value of moisture at the inlet of the tow-dryer from the control processing device. The real-time value of moisture at the inlet of the tow-dryer is collected by using a moisture oven method.

[0095] In this embodiment, samples are collected from the inlet of the wire drying machine and placed in an oven for drying. The weight of the samples before and after drying is recorded, and the moisture content of the samples is calculated based on these data. The actual moisture content is the actual moisture value at the inlet of the wire drying machine, which is used as the data basis for comparison with the predicted value of the moisture estimation model at the inlet of the wire drying machine.

[0096] S145, comparing the predicted value of moisture at the inlet of the tofu drying machine at a certain moment with the real-time value of moisture at the inlet of the tofu drying machine to obtain a comparison result.

[0097] In this embodiment, the actual moisture value at the inlet of the torrefaction machine is compared with the predicted moisture value at the inlet of the torrefaction machine to find the difference between the two. If the difference is large, it may be necessary to adjust the parameters of the model or retrain the model to improve the accuracy of the model.

[0098] Specifically, based on the difference between the predicted moisture value y' at the inlet of the tofu drying machine given by the model and the true value y obtained by manual sampling, for example, manual sampling and model prediction are performed N times at the same time, and the mean absolute error MAE and mean absolute percentage error MAPE between the two sequences Y' and Y are calculated to evaluate the degree of deviation between the predicted value and the true value. Multiple rounds of phased verification can be used to continuously verify the online prediction effect of the model.

[0099] S15, verifying and optimizing the moisture estimation model of the inlet of the tofu drying machine based on the comparison result, so as to automatically adjust the moisture of the outlet of the tofu drying machine in the production stage. Figure 7 , which is a schematic flow chart of S15 in the method for estimating the influence factor of the heating and humidifying process on the moisture content of the drying wire inlet of the present invention. Figure 7 As shown, the S15 comprises the following steps:

[0100] S151, based on the comparison result, online verification of the moisture estimation model at the tofu drying machine inlet is performed.

[0101] In this embodiment, the actual moisture value at the inlet of the torrefaction drying machine is monitored to perform online verification of the moisture estimation model at the inlet of the torrefaction drying machine. Based on the comparison results, the performance and accuracy of the model are analyzed. Then, methods such as adjusting the model parameters are used to improve the prediction performance of the model.

[0102] S152, adjusting and optimizing the parameters of the tofu drying machine inlet moisture estimation model based on the online verification result.

[0103] In this embodiment, the performance and accuracy of the moisture estimation model at the tofu drying machine inlet are analyzed based on the online verification results. If the prediction performance of the model is poor, the parameters of the model need to be adjusted. Based on the verification results, the parameters that need to be adjusted are determined, such as the learning rate, the number of iterations, the number of layers of the neural network, etc. Then, parameter optimization is performed. That is, after determining the parameters that need to be adjusted, parameter optimization is performed. An optimization algorithm (such as gradient descent, stochastic gradient descent, etc.) can be used to find the optimal parameters. The goal of parameter optimization is to minimize the difference between the model prediction value and the actual value, such as the mean square error (MSE) or the mean absolute error (MAE).

[0104] S153, based on the optimized moisture estimation model at the inlet of the tofu drying machine, a new round of data collection, model estimation, comparison, etc. is carried out to realize automatic adjustment of the moisture at the outlet of the tofu drying machine in the production stage.

[0105] In this embodiment, based on the optimized wire dryer inlet moisture estimation model, a new round of data collection and model estimation is performed to obtain a new round of real-time value and predicted value of the wire dryer inlet moisture; the real-time value and predicted value of the wire dryer inlet moisture obtained in each round are subtracted to obtain the increase in the wire dryer inlet moisture; the increase in the wire dryer inlet moisture is then input into the wire dryer moisture control system to enable the control and processing equipment to monitor the wire dryer inlet moisture in real time and complete the automatic adjustment of the wire dryer outlet moisture in the production stage.

[0106] The method for estimating the influencing factor of moisture content at the inlet of the wire drying machine provided by the heating and humidification process can be deployed on three wire drying production lines in the workshop through adjustment and optimization of the system software; it can effectively improve the accuracy of the moisture factor at the inlet of the wire drying machine; by estimating the corrected moisture increase after heating and humidification in real time, and comparing the estimated moisture value with the oven method test, the difference between the two is >0.2%, which accounts for less than 5% of the production batch, thereby effectively improving the accuracy of the calculation of the wall temperature of the wire drying machine. This application provides data support for reducing the difference in the wall temperature of the dual-path wire drying machine, with significant results and easy to promote.

[0107] The protection scope of the method for estimating the influence factor of moisture at the drying wire inlet of the heating and humidification process described in the embodiment of the present application is not limited to the execution order of the steps listed in the present embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present application are included in the protection scope of the present application.

[0108] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for estimating the influence factor of the heating and humidification process on the moisture content of the drying wire inlet as shown in FIG. 1 is implemented.

[0109] At any possible level of technical detail combination, the present application may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present application.

[0110] Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, (but not limited to) an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (non-exhaustive list) of computer readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, for example, a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination thereof. The computer readable storage medium used here is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (for example, a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0111] The computer-readable program described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device through a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives a computer-readable program instruction from the network, and forwards the computer-readable program instruction for storage in the computer-readable storage medium in each computing / processing device. The computer program instruction for performing the operation of the present application can be an assembly instruction, an instruction set architecture (ISA) instruction, a machine instruction, a machine-related instruction, a microcode, a firmware instruction, a state setting data, an integrated circuit configuration data, or a source code or object code written in any combination of one or more programming languages, wherein the programming language includes an object-oriented programming language, such as Smalltalk, C++, etc., and a procedural programming language, such as "C" language or similar programming language. Computer readable program instructions can be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a computer, or completely on a computer or server. In the case of a computer, the computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of computer readable program instructions to personalize electronic circuits, such as programmable logic circuits, field programmable gate arrays (FPGAs) or programmable logic arrays (PLAs), the electronic circuits can execute computer readable program instructions, thereby realizing various aspects of the present application.

[0112] An embodiment of the present application also provides a system for estimating the factors affecting the moisture content at the wire drying inlet during the heating and humidification process. The system for estimating the factors affecting the moisture content at the wire drying inlet during the heating and humidification process can implement the method for estimating the factors affecting the moisture content at the wire drying inlet during the heating and humidification process described in the present application. However, the device for implementing the method for estimating the factors affecting the moisture content at the wire drying inlet during the heating and humidification process described in the present application includes but is not limited to the structure of the system for estimating the factors affecting the moisture content at the wire drying inlet during the heating and humidification process listed in the present embodiment. All structural deformations and replacements of the prior art made according to the principles of the present application are included within the scope of protection of the present application.

[0113] The following will describe in detail the estimation system of the influence factor of moisture in the drying wire inlet of the heating and humidification process provided in this embodiment with reference to the drawings.

[0114] This embodiment provides a system for estimating the influence factor of the heating and humidification process on the moisture content at the drying inlet, including:

[0115] See also Figure 8 , which is a schematic diagram of the principle structure of a system for estimating the influence factor of the heating and humidifying process on the moisture content of the dried silk inlet in one embodiment of the present invention. Figure 8 As shown, the estimation system of the influence factor of the heating and humidifying process on the moisture at the drying wire inlet includes: an acquisition module 81, a model building module 82, an analysis module 83, a comparison module 84, and a verification and optimization module 85.

[0116] The acquisition module 81 is used to acquire the historical data of the heating and humidifying machine in the target area; the historical data of the heating and humidifying machine includes: the small point data of the equipment parameters of the heating and humidifying machine and the historical data of the moisture content of the tofu drying machine inlet.

[0117] In this embodiment, the parameters of the field equipment at various relevant positions of the production line (such as heating and humidifying equipment, etc.) are collected through the SPCD data acquisition system. The small data of the heating and humidifying equipment parameters include but are not limited to: HT steam starting flow, blade expansion steam valve front pressure, blade expansion steam valve back pressure, blade expansion steam valve CV value, blade expansion steam volume flow, blade expansion steam flow, blade expansion inlet moisture content, blade expansion steam temperature, blade expansion temperature, blade expansion electronic scale linear speed, blade expansion electronic scale load code value, blade expansion material flow, blade expansion electronic scale cumulative amount, brand type and other related parameters.

[0118] Specifically, for example: the HT steam starting flow, the blade expansion steam volume flow, the blade expansion steam flow, etc. are collected through a flow sensor at the sampling point at the steam inlet; the blade expansion temperature, the blade expansion steam temperature, etc. are measured and collected by a temperature sensor; and the pressure sensor is used to measure the pressure before the blade expansion steam valve, the pressure after the blade expansion steam valve and other related parameters.

[0119] It should be noted that on-site data acquisition equipment is not limited to various types of sensors, and any equipment that can realize the data acquisition function can be used; the corresponding data acquisition systems can also use: SPCD, DCS, SIS, PLC, etc.

[0120] The historical data of moisture content at the inlet of the tofu drying machine was obtained through manual sampling.

[0121] In this embodiment, since the material between the heating and humidifying machine and the tobacco drying machine on the tobacco production line is high-temperature steam, it is impossible to directly collect data. Therefore, manual sampling is used to obtain data at this position.

[0122] The moisture content of the material at the inlet of the tofu drying machine includes parameters such as moisture content, temperature, material flow rate, etc. These data can be measured and recorded by sensors and measuring instruments.

[0123] The model building module 82 is connected to the acquisition module 81 and is used to build a tofu drying machine inlet moisture estimation model based on the historical data of the heating and humidifying machine.

[0124] Based on the historical data of the heating and humidifying machine, the data is merged according to preset rules to obtain a wide table of heating and humidifying machine equipment data.

[0125] In this embodiment, the collected parameters of the heating and humidifying machine equipment and the historical data of moisture at the inlet of the tofu drying machine obtained by manual sampling are combined according to "batch number", "sampling time", "route type", etc. to form a data wide table. The heating and humidifying machine equipment data wide table is then divided into: a data wide table training set and a data wide table test set.

[0126] Specifically, the data wide table is divided into a data wide table training set and a data wide table test set in a ratio of 6:4.

[0127] The data wide table training set is subjected to feature screening to obtain the input feature variables. In this embodiment, the data wide table training set is subjected to data preprocessing to obtain the processed data wide table training set; an Xgboost model is created based on the processed data wide table training set; wherein the small point data of the heating and humidifying machine equipment parameters in the data wide table training set is used as the independent variable, and the historical data of the moisture content of the tofu drying machine inlet is used as the dependent variable; feature screening is performed through the Xgboost model to obtain the input feature variables.

[0128] Specifically, 600 sample data in the data wide table training set are preprocessed, and the preprocessed data are converted into a format acceptable to the Xgboost model, for example, including scaling or standardizing continuous variables, processing categorical variables, etc., and finally the processed data wide table training set samples are obtained.

[0129] Based on the processed data wide table training set samples, the Xgboost model is created using Python's Xgboost library. The small point data of the heating and humidifying machine equipment parameters in the data wide table training set are used as independent variables, and the historical data of the moisture content at the inlet of the tofu drying machine is used as the dependent variable. This can be achieved by setting the parameters in the Xgboost library, such as the learning rate, maximum depth, objective function, etc.

[0130] Then, feature screening is performed through correlation analysis. The correlation between the features of each sample and the target variable (such as the historical data of moisture at the inlet of the tofu drying machine) is calculated, and then the scoring method is used to sort and score them according to their usefulness to determine which features are useful. Then, based on the ranking of the features, the features with high rankings are selected as the features to be entered into the model; the number of features to be entered into the model can be determined by the cross-validation method.

[0131] A tofu drying machine inlet moisture estimation model is constructed based on the mold entry characteristic variables, and the model is trained.

[0132] In this embodiment, Xgboost is used as a regressor to establish a regression model between the device parameters after feature screening and the moisture at the inlet of the wire drying machine; an Xgboost regressor object is created based on the input feature variables; the historical data of the heating and humidifying machine is input into the Xgboost regressor object for training to obtain a moisture estimation model for the inlet of the wire drying machine; and the parameters of the moisture estimation model for the inlet of the wire drying machine are adjusted by random search to obtain the optimal parameter combination.

[0133] Specifically, Xgboost is used as a regressor to establish a regression model between the equipment parameters after feature screening and the moisture content at the inlet of the torrefaction machine. The regression model is: Y = F(X), where: Y represents the moisture content at the inlet of the torrefaction machine, X = (x1, x2, x3, x4, x5, ..., xn) represents the various equipment parameters mentioned in Step 1, such as: CV value of the blade expansion steam valve, blade expansion steam flow, blade expansion inlet moisture content, blade expansion steam temperature, etc. The extreme gradient boosting tree Xgboost is used as a regressor training model to obtain the regression function F(X) that uses the equipment parameter X to characterize the moisture content at the inlet of the torrefaction machine.

[0134] After creating an Xgboost regressor object based on the input feature variables, the historical data of the heating and humidifying machine is input into the Xgboost regressor object for training to obtain the inlet moisture estimation model of the tofu drying machine; this is achieved through the RandomizedSearchCV function in the sklearn library, which can randomly select model parameters within the specified parameter range to find the best parameter combination. Then, the training data set is input into the Xgboost regressor to let it learn the mapping relationship between the equipment parameters and the inlet moisture of the tofu drying machine.

[0135] The performance of the inlet moisture estimation model of the tofu dryer is evaluated and verified to obtain an optimized inlet moisture estimation model of the tofu dryer.

[0136] In this embodiment, the trained tow-dryer inlet moisture estimation model is evaluated using samples in the data wide table test set, and its R 2 , MSE, MAE, MAPE and other evaluation indicators can be used to verify the stability of the tofu drying machine inlet moisture estimation model and whether the model is overfitting.

[0137] Then optimize the moisture estimation model at the inlet of the tofu drying machine based on the evaluation results. This can be achieved by adjusting the parameters of Xgboost, such as learning rate, maximum depth, objective function, etc. Different values ​​of each parameter may affect the performance of the model, so techniques such as grid search or random search can be used to find the best parameter combination. In this step, you can also use, for example, the GridSearchCV or RandomizedSearchCV function in the sklearn library to automatically find the optimal parameter combination. After traversing all the samples in turn, find the optimal hyperparameter combination, so as to obtain the optimized moisture estimation model and mold entry features of the tofu drying machine, and save them.

[0138] The analysis module 83 is used to perform real-time analysis on the measured data of the heating and humidifying machine based on the inlet moisture estimation model of the wire drying machine, obtain the inlet moisture prediction value of the wire drying machine, and feed back the inlet moisture prediction value of the wire drying machine to the control processing device.

[0139] The production line data acquisition point information is configured and loaded into the tow dryer inlet moisture estimation model.

[0140] In this embodiment, the moisture estimation model at the inlet of the tofu drying machine and the data acquisition point information of the production line are configured and loaded.

[0141] The real-time data of the mold input characteristics is obtained from the control processing device.

[0142] In this embodiment, the control processing device is preferably a PLC device, and real-time device parameters (input model variables) are obtained from the PLC device.

[0143] Specifically, based on Python language, the OPC UA protocol is used to read the digital acquisition points related to the mold-in features in the PLC device, and the real-time data of the mold-in features.

[0144] The real-time data of the mold entry characteristics is input into the wire drying machine inlet moisture estimation model for analysis and calculation to obtain the wire drying machine inlet moisture prediction value, and the wire drying machine inlet moisture prediction value is written into the control processing device.

[0145] In this embodiment, after the model is obtained by training with historical data, the regression function F(X) between the heating and humidifying machine equipment parameter X and the inlet moisture of the tofu drying machine is known. Since the function is obtained by training and fitting with a machine learning model, it cannot be expressed as a calculation formula, but is a model file. However, when a set of X values ​​is input into the model, the inlet moisture value of the tofu drying machine can be predicted, that is, the inlet moisture value can be predicted in real time based on the equipment parameters updated in real time. Finally, the predicted value of the inlet moisture of the tofu drying machine is written into the control processing device.

[0146] The comparison module 84 is used to obtain the real-time value of moisture at the inlet of the tow dryer from the control processing device, and compare it with the predicted value of moisture at the inlet of the tow dryer to obtain a comparison result.

[0147] In this embodiment, the real-time data of the mold entry characteristics at different times are obtained according to the preset interval time. The real-time data of the mold entry characteristics at different times are input into the wire drying machine inlet moisture estimation model, and the predicted values ​​of the wire drying machine inlet moisture at different times are analyzed. The predicted values ​​of the wire drying machine inlet moisture at different times are written into the control processing device through the OPCUA protocol. The real-time value of the wire drying machine inlet moisture is obtained from the control processing device. The real-time value of the wire drying machine inlet moisture is collected by the moisture oven method. The predicted value of the wire drying machine inlet moisture at a certain time is compared with the real-time value of the wire drying machine inlet moisture to obtain a comparison result.

[0148] Specifically, the preset time interval is preferably 5 seconds. That is, every 5 seconds, the real-time data of the mold entry characteristics at the corresponding moment is obtained, and the data is input into the moisture estimation model of the wire drying machine entrance, and the moisture prediction value of the wire drying machine entrance is analyzed and calculated, and written into the PLC device. According to the above steps, the real-time data of the mold entry characteristics at different moments are analyzed respectively to obtain the moisture prediction value of the wire drying machine entrance at different moments. Samples are collected from the entrance of the wire drying machine and placed in an oven for drying. The weight of the samples before and after drying is recorded, and the moisture content of the samples is calculated based on these data. And this actual moisture content is the actual moisture value at the entrance of the wire drying machine, which is used as the data basis for comparing with the predicted value of the moisture estimation model at the entrance of the wire drying machine. By comparing the actual moisture value at the entrance of the wire drying machine with the moisture prediction value at the entrance of the wire drying machine, the difference between the two can be found. If the difference is large, it may be necessary to adjust the parameters of the model or retrain the model to improve the accuracy of the model.

[0149] The verification and optimization module 85 is used to verify and optimize the inlet moisture estimation model of the tofu dryer based on the comparison result, so as to complete the automatic adjustment of the outlet moisture of the tofu dryer in the production stage.

[0150] In this embodiment, based on the comparison results, the wire dryer inlet moisture estimation model is verified online; based on the online verification results, the parameters of the wire dryer inlet moisture estimation model are adjusted and optimized; based on the optimized wire dryer inlet moisture estimation model, a new round of data collection and model estimation, comparison, etc. is performed to achieve automatic adjustment of the wire dryer outlet moisture in the production stage.

[0151] Specifically, the actual moisture value at the inlet of the torrefaction dryer is used to verify the moisture estimation model at the inlet of the torrefaction dryer online. Based on the comparison results, the performance and accuracy of the model are analyzed. Then, methods such as adjusting the model parameters are used to improve the prediction performance of the model.

[0152] According to the online verification results, the performance and accuracy of the tofu drying machine inlet moisture estimation model are analyzed. If the model's prediction performance is poor, the model parameters need to be adjusted. According to the verification results, the parameters that need to be adjusted are determined, such as the learning rate, the number of iterations, the number of layers of the neural network, etc. Then, parameter optimization is performed. That is, after determining the parameters that need to be adjusted, parameter optimization is performed. Optimization algorithms (such as gradient descent, stochastic gradient descent, etc.) can be used to find the optimal parameters. The goal of parameter optimization is to minimize the difference between the model's predicted value and the actual value, such as the mean square error (MSE) or the mean absolute error (MAE).

[0153] Based on the optimized wire dryer inlet moisture estimation model, a new round of data collection and model estimation is carried out to obtain a new round of real-time value and predicted value of the wire dryer inlet moisture; the real-time value and predicted value of the wire dryer inlet moisture obtained in each round are uploaded to the control processing device, so that the control processing device can monitor the wire dryer inlet moisture in real time and complete the automatic adjustment of the wire dryer outlet moisture in the production stage.

[0154] Estimation model for the factors affecting moisture content at the wire drying inlet during the heating and humidification process. A system for estimating the factors affecting moisture content at the wire drying inlet during the heating and humidification process can be built. This system can be deployed on three wire drying production lines in the workshop through adjustment and optimization of the system software. This can effectively improve the accuracy of the moisture factor at the wire drying machine inlet.

[0155] It should be noted that it should be understood that the division of the various modules of the above system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the x module can be a separately established processing element, or it can be integrated in a certain chip of the above system for implementation. In addition, it can also be stored in the memory of the above system in the form of program code, and called and executed by a certain processing element of the above system. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0156] The above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital signal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0157] See also Fig. 9 , which is a schematic diagram of the principle structure of an apparatus for estimating the influence factor of the moisture content in the drying wire entrance of the heating and humidifying process of the present invention in one embodiment. Fig. 9 As shown, this embodiment provides a device for estimating the influence factor of the heating and humidifying process on the moisture content at the wire drying inlet, and the device for estimating the influence factor of the heating and humidifying process on the moisture content at the wire drying inlet comprises: a processor 91 and a memory 92; the memory 92 is used to store computer programs; the processor 91 is connected to the memory 92, and is used to execute the computer program stored in the memory 92, so that the device for estimating the influence factor of the heating and humidifying process on the moisture content at the wire drying inlet performs each step of the method for estimating the influence factor of the heating and humidifying process on the moisture content at the wire drying inlet as described above.

[0158] Preferably, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0159] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0160] In summary, the method and system for estimating the influencing factor of moisture at the drying inlet of the heating and humidifying process provided in the present application have the following beneficial effects:

[0161] The method for estimating the influencing factor of moisture content at the inlet of the wire drying machine provided by the heating and humidification process provided in this application is deployed on three wire drying production lines in the workshop through adjustment and optimization of the system software; it can effectively improve the accuracy of the moisture factor at the inlet of the wire drying machine; by real-time estimation of the corrected moisture increase after heating and humidification, by comparing the estimated moisture value with the oven method test, the difference between the two is >0.2% and accounts for less than 5% of the production batch, thereby effectively improving the accuracy of the calculation of the wall temperature of the wire drying machine. This application provides data support for reducing the difference in the wall temperature of the dual-path wire drying machine, with significant results and easy to promote.

[0162] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for estimating the influence factor of the heating and humidification process on the moisture content at the drying inlet, characterized in that: The following steps are involved: Acquire historical data of the heating and humidifying machine in the target area; the historical data of the heating and humidifying machine includes: small point data of equipment parameters of the heating and humidifying machine and historical data of moisture content in the inlet of the tofu drying machine; Constructing a moisture estimation model for the inlet of the tofu drying machine based on the historical data of the heating and humidifying machine; Based on the inlet moisture estimation model of the wire drying machine, the measured data of the heating and humidifying machine is analyzed in real time to obtain the inlet moisture prediction value of the wire drying machine, and the inlet moisture prediction value of the wire drying machine is fed back to the control processing device; Acquire the real-time moisture value at the inlet of the wire drying machine from the control processing device, and compare it with the predicted moisture value at the inlet of the wire drying machine to obtain a comparison result; Based on the comparison results, the inlet moisture estimation model of the wire drying machine is verified and optimized to complete the automatic adjustment of the outlet moisture of the wire drying machine during the production stage.

2. The method for estimating the influence factor of the heating and humidifying process on the moisture content at the drying inlet of the strands according to claim 1 is characterized in that: Obtaining historical data of heating and humidifying machines in a target area includes the following steps: Obtain the small data of the heating and humidifying equipment parameters through the data acquisition system; wherein, the small data of the heating and humidifying equipment parameters include: HT steam starting flow, blade expansion steam valve front pressure, blade expansion steam valve back pressure, blade expansion steam valve CV value, blade expansion steam volume flow, blade expansion steam flow, blade expansion inlet moisture content, blade expansion steam temperature, blade expansion temperature, blade expansion electronic scale linear speed, blade expansion electronic scale load code value, blade expansion material flow, blade expansion electronic scale cumulative amount, brand type, any one or more combinations thereof; The historical data of moisture content at the inlet of the tofu drying machine was obtained through manual sampling.

3. The method for estimating the influence factor of the heating and humidifying process on the moisture at the drying inlet of the strands according to claim 1, characterized in that: Constructing a tofu drying machine inlet moisture estimation model based on the historical data of the heating and humidifying machine comprises the following steps: Based on the historical data of the heating and humidifying machine, the historical data are merged according to preset rules to obtain a wide table of heating and humidifying machine equipment data; Dividing the wide data table of the heating and humidifying machine equipment into a wide data table training set and a wide data table test set; Performing feature screening on the data wide table training set to obtain input model feature variables; Building a tofu drying machine inlet moisture estimation model based on the mold entry characteristic variables and training the model; The performance of the inlet moisture estimation model of the tofu dryer is evaluated and verified to obtain an optimized inlet moisture estimation model of the tofu dryer.

4. The method for estimating the influence factor of the heating and humidifying process on the moisture at the drying inlet of the strands according to claim 3 is characterized in that: Performing feature screening on the data wide table training set to obtain input model feature variables includes the following steps: Performing data preprocessing on the data wide table training set to obtain a processed data wide table training set; An Xgboost model is created based on the processed data wide table training set; wherein the small point data of the heating and humidifying machine equipment parameters in the data wide table training set is used as an independent variable, and the historical data of the moisture content of the tofu drying machine inlet is used as a dependent variable; The Xgboost model is used to perform feature screening to obtain input feature variables.

5. The method for estimating the influence factor of the heating and humidifying process on the moisture content at the drying inlet of the strands according to claim 3 is characterized in that: Constructing a tofu drying machine inlet moisture estimation model based on the mold input characteristic variables includes the following steps: Using Xgboost as a regressor, a regression model between the equipment parameters after feature screening and the moisture content at the inlet of the tofu drying machine was established; Create an Xgboost regressor object based on the input feature variables; Input the heating and humidifying machine historical data into the Xgboost regressor object for training to obtain a tofu drying machine inlet moisture estimation model; The parameters of the moisture estimation model at the tofu drying machine inlet are adjusted by random search to obtain the optimal parameter combination.

6. The method for estimating the influence factor of the heating and humidifying process on the moisture at the drying inlet of the strands according to claim 1, characterized in that: The following steps are included in the following: performing real-time analysis on the measured data of the heating and humidifying machine based on the inlet moisture estimation model of the tofu drying machine to obtain the inlet moisture prediction value of the tofu drying machine: The production line data acquisition point information is configured and loaded into the tow-dryer inlet moisture estimation model; Acquire the real-time data of the mold input characteristics from the control processing device; The real-time data of the mold entry characteristics is input into the wire drying machine inlet moisture estimation model for analysis and calculation to obtain the wire drying machine inlet moisture prediction value, and the wire drying machine inlet moisture prediction value is written into the control processing device.

7. The method for estimating the influence factor of the heating and humidifying process on the moisture at the drying inlet of the strands according to claim 1, characterized in that: Acquiring the real-time moisture value of the inlet of the wire drying machine from the control processing device and comparing it with the predicted moisture value of the inlet of the wire drying machine to obtain the comparison result includes the following steps: According to the preset interval time, obtain the real-time data of the mold input characteristics at different times; Input the real-time data of the mold entry characteristics at different times into the inlet moisture estimation model of the tofu drying machine, and analyze the predicted values ​​of the inlet moisture of the tofu drying machine at different times; The predicted moisture value of the tofu drying machine inlet at different times is written into the control processing device through the OPCUA protocol; Acquire the real-time moisture value at the inlet of the wire drying machine from the control processing device, and collect the real-time moisture value at the inlet of the wire drying machine by using the moisture oven method; The predicted value of moisture at the inlet of the torrefaction machine at a certain moment is compared with the real-time value of moisture at the inlet of the torrefaction machine to obtain a comparison result.

8. The method for estimating the influence factor of the heating and humidifying process on the moisture at the drying inlet of the strands according to claim 1, characterized in that: The method of verifying and optimizing the moisture estimation model of the inlet of the wire drying machine based on the comparison result to automatically adjust the moisture at the outlet of the wire drying machine in the production stage includes the following steps: Based on the comparison results, the moisture estimation model for the tofu drying machine inlet is verified online; Adjusting and optimizing the parameters of the moisture estimation model at the inlet of the tofu drying machine based on the online verification results; Based on the optimized moisture estimation model at the inlet of the tofu drying machine, a new round of data collection and model estimation is carried out to obtain the real-time value and predicted value of the moisture at the inlet of the tofu drying machine in a new round; Subtract the real-time value of moisture at the inlet of the silk drying machine obtained in each round from the predicted value to obtain the increase in moisture at the inlet of the silk drying machine; The increase in moisture at the inlet of the wire dryer is then input into the moisture control system of the wire dryer, so that the control processing equipment can monitor the moisture at the inlet of the wire dryer in real time and automatically adjust the moisture at the outlet of the wire dryer during the production stage.

9. A system for estimating the influence factor of the heating and humidification process on the moisture content at the drying inlet, characterized in that: include: An acquisition module is used to acquire historical data of heating and humidifying machines in a target area; The historical data of the heating and humidifying machine include: small point data of equipment parameters of the heating and humidifying machine and historical data of moisture content in the inlet of the tofu drying machine; A model building module, used to build a tofu drying machine inlet moisture estimation model based on the historical data of the heating and humidifying machine; An analysis module is used to perform real-time analysis on the measured data of the heating and humidifying machine based on the moisture estimation model for the inlet of the wire drying machine, obtain a predicted value of the moisture for the inlet of the wire drying machine, and feed the predicted value of the moisture for the inlet of the wire drying machine back to the control processing device; A comparison module, used for obtaining the real-time value of moisture at the inlet of the wire drying machine from the control processing device, and comparing it with the predicted value of moisture at the inlet of the wire drying machine to obtain a comparison result; The verification and optimization module is used to verify and optimize the moisture estimation model at the inlet of the tow dryer based on the comparison result, so as to complete the automatic adjustment of the moisture at the outlet of the tow dryer in the production stage.

10. A device for estimating the influence factor of the heating and humidification process on the moisture content at the drying inlet, characterized in that: include: Processor and memory; The memory is used to store computer programs; The processor is connected to the memory and is used to execute the computer program stored in the memory so that the device for estimating the influence factor of the moisture at the wire drying inlet of the heating and humidification process executes the method for estimating the influence factor of the moisture at the wire drying inlet of the heating and humidification process described in any one of claims 1 to 8.

Images