CN118034195A - Method and device for controlling water addition to materials during material flow through production equipment - Google Patents

Method and device for controlling water addition to materials during material flow through production equipment Download PDF

Info

Publication number
CN118034195A
CN118034195A CN202410044062.6A CN202410044062A CN118034195A CN 118034195 A CN118034195 A CN 118034195A CN 202410044062 A CN202410044062 A CN 202410044062A CN 118034195 A CN118034195 A CN 118034195A
Authority
CN
China
Prior art keywords
water
historical
value
obtaining
water content
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202410044062.6A
Other languages
Chinese (zh)
Inventor
李毅
王桂铝
王征
皋元崚
张清
王荣康
杨学良
张一圳
陈定玮
刘友娟
赵婷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hongta Tobacco Group Co Ltd
Original Assignee
Hongta Tobacco Group Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hongta Tobacco Group Co Ltd filed Critical Hongta Tobacco Group Co Ltd
Priority to CN202410044062.6A priority Critical patent/CN118034195A/en
Publication of CN118034195A publication Critical patent/CN118034195A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32252—Scheduling production, machining, job shop

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The application provides a method and a device for controlling water addition of materials when the materials flow through production equipment. The method comprises the following steps: obtaining a prediction model; correcting the historical predicted value output by the predicted model to obtain a corrected value of the predicted model; when the material flows through the production equipment, the current inlet material flow rate when the material enters the production equipment, the current inlet material water content when the material enters the production equipment and the current water adding amount of the material when the material is in the production equipment are used as input variables of a prediction model, so that a current prediction value output by the prediction model is obtained; based on the current predicted value and the corrected value, obtaining the water content of the predicted outlet material; obtaining an expected adjustment water quantity for the material based on a preset value for predicting the water content of the outlet material and the water content of the outlet material; the water addition to the material is controlled based on the desired amount of water to be adjusted. The method and the device provided by the application can enable the control of water addition to the materials to be more accurate and stable.

Description

Method and device for controlling water addition to materials during material flow through production equipment
Technical Field
The application relates to the field of material water adding control, in particular to a method and a device for controlling material water adding when materials flow through production equipment.
Background
In industrial production processes, it is sometimes necessary to control the water addition of the material so that the material, after flowing through the production equipment, has a desired water content of the outlet material, so as to facilitate the production and processing of subsequent processes. Taking tobacco production as an example, in the redrying and moistening process, the stability of the moisture content of the redrying and moistening outlet has a great influence on the processing and production of subsequent procedures. In the traditional production process, the value of the water adding amount is obtained based on feedback calculation, and the feedback-based method is that when the sensor detects that the actual value of the water content of the outlet material is higher than a set value, the value of the water adding amount starts to be reduced; when the actual value of the water content of the outlet material is detected to be lower than the set value, the value of the water adding amount starts to increase. This way of adjustment has a certain hysteresis.
In the related art, there are some water addition control methods based on feedforward. For example, a model is built based on historical data, variables such as inlet water content and the like are acquired based on sensors, the variables are used as input variables of the model, and the expected regulated water quantity is calculated. However, the feedforward control mode has high precision requirement on the model, the expected adjustment water quantity is influenced by various factors, the input variable of the model in the related technology is single, the precision is low, the feedforward control effect is poor, and the fluctuation of the water content of the outlet is caused; in addition, a large number of interference factors exist in the production process, such as the water absorption characteristic, the air return temperature and the like of materials, and the feed-forward water adding amount directly calculated according to a fixed model cannot enable the water content of an outlet to be stable within an ideal range.
Disclosure of Invention
To at least partially solve the above problems, according to one aspect of the present application, an embodiment of the present application provides a method for controlling water addition to a material as the material flows through a production facility, comprising: obtaining a prediction model; correcting a historical predicted value output by the prediction model based on a plurality of historical data when the material flows through the production equipment to obtain a corrected value for the prediction model, wherein the historical data comprise a historical inlet material flow rate when the material enters the production equipment, a historical inlet material water content when the material enters the production equipment, a historical outlet material water content when the material flows out of the production equipment and a historical water adding amount of the material in the production equipment; when the material flows through the production equipment, the current inlet material flow rate when the material enters the production equipment, the current inlet material water content when the material enters the production equipment and the current water adding amount of the material when the material is in the production equipment are used as input variables of a prediction model, so that a current prediction value output by the prediction model is obtained; based on the current predicted value and the corrected value, obtaining the water content of the predicted outlet material; obtaining an expected adjustment water quantity for the material based on a preset value for predicting the water content of the outlet material and the water content of the outlet material; controlling the water addition to the material based on the desired amount of water to be adjusted; wherein, the prediction model is established by the following steps: obtaining a plurality of historical data of the material flowing through the production equipment; obtaining an initial model, taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount as input variables of the initial model, taking the historical outlet material water content as output variables of the initial model, training the initial model, and establishing a prediction model.
In some preferred embodiments, correcting the historical predicted value output by the predictive model based on a plurality of historical data of the material flowing through the production facility to obtain a corrected value for the predictive model includes: obtaining historical data of the latest M sampling periods; taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount of historical data of each sampling period in M sampling periods as input variables of a prediction model to respectively obtain a first prediction value output by the prediction model; obtaining a first predicted value and a difference value of the water content of a historical outlet material in the corresponding historical data; based on a plurality of differences corresponding to the M sampling periods, a first average value of the plurality of differences is obtained, and based on the first average value, a correction value of the prediction model is obtained.
In some preferred embodiments, correcting the historical predicted value output by the prediction model based on a plurality of historical data of the material flowing through the production facility to obtain a corrected value for the prediction model, further comprising: obtaining historical data of N recent sampling periods, wherein N is less than M; taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount of historical data of each sampling period in N sampling periods as input variables of a prediction model to respectively obtain second predicted values output by the prediction model; obtaining a second predicted value and a difference value of the water content of the historical outlet material in the corresponding historical data; obtaining a second average value of the plurality of difference values based on the plurality of difference values respectively corresponding to the N sampling periods, wherein the first average value has a first weight, and the second average value has a second weight; and obtaining a correction value for the prediction model based on the first average value and the second average value and the first weight and the second weight.
In some preferred embodiments, the first weight is greater than the second weight.
In some preferred embodiments, obtaining the desired adjusted water amount for the material based on the predicted outlet material water content and the preset value of the outlet material water content comprises: obtaining the water absorption coefficient of the material; and obtaining the expected regulating water quantity of the material based on the difference value between the predicted water content of the outlet material and the preset value of the water content of the outlet material, the current inlet material flow and the water absorption coefficient.
In some preferred embodiments, obtaining the water absorption coefficient of the material comprises: calculating a difference value between the water content of the historical inlet material and the water content of the corresponding historical outlet material in the historical data; establishing a linear relation between a difference value between the water content of the historical inlet material and the water content of the corresponding historical outlet material, and the flow rate of the historical inlet material and the historical water adding amount; based on the linear relation, the water absorption coefficient of the material is obtained.
In some preferred embodiments, the production facility charges the material through a first and a second water charger, the first and second water charger being disposed at different locations of the production facility along a flow direction of the material, wherein the historical water charge comprises a historical first water charger charge and a historical second water charger charge, the historical first water charger charge and the historical second water charger charge being two input variables of the raw model and the predictive model; the current water adding quantity comprises the current water adding quantity of the first water adding device and the current water adding quantity of the second water adding device, and the current water adding quantity of the first water adding device and the current water adding quantity of the second water adding device are used as two input variables of a prediction model.
In some preferred embodiments, the first water applicator has a nominal water addition, wherein controlling the water addition to the material based on the desired amount of water adjustment comprises: controlling the first water feeder to feed water to the material according to the current water feeding amount of the first water feeder according to the water feeding amount adjusted by the expected water feeding amount; and controlling the first water feeder to feed water to the material in the rated water feeding amount in response to the fact that the current water feeding amount of the first water feeder is larger than the rated water feeding amount according to the expected water feeding amount, and controlling the second water feeder to feed water to the material by taking the difference between the current water feeding amount of the first water feeder and the rated water feeding amount according to the expected water feeding amount.
In some preferred embodiments, the various variables in the history data are Kalman filtered and/or time delays of the various variables are aligned prior to training the initial model using the history data.
In some preferred embodiments, the initial model and the predictive model are long-term memory neural network models.
According to another aspect of the present application, there is provided an apparatus for controlling the addition of water to a material as it flows through a production facility, the apparatus comprising a memory and a processor, the memory for storing instructions; the processor is coupled to the memory and is configured to execute the method provided by any of the embodiments of the present application based on the instructions stored in the memory.
According to the method and the device for controlling water adding of the materials when the materials flow through the production equipment, the prediction model is used for predicting the water content of the outlet materials instead of directly predicting the water adding amount, so that influence factors between input variables and output variables of the prediction model are reduced, and the accuracy of the prediction model is improved; and the water content of the predicted outlet material is determined together based on the predicted value and the corrected value, the water content of the predicted outlet material can be adjusted timely through adjusting the corrected value, the situation that the influence of interference factors cannot be processed timely when the prediction is performed through a fixed model is avoided, and the control stability is improved; in addition, the influence of factors such as inlet material flow and the like on the water content of the outlet material is considered by the input variables of the prediction model, and the prediction model is more reasonable and accurate.
Drawings
FIG. 1 is a flow chart of a method for controlling the addition of water to a material as it flows through a production facility, in accordance with an embodiment of the present application.
Detailed Description
The technical solutions of the preferred embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. It is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments, and that the various embodiments and technical features can be combined with each other without collision. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to fall within the scope of the application as claimed.
The embodiment of the application provides a method for controlling water addition to a material when the material flows through production equipment. The method provided by the embodiment of the application can be applied to various working procedures or operation flows which need to control the water content or the water adding amount of materials, the type and the material type of the working procedures or the operation flows are not limited, and taking the cigarette production process as an example, the method provided by the embodiment of the application can be applied to the leaf wetting and water adding control of the redrying working procedures, and can also be applied to the water adding control of the working procedures such as shred making and moisture regaining, shred making and leaf wetting.
FIG. 1 is a flow chart of a method for controlling the addition of water to a material as it flows through a production facility, in accordance with an embodiment of the present application. As shown in fig. 1, the method provided by the embodiment of the application may include the following steps: step S101, obtaining a prediction model; step S102, correcting a historical predicted value output by a predicted model to obtain a corrected value of the predicted model; step S103, obtaining a current predicted value output by a prediction model; step S104, based on the current predicted value and the corrected value, obtaining the predicted outlet material water content; step S105, obtaining expected adjustment water quantity of the materials based on the preset values of the predicted outlet material water content and the outlet material water content; and step S106, controlling water addition of the materials based on the expected water amount adjustment.
Specifically, in step S101, the prediction model is established by: obtaining a plurality of historical data of the material flowing through the production equipment; obtaining an initial model, taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount as input variables of the initial model, taking the historical outlet material water content as output variables of the initial model, training the initial model, and establishing a prediction model. After the initial model is trained by using the historical data, the influence coefficients of all variables in the initial model can be adjusted to form a final prediction model, and the final prediction model can be better fit with the historical data. Illustratively, the initial model and the predictive model may be models built based on decision trees, BP neural networks, multiple regressions, least squares, constrained least squares, and the like. Preferably, in some embodiments, the initial model and the predictive model are long-short term memory (LSTM) neural network models that have a better fit to the time series tasks.
In step S102, the historical predicted value output by the prediction model is corrected based on a plurality of historical data when the material flows through the production equipment, so as to obtain a correction value for the prediction model, where the historical data includes a historical inlet material flow rate when the material enters the production equipment, a historical inlet material water content when the material enters the production equipment, a historical outlet material water content when the material flows out of the production equipment, and a historical water adding amount of the material in the production equipment. The historical data can be acquired through the sensor when the material flows through the production equipment, and a fixed sampling period can be set when the historical data is acquired, and a group of historical data is acquired in each period.
The historical inlet material flow is the amount of material entering the production facility per unit time contained in the historical data, reflecting how much material is flowing through the production facility; the water content of the material at the inlet of the production equipment, which is contained in the historical data, can be acquired near the inlet of the production equipment, and reflects the water content of the material before water is added in the process of flowing through the production equipment; the water content of the material at the outlet of the production equipment, which is contained in the historical data, can be collected near the outlet of the production equipment, and reflects the water content of the material after being added with water in the process of flowing through the production equipment; the historical water adding amount is the amount of water for supplementing the material in the process of flowing through the production equipment by the material contained in the historical data, and can be obtained through a flowmeter arranged on the water adding device, and reflects the amount of water additionally obtained by the material after flowing through the production equipment. According to the application, the historical data containing the variables can be used for more reasonably and accurately establishing the prediction model.
It will be appreciated that the data for each variable contained in the historical data is a measurement obtained by the sensor with some error from the true value, and therefore, in some embodiments, the values of each variable in the historical data are kalman filtered to varying degrees before the initial model is trained using the historical data to make the predictive model more accurate.
In addition, the material flows through the production facility in a certain time, so that the history data is actually time-series data. Specifically, assuming that the time required from entering the production equipment to being added with water is t 1 and the time required for the material to flow through the production equipment is t 2, the water adding amount corresponding to the historical inlet material flow rate and the historical inlet material water content at the time k is at the time k+t 1, and the corresponding historical outlet material water content is at the time k+t 2. Thus, in some embodiments, the time delays of the various variables in the history data are aligned to reduce errors caused by the time delays before training the initial model using the history data.
In this embodiment, after a prediction model is obtained through historical data training, a specified number of historical data are substituted into the prediction model, so that the prediction model outputs a historical predicted value, then the output historical predicted value is compared with the water content of the historical outlet material in the corresponding historical data to obtain the deviation of the historical predicted value and the recorded value, the deviation of the plurality of historical data is averaged to obtain a correction value of the prediction model, and the correction value can correct the deviation between the predicted value and the actual recorded value of the prediction model. In this embodiment, the history data used for correcting the prediction model may be the history data used for training the prediction model, the history data collected recently in the production process, or a combination of the two. In this embodiment, the prediction model does not directly predict the desired water quantity, but directly predicts the water content of the outlet material, so that the output predicted value can be corrected in time according to the measured value of the water content of the outlet material, and the corrected value of the prediction model can be obtained in time.
In step S103, when the material flows through the production device, the current inlet material flow rate when the material enters the production device, the current inlet material water content when the material enters the production device, and the current water adding amount of the material when the material is in the production device are used as input variables of the prediction model, so as to obtain the current prediction value output by the prediction model. After the prediction model is obtained, in the process of predicting the water content of the outlet material under the current condition by using the prediction model, the corresponding current inlet material flow, the current inlet material water content and the current water adding amount are only required to be collected and substituted into the prediction model, so that the corresponding current prediction value is calculated.
In step S104, the predicted outlet material water content is obtained based on the current predicted value and the corrected value. For example, the sum of the current predicted value and the corrected value may be used as the predicted outlet material moisture content.
In step S105, a desired amount of water to be adjusted for the material is obtained based on the predicted outlet material water content and the preset value of the outlet material water content. The preset value of the water content of the outlet material is the value of the water content that the material is expected to reach, which can be determined according to different requirements. In this embodiment, a deviation between the water content of the outlet material and the preset value of the water content of the outlet material is predicted, and the deviation is reduced by increasing or decreasing the water of the desired water amount to be adjusted based on the current water adding amount, so that the water content of the material is as close as possible to the preset value of the water content of the outlet material, thereby being beneficial to the production and processing of the subsequent procedures.
In step S105, the water addition to the material is controlled based on the desired amount of water to be adjusted. After the desired amount of water is obtained, the water addition to the material may be controlled, for example, by controlling a water adder for adding water to the material flowing through the production facility, such that the water adder adds water to the material based on the desired amount of water. The water feeder can be provided with a flowmeter and an electromagnetic valve, the water quantity added by the water feeder can be monitored through the flowmeter, and the starting and stopping of water feeding and the water flow during water feeding can be controlled through adjusting the opening degree of the electromagnetic valve.
According to the method for controlling water adding of the materials when the materials flow through the production equipment, provided by the embodiment of the application, the prediction model is used for predicting the water content of the outlet materials instead of directly predicting the water adding amount, so that the influence factors between the input variable and the output variable of the prediction model are reduced, and the accuracy of the prediction model is improved; and the water content of the predicted outlet material is determined together based on the predicted value and the corrected value, the water content of the predicted outlet material can be adjusted timely through adjusting the corrected value, the situation that the influence of interference factors cannot be processed timely when the prediction is performed through a fixed model is avoided, and the control stability is improved; in addition, the influence of factors such as inlet material flow and the like on the water content of the outlet material is considered by the input variables of the prediction model, and the prediction model is more reasonable and accurate.
In some embodiments, correcting the historical predicted value output by the predictive model based on a plurality of historical data of the material flowing through the production facility to obtain a corrected value for the predictive model includes: obtaining historical data of the latest M sampling periods; taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount of historical data of each sampling period in M sampling periods as input variables of a prediction model to respectively obtain a first prediction value output by the prediction model; obtaining a first predicted value and a difference value of the water content of a historical outlet material in the corresponding historical data; based on a plurality of differences corresponding to the M sampling periods, a first average value of the plurality of differences is obtained, and based on the first average value, a correction value of the prediction model is obtained.
In this embodiment, the historical data of the last M sampling periods may be the historical data of the last M sampling periods, which is acquired in sequence with the time origin at this time when the water content of the material outlet is predicted by using the prediction model in the production process, and is closest to the time origin in time. The specific content of the historical data of the M sampling periods depends on the time point of the prediction of the outlet water content by using a prediction model, so that the selected historical data has stronger correlation with the time point of the prediction of the outlet water content. In the embodiment, when the prediction model is used for prediction, the correction value updated in real time is obtained through the historical data of the latest M sampling periods, so that the timeliness of the correction value is improved, the influence of the sudden interference factors in the production process on the water content of the predicted material outlet is reduced, and the accuracy of the prediction result is improved.
In some embodiments, correcting the historical predicted value output by the prediction model based on a plurality of historical data of the material flowing through the production facility to obtain a corrected value for the prediction model, further comprising: obtaining historical data of N recent sampling periods, wherein N is less than M; taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount of historical data of each sampling period in N sampling periods as input variables of a prediction model to respectively obtain second predicted values output by the prediction model; obtaining a second predicted value and a difference value of the water content of the historical outlet material in the corresponding historical data; obtaining a second average value of the plurality of difference values based on the plurality of difference values respectively corresponding to the N sampling periods, wherein the first average value has a first weight, and the second average value has a second weight; and obtaining a correction value for the prediction model based on the first average value and the second average value and the first weight and the second weight.
The historical data of the last N sampling periods may be the historical data of the last N sampling periods, which is acquired in sequence with the time origin at the time when the water content of the material outlet is predicted by using the prediction model in the production process, and is closest to the time origin in time. In this embodiment, N < M, the history data of the last N sampling periods is closer to the time point when the prediction of the outlet water content is performed than the history data of the last M sampling periods. It will be appreciated that the first average value obtained based on the historical data of the last M sampling periods reflects a deviation over a longer period of time; the second average value obtained based on the history of the last N sampling periods reflects the deviation over a shorter period of time. Different weights are given to a first average value obtained based on M sampling periods and a second average value obtained based on N sampling periods, and correction values are obtained by multiplying the first average value and the second average value by the weights and then adding the multiplied values. In this embodiment, the correction value is updated by the long and short sampling periods, so that the long-term trend of the correction value can be captured, and the change of the short-term trend of the correction value can be responded in time, so that the obtained correction value has higher reliability, and the accuracy of the prediction result is improved. Illustratively, M may be 30, 40, 50, 60, 70 or other ones of these values; n may be 2,3,4 or other values less than or equal to 10.
In some embodiments, the first weight is greater than the second weight. By setting the first weight to be larger than the second weight, the influence of deviation in a long sampling period on the correction value is larger, and the correction value is more stable, so that an abnormal value of short-term deviation change can be filtered, the correction value is prevented from being suddenly changed under the influence of the abnormal value, and the accuracy of a prediction result is prevented from being influenced. Illustratively, the first weight may range from 0.7 to 0.9 and the second weight may range from 0.1 to 0.3. Preferably, the first weight is 0.8 and the second weight is 0.2.
In some embodiments, obtaining the desired adjusted water amount for the material based on the predicted outlet material water content and the preset value of the outlet material water content comprises: obtaining the water absorption coefficient of the material; and obtaining the expected regulating water quantity of the material based on the difference value between the predicted water content of the outlet material and the preset value of the water content of the outlet material, the current inlet material flow and the water absorption coefficient.
The water absorption coefficient of the material may reflect the amount of water absorbed when the unit water content of the material per unit mass changes, and in this embodiment, in order to reduce the difference between the predicted water content of the outlet material and the preset value of the water content of the outlet material, it is necessary to increase or decrease the corresponding amount of water according to the difference, and the corresponding amount is desired to adjust the water amount. For example, the desired amount of water to be regulated may be the product of the water absorption coefficient, the difference, and the current inlet material flow corresponding to its time delay.
In some embodiments, obtaining the water absorption coefficient of the material comprises: calculating a difference value between the water content of the historical inlet material and the water content of the corresponding historical outlet material in the historical data; establishing a linear relation between a difference value between the water content of the historical inlet material and the water content of the corresponding historical outlet material, and the flow rate of the historical inlet material and the historical water adding amount; based on the linear relation, the water absorption coefficient of the material is obtained. In this embodiment, the water absorption coefficient of the material is calculated by actually measuring the difference value between the water content of the historical inlet material and the water content of the corresponding historical outlet material in the historical data, the flow rate of the historical inlet material and the historical water adding amount, so that the historical data is fully utilized, the water absorption coefficient of the material is not required to be measured, and the accuracy of the water absorption coefficient can be improved. Illustratively, a plurality of historical data may be linearly fitted to obtain the water absorption coefficient of the material. Of course, in other embodiments, the water absorption coefficient of the material may also be obtained by measurement or empirical values.
In some embodiments, the production facility charges water to the material through a first water charger and a second water charger, the first water charger and the second water charger being disposed at different locations of the production facility along a flow direction of the material, wherein the historical water charge comprises a historical first water charger charge and a historical second water charger charge, the historical first water charger charge and the historical second water charger charge being two input variables of the original model and the predictive model; the current water adding quantity comprises the current water adding quantity of the first water adding device and the current water adding quantity of the second water adding device, and the current water adding quantity of the first water adding device and the current water adding quantity of the second water adding device are used as two input variables of a prediction model.
In this embodiment, a plurality of water feeders are set, and the plurality of water feeders are distributed along the flow direction of the material, so that water feeding to the material is more uniform, and when an original model is trained, a prediction model is corrected and the prediction model is used for prediction under the condition that a plurality of water feeders exist, the water amounts of the plurality of water feeders are not combined into one input variable, but the water amounts of the water feeders are taken as independent input variables, so that the mode can be closer to the actual working condition, and the prediction model and the predicted value obtained by the prediction model are more accurate. In the redrying and moistening process, the first water feeder may be a water feeder arranged in the front chamber, and the second water feeder may be a water feeder arranged in the rear chamber.
In some embodiments, the first water applicator has a nominal water addition, wherein controlling the water addition to the material based on the desired adjusted water amount comprises: controlling the first water feeder to feed water to the material according to the current water feeding amount of the first water feeder according to the water feeding amount adjusted by the expected water feeding amount; and controlling the first water feeder to feed water to the material in the rated water feeding amount in response to the fact that the current water feeding amount of the first water feeder is larger than the rated water feeding amount according to the expected water feeding amount, and controlling the second water feeder to feed water to the material by taking the difference between the current water feeding amount of the first water feeder and the rated water feeding amount according to the expected water feeding amount.
For example, the rated water addition amount of the first water adder may be 200. At a certain moment, the current water adding amount of the first water adding device is 195, the current water adding amount of the second water adding device is 100, the calculated expected adjusting water amount is +3, at the moment, the current water adding amount of the first water adding device is 198 according to the expected adjusting water amount, the adjusted water adding amount is smaller than the rated water adding amount of the first water adding device, at the moment, the first water adding device is controlled to add water according to the 198 water adding amount, and the second water adding device is controlled to add water according to the 100 water adding amount; at another moment, the current water adding quantity of the first water adding device is 195, the current water adding quantity of the second water adding device is 100, the calculated expected adjusting water quantity is +8, at the moment, the current water adding quantity of the first water adding device is 203 according to the expected adjusting water quantity, the water adding quantity is larger than the rated water adding quantity of the first water adding device, the difference value between the current water adding quantity of the first water adding device and the rated water adding quantity is +3 according to the expected adjusting water quantity, the second water adding device is +3 on the basis of the current water adding quantity of the second water adding device, at the moment, the first water adding device is controlled to add water with the water adding quantity of 200, and the second water adding device is controlled to add water with the water adding quantity of 103.
In this embodiment, by limiting the water adding amount of the first water adder and adjusting the control strategy for compensating the water adding amount of the second water adder, predictive control can be facilitated, and control logic is simplified, so that the water adder can respond to the adjustment of the water adding amount more quickly.
According to another aspect of the present application, there is provided an apparatus for controlling the addition of water to a material as it flows through a production facility, the apparatus comprising a memory and a processor, the memory for storing instructions; the processor is coupled to the memory and is configured to execute the method provided by any of the embodiments of the present application based on the instructions stored in the memory.
Finally, it is noted that the above-mentioned preferred embodiments illustrate rather than limit the application, and that, although the application has been described in detail with reference to the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the application as defined by the appended claims; the dimensions in the drawings and the embodiments are irrelevant to specific objects, are not used for limiting the protection scope of the application, and the objects can be selected and changed according to actual needs.

Claims (11)

1. A method for controlling the addition of water to a material as it flows through a production facility, comprising:
obtaining a prediction model;
Correcting a historical predicted value output by the prediction model based on a plurality of historical data when the material flows through production equipment to obtain a corrected value for the prediction model, wherein the historical data comprise a historical inlet material flow rate when the material enters the production equipment, a historical inlet material water content when the material enters the production equipment, a historical outlet material water content when the material flows out of the production equipment and a historical water adding amount of the material in the production equipment;
When the material flows through the production equipment, the current inlet material flow rate when the material enters the production equipment, the current inlet material water content when the material enters the production equipment and the current water adding amount of the material when the material is in the production equipment are used as input variables of the prediction model, so that a current prediction value output by the prediction model is obtained;
Based on the current predicted value and the corrected value, obtaining the predicted water content of the outlet material;
obtaining expected adjustment water quantity of the materials based on the predicted outlet material water content and the preset value of the outlet material water content;
Controlling the water addition of the material based on the desired adjusted water amount;
the prediction model is established through the following steps:
Obtaining a plurality of historical data of the material flowing through the production equipment;
Obtaining an initial model, taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount as input variables of the initial model, taking the historical outlet material water content as output variables of the initial model, training the initial model, and establishing the prediction model.
2. The method of claim 1, wherein modifying the historical predictions output by the predictive model based on a plurality of historical data of the material flowing through the production facility to obtain modified values for the predictive model comprises:
obtaining historical data of the latest M sampling periods;
Taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount of historical data of each sampling period in M sampling periods as input variables of the prediction model to respectively obtain a first predicted value output by the prediction model;
Obtaining a difference value of the water content of the historical outlet materials in the first predicted value and the corresponding historical data;
based on a plurality of differences corresponding to the M sampling periods respectively, a first average value of the plurality of differences is obtained,
And obtaining a correction value for the prediction model based on the first average value.
3. The method of claim 2, wherein correcting the historical predictions output by the predictive model based on a plurality of historical data of the material flowing through the production facility to obtain corrected values for the predictive model further comprises:
obtaining historical data of N recent sampling periods, wherein N is less than M;
Taking the historical inlet material flow, the historical inlet material water content and the historical water adding amount of historical data of each sampling period in N sampling periods as input variables of the prediction model to respectively obtain second predicted values output by the prediction model;
obtaining a difference value of the water content of the historical outlet materials in the second predicted value and the corresponding historical data;
Obtaining a second average value of the plurality of difference values based on the plurality of difference values respectively corresponding to the N sampling periods, wherein the first average value has a first weight, and the second average value has a second weight;
and obtaining a correction value of the prediction model based on the first average value and the second average value and the first weight and the second weight.
4. A method according to claim 3, wherein the first weight is greater than the second weight.
5. The method of claim 1, wherein obtaining a desired adjusted water amount for the material based on the predicted outlet material water content and the preset value of outlet material water content comprises:
obtaining the water absorption coefficient of the material;
and obtaining the expected adjustment water quantity of the material based on the difference value between the predicted water content of the outlet material and the preset value of the water content of the outlet material, the current inlet material flow and the water absorption coefficient.
6. The method of claim 5, wherein obtaining the water absorption coefficient of the material comprises:
calculating a difference value between the water content of the historical inlet material and the water content of the corresponding historical outlet material in the historical data;
Establishing a linear relation between a difference value between the water content of the historical inlet material and the water content of the corresponding historical outlet material, and the flow rate of the historical inlet material and the historical water adding amount;
and obtaining the water absorption coefficient of the material based on the linear relation.
7. The method of claim 1, wherein the production facility charges the material with a first water charger and a second water charger, the first water charger and the second water charger being disposed at different locations of the production facility along a flow direction of the material, wherein,
The historical water adding quantity comprises a historical first water adding device water adding quantity and a historical second water adding device water adding quantity, and the historical first water adding device water adding quantity and the historical second water adding device water adding quantity are used as two input variables of the original model and the prediction model;
The current water adding amount comprises the current water adding amount of the first water adding device and the current water adding amount of the second water adding device, and the current water adding amount of the first water adding device and the current water adding amount of the second water adding device are used as two input variables of the prediction model.
8. The method of claim 7, wherein the first water applicator has a nominal water addition, wherein controlling water addition to the material based on the desired adjusted water amount comprises:
Controlling the first water feeder to feed water to the material according to the current water feeding amount of the first water feeder according to the water feeding amount adjusted by the expected adjusted water feeding amount in response to the current water feeding amount of the first water feeder being smaller than or equal to the rated water feeding amount;
And controlling the first water feeder to feed water to the material in the rated water feeding amount in response to the fact that the current water feeding amount of the first water feeder is larger than the rated water feeding amount according to the expected water feeding amount, and controlling the second water feeder to feed water to the material by taking the difference between the current water feeding amount of the first water feeder and the rated water feeding amount according to the expected water feeding amount.
9. The method according to claim 1, characterized in that each variable in the history data is kalman filtered and/or the time delays of each variable are aligned before training the initial model using the history data.
10. The method of any one of claims 1-9, wherein the initial model and the predictive model are long-term memory neural network models.
11. An apparatus for controlling the addition of water to a material as it flows through a production facility, the apparatus comprising:
a memory for storing instructions;
A processor coupled to the memory, the processor configured to perform the method of any of claims 1-9 based on instructions stored by the memory.
CN202410044062.6A 2024-01-11 2024-01-11 Method and device for controlling water addition to materials during material flow through production equipment Pending CN118034195A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410044062.6A CN118034195A (en) 2024-01-11 2024-01-11 Method and device for controlling water addition to materials during material flow through production equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410044062.6A CN118034195A (en) 2024-01-11 2024-01-11 Method and device for controlling water addition to materials during material flow through production equipment

Publications (1)

Publication Number Publication Date
CN118034195A true CN118034195A (en) 2024-05-14

Family

ID=91003355

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410044062.6A Pending CN118034195A (en) 2024-01-11 2024-01-11 Method and device for controlling water addition to materials during material flow through production equipment

Country Status (1)

Country Link
CN (1) CN118034195A (en)

Similar Documents

Publication Publication Date Title
JP3949164B2 (en) A feedback method for controlling nonlinear processes.
EP0710901B1 (en) Multivariable nonlinear process controller
CN110101106B (en) Moisture control method and system for dampening and humidifying process based on fuzzy feedforward feedback algorithm
US20120221156A1 (en) Method and system for controlling a multivariable system with limits
US5566065A (en) Method and apparatus for controlling multivariable nonlinear processes
US7653445B2 (en) Apparatus and method for model-based control
CN101479682B (en) Flow controller delivery of a specified quantity of a fluid
JP2002523821A (en) Controller for determining optimal tuning parameters used in a process control system and method of operating the controller
CN112914139B (en) Method and system for controlling water adding amount in loosening and moisture regaining process
CN110955143B (en) Compound control method for first-order inertia pure lag process
CN102987538B (en) Air-flowing type dried tobacco water control method
CN101995822A (en) Grey active disturbance rejection control method of long time-delay system
CN113491341B (en) Method for controlling tobacco moisture regain and water adding flow based on historical production data modeling
CN115251445B (en) Control method for moisture content of tobacco leaves at outlet of loosening and conditioning machine
CN114166328B (en) Method for self-adjusting weighing sensor
CN112263012B (en) Moisture content control method of redrying machine based on formula parameter library
CN108536087B (en) Control method and control device for moisture regain and humidification of tobacco leaves and tobacco sheets
CN111694275B (en) Equipment control method and device and electronic equipment
CN118034195A (en) Method and device for controlling water addition to materials during material flow through production equipment
CN109964180B (en) Device and method for determining parameters of a control device
CN111650829A (en) Embedded PID module parameter adjusting method, system and device based on artificial intelligence
AU610166B2 (en) Multivariable adaptive feedforward controller
CN112263015B (en) Method for controlling discharge flow of cabinet type feeding machine
JPH01308320A (en) Quarry amount control device
JPS5936803A (en) Method for correcting feedforward model

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination