Thin plate cut tobacco dryer cylinder wall temperature regulation and control method, electronic equipment and program product

Abstract

The invention provides a thin-plate cut-tobacco dryer cylinder wall temperature regulation and control method, electronic equipment and a program product, and the method comprises the steps: collecting an environment temperature, a target moisture content, a tobacco leaf instantaneous flow rate and a cut tobacco temperature, and inputting into a trained prediction model to obtain a cylinder wall prediction temperature; adjusting the temperature of the cylinder wall to the predicted temperature of the cylinder wall; obtaining the actual outlet moisture content of the tobacco shreds; obtaining an initial PID control parameter based on a first difference value between the actual outlet water content and the target outlet water content and a change rate of the first difference value along with time; the initial PID control parameters are input into an optimization model, and the optimization model optimizes the initial PID control parameters based on a genetic algorithm; and adjusting the opening degree of a steam valve based on the optimized PID control parameters so as to adjust the temperature of the cylinder wall. In the technical scheme provided by the invention, the problem that the cut tobacco flow and the environment temperature suddenly change, and consequently the cut tobacco quality is reduced is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of tobacco leaf processing, and more particularly, to a method for regulating the temperature of the cylinder wall of a thin plate tobacco dryer, an electronic device, and a program product. Background Art

[0002] In the prior art, based on the moisture content of the tobacco leaves at the outlet, the temperature of the cylinder wall of the tobacco dryer is regulated. Specifically, the actual moisture content of the tobacco leaves at the outlet is collected in real time, and the PID controller regulates the temperature of the cylinder wall of the tobacco dryer based on the difference between the actual moisture content of the tobacco leaves and the target moisture content of the tobacco leaves at the outlet and the change rate of the difference. This method relying solely on PID regulation has drawbacks, mainly including: due to the lag of PID regulation, when there are large fluctuations in the ambient temperature and the tobacco leaf flow rate, the increase in the tobacco leaf flow rate will increase the heat load of the cylinder wall, resulting in a shortening of the temperature response delay time and an increase in the system gain. At this time, a fixed Kp cannot match the new heat conduction rate, which may cause overshoot or oscillation. Moreover, an increase in the ambient temperature will reduce the heat exchange efficiency, and humidity changes affect the moisture evaporation rate, resulting in a non-linear shift of the system transfer function. The linear control of traditional PID cannot compensate for this shift, and the integral term Ki may cause integral saturation due to error accumulation. Therefore, relying solely on traditional PID control will lead to a decline in the quality of tobacco leaves. Summary of the Invention

[0003] In view of this, the purpose of the embodiments of the present application is to provide a method for regulating the temperature of the cylinder wall of a thin plate tobacco dryer, which can improve the problem that relying solely on the traditional PID control method to regulate the temperature of the cylinder wall will lead to a decline in the quality of tobacco leaves.

[0004] To achieve the above technical objectives, the technical solutions adopted in the present application are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for regulating the temperature of the cylinder wall of a thin plate tobacco dryer, the method comprising:

[0006] Collect the ambient temperature, target moisture content, instantaneous tobacco leaf flow rate, and tobacco leaf temperature;

[0007] Input the ambient temperature, target outlet moisture content, instantaneous tobacco leaf flow rate, and tobacco leaf temperature into a trained prediction model to obtain a predicted cylinder wall temperature;

[0008] Adjust the temperature of the cylinder wall of the thin plate tobacco dryer to the predicted cylinder wall temperature;

[0009] Obtain the actual outlet moisture content of the tobacco leaves;

[0010] Based on the first difference between the actual outlet moisture content and the target outlet moisture content, and the change rate of the first difference over time, obtain initial PID control parameters;

[0011] Input the initial PID control parameters into an optimization model. The optimization model is based on a genetic algorithm to optimize the initial PID control parameters and obtain optimized PID control parameters.

[0012] Based on the optimized PID control parameters, adjust the opening degree of the steam valve to adjust the temperature of the cylinder wall so that the first difference is less than or equal to the first difference threshold.

[0013] Further, after adjusting the opening degree of the steam valve based on the optimized PID control parameters, the method further includes:

[0014] Obtain the actual cylinder wall temperatures collected by all temperature sensors, where all the temperature sensors are evenly distributed on the temperature measurement surface of the cylinder wall of the cut tobacco dryer.

[0015] Based on a clustering algorithm, cluster all the actual cylinder wall temperatures into several clusters.

[0016] Based on the actual cylinder wall temperatures corresponding to the cluster centers of each cluster, obtain the median value of the actual cylinder wall temperatures.

[0017] Calculate the second difference between each actual cylinder wall temperature and the median value of the actual cylinder wall temperatures, and select the cluster centers with the second difference greater than the difference threshold.

[0018] Send the positions corresponding to the cluster centers with the second difference greater than the difference threshold to a display terminal.

[0019] Further, the step of inputting the ambient temperature, target outlet moisture content, instantaneous tobacco leaf flow rate, and cut tobacco temperature into the trained prediction model to obtain the predicted cylinder wall temperature includes:

[0020] Preset a deep learning model to obtain a pre-training prediction model.

[0021] Collect several groups of historical data, where the historical data includes historical ambient temperature, historical outlet moisture content, historical instantaneous tobacco leaf flow rate, and historical cut tobacco temperature corresponding in time series.

[0022] Divide all the historical data into a training set and a test set, and input them into the pre-training prediction model to obtain the trained prediction model, so that inputting the ambient temperature, target outlet moisture content, instantaneous tobacco leaf flow rate, and cut tobacco temperature into the trained prediction model can obtain the predicted cylinder wall temperature.

[0023] Further, the step of inputting the initial PID control parameters into an optimization model. The optimization model is based on a genetic algorithm to optimize the initial PID control parameters and obtain optimized PID control parameters includes:

[0024] Based on the preset range of PID control parameters, a number of real - number discrete points of PID control parameters are obtained, and one of the real - number discrete points of PID control parameters is the initial PID control parameter;

[0025] Based on the real - number discrete points of PID control parameters and the fitness function, the fitness corresponding to the real - number discrete points of PID control parameters is obtained, and the fitness is used to characterize the first difference;

[0026] Take the real - number discrete points of PID control parameters whose fitness is less than or equal to the fitness threshold as the parental discrete points, and perform iteration until the optimized PID control parameters are obtained;

[0027] The iteration includes:

[0028] Based on the parental discrete points, perform crossover and / or mutation to obtain offspring discrete points;

[0029] Input the offspring discrete points into the fitness function;

[0030] Select the offspring discrete points whose fitness is less than or equal to the fitness threshold as the parental discrete points;

[0031] Repeat the steps of performing crossover and / or mutation based on the parental discrete points to obtain offspring discrete points; input the offspring discrete points into the fitness function; select the offspring discrete points whose fitness is less than or equal to the fitness threshold as the parental discrete points.

[0032] Furthermore, the fitness function is:

[0033] Fitness=|Q actual -Q target |

[0034] Where Fitness represents the fitness;

[0035] Q actual represents the simulated outlet moisture content;

[0036] Q target represents the target moisture content;

[0037] The step of obtaining the fitness corresponding to the real - number discrete points of PID control parameters based on the real - number discrete points of PID control parameters and the fitness function includes:

[0038] Input the real - number discrete points of PID control parameters into the simulation system, and the simulation system outputs Q actual ;

[0039] Based on the fitness function, calculate the fitness.

[0040] Further, adjusting the opening degree of the steam valve based on the optimized PID control parameters includes:

[0041] Inputting the optimized PID control parameters, the rate of change, and the first difference into an opening degree algorithm to obtain the opening degree of the steam valve, where the opening degree algorithm is:

[0042]

[0043] where k p represents the optimized proportional coefficient;

[0044] k i represents the optimized integral coefficient;

[0045] k d represents the optimized differential coefficient;

[0046] e(t) represents the first difference.

[0047] In a second aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory coupled to each other. The memory stores a computer program. When the computer program is executed by the processor, the electronic device executes the above method.

[0048] In a third aspect, an embodiment of the present application provides a computer program product. A computer program is stored in a computer-readable storage medium. When the computer program runs on a computer, the computer executes the above method.

[0049] The invention adopting the above technical solution has the following advantages:

[0050] In the technical solution provided by the present application, a trained prediction model is used to dynamically predict the required temperature of the cylinder wall in combination with variables such as environmental temperature and tobacco leaf flow rate, so as to timely change the temperature of the cylinder wall and avoid the problem of deterioration of tobacco leaf quality caused by sudden changes in tobacco leaf flow rate and environmental temperature.

[0051] In the technical solution provided by the present application, the initial PID parameters are optimized based on a genetic algorithm to solve the problem of mismatch of fixed parameters when the tobacco leaf flow rate suddenly changes or the environmental temperature and humidity change.

[0052] In the technical solution of the present application, the initial PID control parameters are optimized through an optimization model (based on a genetic algorithm), and a more reasonable control strategy for the opening degree of the steam valve can be obtained. This can not only reduce the waste of steam, lower energy consumption, but also improve the regulation accuracy and stability of the cylinder wall temperature. Description of the Drawings

[0053] This application can be further illustrated by non-limiting embodiments given in the accompanying drawings. It should be understood that the following drawings only show some embodiments of this application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0054] Figure 1 It is a flowchart of the method provided by the embodiment of this application.

[0055] Figure 2 It is a sub-flowchart of S120 provided by the embodiment of this application.

[0056] Figure 3 It is a sub-flowchart of S160 provided by the embodiment of this application.

[0057] Figure 4 It is a sub-flowchart of S180 provided by the embodiment of this application. Detailed implementation manners

[0058] The following will describe this application in detail in combination with the accompanying drawings and specific embodiments. It should be noted that in the drawings or the description of the specification, similar or identical parts use the same figure numbers. The implementation manners not shown or described in the drawings are in the forms known to those of ordinary skill in the relevant art. In the description of this application, the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0059] An electronic device provided by an embodiment of this application may include a processing module and a storage module. A computer program is stored in the storage module. When the computer program is executed by the processing module, the electronic device can execute the corresponding steps in the following method for regulating the temperature of the cylinder wall of a thin plate tobacco dryer.

[0060] Please refer to Figure 1 , this application also provides a method for regulating the temperature of the cylinder wall of a thin plate tobacco dryer. Among them, the method for regulating the temperature of the cylinder wall of a thin plate tobacco dryer may include the following steps:

[0061] S110: Collect the ambient temperature, target moisture content, instantaneous tobacco leaf flow rate, and cut tobacco temperature;

[0062] S120: Input the ambient temperature, target outlet moisture content, instantaneous tobacco leaf flow rate, and cut tobacco temperature into the trained prediction model to obtain the predicted cylinder wall temperature;

[0063] S130: Adjust the temperature of the cylinder wall of the thin plate tobacco dryer to the predicted cylinder wall temperature;

[0064] S140: Obtain the actual outlet moisture content of the cut tobacco;

[0065] S150: Obtain initial PID control parameters based on the first difference between the actual outlet moisture content and the target outlet moisture content, and the rate of change of the first difference over time;

[0066] S160: Input the initial PID control parameters into an optimization model. The optimization model is based on a genetic algorithm to optimize the initial PID control parameters and obtain optimized PID control parameters;

[0067] S170: Adjust the opening of the steam valve based on the optimized PID control parameters to adjust the cylinder wall temperature, so as to make the first difference less than or equal to the first difference threshold.

[0068] The following will elaborate on each step of the method for regulating the cylinder wall temperature of the thin plate tobacco dryer in detail as follows:

[0069] In S110, it is necessary to collect the ambient temperature, target moisture content, instantaneous tobacco leaf flow rate, and cut tobacco temperature. Among them, the target moisture content in this embodiment can be determined during the product design stage and then input in advance by relevant engineers.

[0070] The ambient temperature can usually be collected by installing a temperature sensor. Common temperature sensors include thermocouples, thermistors (such as PT100 or NTC), infrared sensors, etc. These sensors can be fixed at appropriate positions near the tobacco dryer to monitor and collect ambient temperature data in real time. When selecting a sensor, factors such as the measurement range, accuracy, stability, and working environment need to be considered to ensure the accuracy and reliability of the collected data.

[0071] The collection of the instantaneous cut tobacco flow rate can be achieved by installing a flow sensor on the cut tobacco conveying path. The flow sensor can monitor the flow rate of cut tobacco in real time and convert it into an electrical signal for collection. Common flow sensors include, but are not limited to, electronic belt scales. When selecting a flow sensor, factors such as the physical properties of cut tobacco (such as density, viscosity, etc.), flow range, and measurement accuracy need to be considered.

[0072] The collection of the cut tobacco temperature can be achieved by installing a temperature sensor on the tobacco leaf conveying path. These temperature sensors can be fixed at positions where the tobacco leaf is in contact or non-contact to monitor and collect the temperature data of the tobacco leaf in real time. Similar to collecting the ambient temperature, when selecting a temperature sensor, factors such as the measurement range, accuracy, stability, and working environment also need to be considered.

[0073] In S120, the prediction model can be a Deep Gated Recurrent Unit (DGRU) network model or a Convolutional Neural Network (CNN) combined with a Recurrent Neural Network (RNN). The reasons are as follows: The DGRU network is particularly suitable for processing time series data and can capture the dynamic features in the data. When dealing with problems such as the prediction of the cylinder wall temperature of a cut tobacco dryer, the DGRU network can extract the deep non-linear dynamic features in industrial data by stacking gated recurrent units, thereby improving the prediction accuracy. When the DGRU algorithm predicts the cylinder wall temperature, the prediction error is small, and accurate dynamic prediction of the cylinder wall temperature can be achieved.

[0074] The CNN performs well in processing images and local feature extraction, while the RNN is good at processing time series data. By combining the CNN and the RNN, the relevant features of tobacco leaves and cut tobacco (such as texture, color, etc.) can be extracted first using the CNN, and then the RNN can be used to capture the changing patterns of these features over time, so as to more accurately predict the cylinder wall temperature. This combined model can make full use of the advantages of the two networks and improve the accuracy and robustness of the prediction.

[0075] As Figure 2 shown, S120 includes the following steps:

[0076] S121: Preset a deep learning model to obtain a pre-training prediction model;

[0077] S122: Collect several groups of historical data, where the historical data includes historical ambient temperature, historical outlet moisture content, historical instantaneous tobacco leaf flow rate, and historical cut tobacco temperature corresponding in time series;

[0078] S123: Divide all the historical data into a training set and a test set, and input them into the pre-training prediction model to obtain the post-training prediction model. By inputting the ambient temperature, target outlet moisture content, instantaneous tobacco leaf flow rate, and cut tobacco temperature into the post-training prediction model, the predicted cylinder wall temperature can be obtained.

[0079] The historical data can be obtained in the following ways: Query the enterprise database from the historical database: If the enterprise has established a complete production database, the required historical data can be directly queried from the historical database. These data usually include various parameters and indicators in the production process, such as temperature, flow rate, moisture content, etc. Industry database: In addition, the industry database or public data sets can also be queried to understand the historical data under similar production conditions. These data can provide valuable references for model training.

[0080] In this embodiment, after obtaining the trained prediction model, the collected ambient temperature, target outlet moisture content, instantaneous tobacco leaf flow rate, and cut tobacco temperature are used as input values, and the trained prediction model outputs the predicted cylinder wall temperature. Therefore, through S120, when parameters such as the ambient temperature and instantaneous tobacco leaf flow rate suddenly change, the predicted cylinder wall temperature can be accurately obtained, enabling the cylinder wall temperature to be corrected in a timely manner and avoiding too large a difference in moisture content due to the cylinder wall temperature not having time to change.

[0081] In S140, the actual outlet moisture content of the cut tobacco can be measured using a professional tobacco moisture detector. Such an instrument usually features high precision and rapid measurement, and can accurately reflect the moisture content of the cut tobacco.

[0082] In S150, the PID control principle is to adjust based on the difference between the current state of the controlled object and the set value to output an appropriate control signal, enabling the state of the controlled object to gradually approach and stabilize near the set value. PID control consists of three basic parts: proportional (P), integral (I), and derivative (D), each with different functions and roles: Proportional (P) control: Adjusts the output of the controller according to the magnitude of the current error (i.e., the difference between the set value and the actual value). When the error is large, the output signal is also large, thereby accelerating the response speed of the controlled object. Proportional control can make the system quickly approach the set value but may cause overshoot and oscillation. Integral (I) control: Adjusts the output of the controller according to the accumulated amount of the error over time. The role of integral control is to eliminate the steady-state error of the system and ensure that the system finally stabilizes near the set value. It can continuously adjust the output to accumulate and eliminate the error, but may also cause overshoot and oscillation. Derivative (D) control: Adjusts the output of the controller according to the rate of change of the error. Derivative control can predict the future state change trend of the system, thereby reducing overshoot and improving the stability of the system. The introduction of derivative control can suppress the oscillation of the system but will also increase the sensitivity of the system to noise. In PID control, these three parts calculate the output signal through different weight combinations to achieve precise control of the controlled object. The specific implementation process is as follows: Error calculation: The system first detects the state of the controlled object in real time through a sensor and compares it with the set value to obtain the error. PID calculation: The PID controller converts the error into an output signal according to the weights of the proportional, integral, and derivative parts. This output signal is a linear combination obtained by performing proportional, integral, and derivative operations on the error.

[0083] Therefore, in this step, based on the first difference and the rate of change, according to the above PID control principle, the initial PID control parameters can be obtained, where the initial PID control parameters include the initial integral coefficient, the initial derivative coefficient, and the initial proportional coefficient.

[0084] In S160, asFigure 3 As shown in the figure, it may specifically include the following steps:

[0085] S161: Based on the preset range of the PID control parameters, obtain a number of real discrete points of the PID control parameters, where one of the real discrete points of the PID control parameters is the initial PID control parameter;

[0086] S162: Based on the real discrete points of the PID control parameters and the fitness function, obtain the fitness corresponding to the real discrete points of the PID control parameters;

[0087] S163: Take the real discrete points of the PID control parameters whose fitness is less than or equal to the fitness threshold as the parental discrete points, and perform iteration until the optimized PID control parameters are obtained.

[0088] The iteration includes:

[0089] Based on the parental discrete points, perform crossover and / or mutation to obtain offspring discrete points;

[0090] Input the offspring discrete points into the fitness function;

[0091] Select the offspring discrete points whose fitness is less than or equal to the fitness threshold as the parental discrete points;

[0092] Repeat the steps of performing crossover and / or mutation based on the parental discrete points to obtain offspring discrete points; input the offspring discrete points into the fitness function; select the offspring discrete points whose fitness is less than or equal to the fitness threshold as the parental discrete points.

[0093] The process of the genetic algorithm mentioned in S161 includes:

[0094] Initialization: Set the evolution generation counter t = 0, set the maximum number of generations T, and randomly generate M individuals as the initial population P(0). Selection operation: Apply the selection operator to the population. The purpose of selection is to directly inherit the optimized individuals to the next generation or generate new individuals through pairing and crossover and then inherit them to the next generation. The selection operation is based on the fitness evaluation of the individuals in the population. Crossover operation: Apply the crossover operator to the population. Crossover refers to the operation of replacing and recombining part of the structures of two parent individuals to generate new individuals. It plays a core role in the process of biological evolution in nature and also plays a core role in genetic algorithms. Mutation operation: Apply the mutation operator to the population. That is, make changes to the gene values at certain gene loci of the individual strings in the population. The basic contents of the mutation operator include real-value mutation and binary mutation, etc. Termination condition judgment: If t = T, then output the individual with the maximum fitness obtained during the evolution process as the optimal solution and terminate the calculation. The termination conditions also include that the fitness of the optimal individual reaches a given threshold, or when the fitness of the optimal individual and the population fitness no longer increase.

[0095] Therefore, in step 161, the preset range of the PID control parameters can be the ranges of the proportional coefficient, the differential coefficient, and the integral coefficient. The range cannot be too large. For example, Kp ∈ [0, 15], Ki ∈ [0, 5], Kd ∈ [0, 10]. Then randomly extract discrete points within these ranges, integrate these discrete points to form an initialized population. The initialized population includes several real-number discrete points of the PID control parameters, and the initial PID control parameters are also one of the real-number discrete points of the PID control parameters, which can contribute to the rapid convergence of the genetic algorithm. For example, the real-number discrete points {Kp, Ki, Kd} of the PID control parameters can be {8, 1, 2}, {3.5, 2.3, 2}, etc. These values are all extracted from the preset range.

[0096] In S162, the fitness function of this embodiment is:

[0097] Fitness = |Q actual -Q target |

[0098] Among them, Fitness represents the fitness;

[0099] Q actual represents the simulated outlet moisture content, which is obtained through the simulation system. The simulation system can be a finite element simulation system.

[0100] Q target represents the target moisture content, which is obtained from S110.

[0101] In this embodiment, all real discrete points of the PID control parameters are input into the fitness function, and all fitness values are calculated. The way to calculate the fitness is as follows: Input the real discrete points of the PID control parameters into the finite element simulation system, and output Q actual , and then use Q actual to calculate the corresponding fitness.

[0102] In S163, the real discrete points of the PID control parameters corresponding to the fitness less than the fitness threshold are used as the parent discrete points. The fitness threshold can represent the maximum allowable value of the first difference, and the fitness threshold can be designed according to the required quality during the product design stage of tobacco leaves.

[0103] In S163, the iterative steps include:

[0104] Based on the parent discrete points, perform crossover and / or mutation to obtain offspring discrete points;

[0105] Input the offspring discrete points into the fitness function;

[0106] Screen the offspring discrete points with the fitness less than or equal to the fitness threshold as the parent discrete points;

[0107] Repeat the steps of performing crossover and / or mutation based on the parent discrete points to obtain offspring discrete points; input the offspring discrete points into the fitness function; screen the offspring discrete points with the fitness less than or equal to the fitness threshold as the parent discrete points.

[0108] Among them, 1. Example of crossover operation Suppose we have two parent individuals, and their PID parameters (Kp, Ki, Kd) are encoded as genotypes as follows: Parent individual A: Kp = 2.0, Ki = 1.0, Kd = 0.5 (encoded as genotype: 1010001101. Here, for simplicity, we use binary encoding, and each gene bit represents a part of a parameter. The actual encoding method may be more complex) Parent individual B: Kp = 2.5, Ki = 0.8, Kd = 0.3 (encoded as genotype: 1010110010) We use the single-point crossover method. Randomly select a crossover point (such as the 5th bit), and then exchange the part of the genes after this point: The crossed offspring individual A': Kp = 2.0, Ki = 0.8, Kd = 0.3 (new genotype: 1010000010) The crossed offspring individual B': Kp = 2.5, Ki = 1.0, Kd = 0.5 (new genotype: 1010111101) In this way, we generate two new offspring individuals A' and B' through the crossover operation. They inherit part of the genes of parent individuals A and B respectively, and at the same time introduce new gene combinations. 2. Example of mutation operation Next, we perform a mutation operation on the offspring individual A'. We randomly select a gene bit (such as the 7th bit), and then change the gene at this bit (such as from 0 to 1): The mutated offspring individual A": Kp = 2.0 (slightly changed, but for simplicity of explanation here, the actual change may be more complex), Ki = 0.8 (slightly changed, but the encoding here does not directly reflect the specific numerical change), Kd = a value after a slight adjustment (since we changed the gene bit, the specific value of Kd will change, but it cannot be directly given here because the actual mutation operation may involve more complex decoding and calculation processes).

[0109] Through the above steps, this embodiment discloses the process of optimizing the initial PID control parameters according to the genetic algorithm.

[0110] The PID parameter combination optimized by the genetic algorithm in this embodiment can usually make the system have better dynamic performance and steady-state performance. Specifically, the optimized PID controller can respond to system changes faster, reduce overshoot and oscillation phenomena, and improve the stability and accuracy of the system. This helps to improve the performance of the entire control system and meet higher control requirements.

[0111] The genetic algorithm will consider various possible parameter combinations during the optimization process and screen out the optimal solution through iterative evolution. This comprehensive search method helps to enhance the robustness of the system, enabling the system to maintain stable operation in the face of external disturbances or parameter changes.

[0112] In S170, the optimized PID control parameters, rate of change, and first difference are input into the opening algorithm to obtain the opening of the steam valve. The opening algorithm is as follows:

[0113]

[0114] where k p represents the optimized proportional coefficient;

[0115] k i represents the optimized integral coefficient;

[0116] k d represents the optimized differential coefficient;

[0117] e(t) represents the first difference.

[0118] In this embodiment, after S170, S180 is further included. As Figure 4 shown, S180 includes the following steps:

[0119] S181: Obtain the actual barrel wall temperature collected by all temperature sensors, where all the temperature sensors are evenly distributed on the temperature measurement surface of the barrel wall of the cut tobacco dryer;

[0120] S182: Based on the clustering algorithm, cluster all the actual barrel wall temperatures into several clusters;

[0121] S183: Based on the actual barrel wall temperature corresponding to the clustering center of each cluster, obtain the median value of the actual barrel wall temperature;

[0122] S184: Calculate the second difference between each actual barrel wall temperature and the median value of the actual barrel wall temperature, and screen out the clustering centers with the second difference greater than the difference threshold;

[0123] S185: Send the positions corresponding to the clustering centers with the second difference greater than the difference threshold to the display terminal.

[0124] In S181, all the temperature sensors are used to collect the actual barrel wall temperature of the corresponding area of the barrel wall, which can be realized by arranging one temperature sensor in each area. In this embodiment, the temperature sensor can be a wireless temperature sensor. The wireless temperature sensor is built-in with a battery, the wireless temperature sensor contacts the barrel wall, and the wireless temperature sensor communicates with the data acquisition system through a wireless signal to realize the acquisition of the barrel wall temperature.

[0125] In S182, the difference between the actual values of the barrel wall temperatures in each cluster is small, which can indicate that in the same cluster, the temperature difference in the area corresponding to the actual values of the barrel wall temperatures is small.

[0126] In S183 and S184, when the gap from the median value is too large, it indicates that there is a problem of uneven temperature distribution on the cylinder wall, which means that the difference between this clustering center and other clustering centers is too large, and there is an uneven temperature distribution problem on the cylinder wall. Therefore, it is necessary to send the position information of this area to the display terminal. After receiving the relevant reminder, the relevant engineers can perform maintenance in a timely manner. The position can be sent in the form of coordinates, or the position of this clustering center can be displayed on the three-dimensional model of the cut tobacco dryer on the display terminal.

[0127] It should be noted that those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described electronic device can refer to the corresponding processes of each step in the foregoing method, and will not be elaborated herein.

[0128] The embodiment of the present application also provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium. When the computer program runs on a computer, it causes the computer to execute the method for regulating the temperature of the cylinder wall of the cut tobacco dryer as described in the above embodiment.

[0129] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware, or can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes several instructions to enable a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0130] In the embodiments provided by the present application, it should be understood that the disclosed method can also be implemented in other ways. The method embodiments described above are only illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of the methods and computer program products according to multiple embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, the program segment, or the part of code includes one or more executable instructions for implementing the specified logical function. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions. In addition, the various functional modules in the embodiments of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0131] The above are only examples of the present application and are not intended to limit the protection scope of the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A method for controlling the temperature of a thin-plate wire drying machine barrel wall, characterized in that: The method comprises: Collect ambient temperature, target moisture content, instantaneous flow rate of tobacco leaves and temperature of tobacco cuts; The ambient temperature, target outlet moisture content, instantaneous flow rate of tobacco leaves and temperature of cut tobacco are input into the trained prediction model to obtain the predicted temperature of the cylinder wall; Adjusting the temperature of the cylinder wall of the thin-plate torrefaction drying machine to the predicted cylinder wall temperature; Obtain the actual outlet moisture content of tobacco; Obtaining initial PID control parameters based on a first difference between the actual outlet moisture content and the target outlet moisture content, and a rate of change of the first difference over time; The initial PID control parameters are input into an optimization model, wherein the optimization model optimizes the initial PID control parameters based on a genetic algorithm to obtain optimized PID control parameters; Based on the optimized PID control parameters, the opening of the steam valve is adjusted to adjust the drum wall temperature so that the first difference is less than or equal to a first difference threshold.

2. The method according to claim 1, characterized in that After adjusting the opening of the steam valve based on the optimized PID control parameters, the method further includes: Acquiring the actual temperature of the drum wall collected by all temperature sensors, wherein all the temperature sensors are evenly distributed on the temperature measuring surface of the drum wall of the tofu drying machine; Based on the clustering algorithm, all the actual temperatures of the cylinder wall are clustered into several clusters; Based on the actual temperature of the cylinder wall corresponding to the cluster center of each cluster, a median value of the actual temperature of the cylinder wall is obtained; Calculate the second difference between each of the actual cylinder wall temperatures and the median of the actual cylinder wall temperatures, and select the cluster center whose second difference is greater than a difference threshold; The position corresponding to the cluster center where the second difference is greater than the difference threshold is sent to the display terminal.

3. The method according to claim 1, characterized in that The step of inputting the ambient temperature, target outlet moisture content, instantaneous flow rate of tobacco leaves and cut tobacco temperature into the trained prediction model to obtain the predicted cylinder wall temperature includes: Preset the deep learning model to obtain the pre-training prediction model; Collecting several groups of historical data, the historical data including historical ambient temperature, historical outlet moisture content, historical tobacco leaf instantaneous flow rate and historical tobacco cut temperature corresponding to the time sequence; All the historical data are divided into a training set and a test set, and input into the pre-training prediction model to obtain the post-training prediction model. The ambient temperature, target outlet moisture content, instantaneous flow rate of tobacco leaves and tobacco shred temperature are input into the post-training prediction model to obtain the predicted temperature of the barrel wall.

4. The method according to claim 1, characterized in that: The initial PID control parameters are input into an optimization model, and the optimization model optimizes the initial PID control parameters based on a genetic algorithm to obtain optimized PID control parameters, including: Based on a preset range of the PID control parameter, a plurality of real discrete points of the PID control parameter are obtained, wherein one of the real discrete points of the PID control parameter is an initial PID control parameter; Based on the real discrete points of the PID control parameters and the fitness function, obtaining the fitness corresponding to the real discrete points of the PID control parameters, wherein the fitness is used to characterize the first difference; The real discrete points of the PID control parameters corresponding to the fitness values ​​being less than or equal to the fitness threshold are used as parent discrete points, and iterated until the optimized PID control parameters are obtained; The iterations include: Based on the parent discrete points, crossover and / or mutation are performed to obtain offspring discrete points; Inputting the offspring discrete points into the fitness function; Selecting the offspring discrete points whose fitness is less than or equal to the fitness threshold as the parent discrete points; Repeat the steps of performing crossover and / or mutation based on the parent discrete points to obtain offspring discrete points; inputting the offspring discrete points into the fitness function; and selecting offspring discrete points whose fitness is less than or equal to a fitness threshold as parent discrete points.

5. The method according to claim 4, characterized in that The fitness function is: Fitness=|Q actual -Q target | Among them, Fitness means fitness; Q actual represents the simulated outlet moisture content; Q target represents the target moisture content; The obtaining, based on the real discrete points of the PID control parameters and the fitness function, the fitness corresponding to the real discrete points of the PID control parameters comprises: The real discrete points of the PID control parameters are input into the simulation system, and the simulation system outputs Q actual ; Based on the fitness function, the fitness is calculated.

6. The method according to claim 1, characterized in that The adjusting the opening of the steam valve based on the optimized PID control parameter comprises: The optimized PID control parameter, the change rate and the first difference are input into the opening algorithm to obtain the opening of the steam valve. The opening algorithm is: Among them, k p Represents the optimized proportional coefficient; k i Represents the optimized integral coefficient; k d represents the optimized differential coefficient; e(t) represents the first difference.

7. An electronic device, characterized in that: The electronic device comprises a processor and a memory coupled to each other, wherein the memory stores a computer program. When the computer program is executed by the processor, the electronic device executes the method according to any one of claims 1 to 6.

8. A computer program product, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 6.

Images