A temperature balancing control method and device for a cigarette factory production area

Abstract

The application discloses a temperature balance control method for a production area of a cigarette factory, and belongs to the technical field of temperature control. The method comprises the following steps: dividing the production area into multiple temperature regulation areas; collecting temperature data of each temperature regulation area in real time and generating a plane heat map; generating a branch valve opening control value according to a difference in the temperature regulation area, so as to adjust a temperature difference between the multiple temperature regulation areas; dividing the plane heat map into multiple temperature units by using a clustering algorithm; calculating an average temperature of each temperature unit respectively, and generating a real-time temperature matrix; generating an air outlet valve opening control value according to the temperature matrix, so as to adjust a temperature difference between the multiple temperature units; and dividing the production area into multiple temperature regulation areas, adjusting a temperature difference between the regulation areas, and then dividing the temperature regulation areas into multiple temperature units, and finely adjusting a temperature difference between the temperature units. Through the area division and twice temperature adjustment, the method has higher fineness, and ensures temperature uniformity and stability of the whole production area.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of production control of cigarette factories, and more particularly to a temperature balance control method and device for a production area of a cigarette factory. BACKGROUND

[0002] With the continuous development of modern industrial production technology and the continuous improvement of product processing precision, the requirements for fine processing technology of tobacco are continuously improved. Since the raw and auxiliary materials and the production process in the production process of a cigarette factory have relatively strict requirements on the temperature and humidity of the environment, the deviation and fluctuation of the temperature and humidity will affect the quality of the products, which puts forward higher requirements on the production environment of the cigarette factory. The temperature and humidity control of the production and storage of the cigarette factory mostly adopts centralized control of large process air conditioners, which generally adopts a full air system to supply and provide a stable and accurate temperature and humidity environment for the controlled area.

[0003] Taking a process air conditioner and its control area in a cigarette wrapping workshop as an example, the control strategy adopts the average value of a group of temperature and humidity sensors in the area as the current space humidity state, controls the average temperature of the production area environment through the central value, and uses some advanced control theory methods to control the fluctuation of the average temperature of the environment within a small range. Since the average value algorithm is used, there are many heat source factors in the area, and even if the average value meets the technical requirements, the variance of all temperature data is large, and the value difference between the data is large, which causes the temperature deviation of the maximum and minimum value areas to affect the production quality. SUMMARY

[0004] An object of the present application is to provide a new technical solution for a temperature balance control method for a production area of a cigarette factory, which has higher precision after area division and twice temperature adjustment, and ensures the temperature uniformity and stability of the entire production area.

[0005] According to a first aspect of the present application, a temperature balance control method for a production area of a cigarette factory is provided, comprising:

[0006] dividing the production area into a plurality of temperature adjustment areas;

[0007] collecting temperature data of each temperature adjustment area in real time, and generating a plane thermal map;

[0008] generating a branch valve opening control value according to the difference in the temperature adjustment area, to adjust the temperature difference between the plurality of temperature adjustment areas;

[0009] using a clustering algorithm to divide the plane thermal map into a plurality of temperature units;

[0010] calculating the average temperature of each temperature unit respectively, and generating a real-time temperature matrix;

[0011] The outlet valve opening control value is generated based on the temperature matrix to adjust the temperature difference between the multiple temperature units.

[0012] Optionally, adjusting the temperature difference between the plurality of temperature regulation zones specifically includes:

[0013] Obtain the predetermined temperature value for each temperature regulation zone;

[0014] The temperature data of each temperature regulation zone is collected in real time and compared with the predetermined temperature value of the temperature regulation zone to obtain the temperature difference of each temperature regulation zone.

[0015] Multiple deviation values ​​are obtained by comparing the temperature differences.

[0016] The deviation is input into the corresponding PID controller to obtain the corresponding branch valve opening control value.

[0017] Optionally, the step of using a clustering algorithm to divide the planar heatmap into multiple temperature units specifically includes:

[0018] The planar heatmap is segmented into pixels, and multidimensional feature vector data points about the pixels and their corresponding temperature values ​​are generated.

[0019] Obtain the number of categories, and randomly select several values ​​from the temperature values ​​as the initial cluster centers;

[0020] Based on the initial cluster center, calculate the Manhattan distance for each data point, and assign the data points to the corresponding categories based on the Manhattan distance;

[0021] Calculate the mean of the data for each category and use it as the new cluster center;

[0022] The mean of the data for each category is repeatedly calculated and used as the iterative cluster center until the termination condition is met, generating the cutoff cluster center.

[0023] Optionally, the termination condition is that the iterative clustering center point is less than a set threshold; or the iterative clustering center point reaches a preset number of iterations.

[0024] Alternatively, the Manhattan distance can be calculated using the following formula:

[0025] d n,m =|n1-m1|+|n2-m2|;

[0026] Where, d n,m Let (n1, n2) represent the Manhattan distance, (n1, n2) be the data points, and (m1, m2) represent the cutoff cluster center points.

[0027] Optionally, the adjusting the temperature difference between the plurality of temperature units specifically comprises:

[0028] establishing an initial correlation model of the temperature matrix and the outlet valve opening degree;

[0029] obtaining a prediction model according to the current temperature matrix and the current opening degree value of the outlet valve;

[0030] calculating a performance index of the prediction model;

[0031] designing a control input sequence according to the prediction model and the performance index, minimizing the performance index, and obtaining a final correlation model.

[0032] Optionally, the initial correlation model is a discrete-time linear time-invariant system model, specifically:

[0033] x(k+1) = Ax(k) + Bu(k);

[0034] wherein x(k) is a system state, u(k) is a control input, and A and B are system matrices.

[0035] Optionally, the performance index calculation formula of the prediction model is:

[0036]

[0037] wherein y i is an actual temperature value of the i-th temperature unit at the current time, r i is a target temperature value of the temperature unit, and N is the number of temperature units.

[0038] Optionally, the method further comprises adjusting the stability of the final correlation model, specifically comprising:

[0039] obtaining the opening degree of the actual output, and collecting the temperature values of the temperature units in a unit time in real time;

[0040] and calculating a stability coefficient according to the target temperature value;

[0041] when the stability coefficient meets the threshold condition, obtaining the final correlation model of the temperature matrix and the outlet valve opening degree.

[0042] According to a second aspect of the present application, a temperature balancing control device for a production area of a cigarette factory comprises:

[0043] a division module that divides the production area into a plurality of temperature regulation areas;

[0044] a collection module that collects temperature data of each temperature regulation area in real time and generates a plane heat map;

[0045] A plurality of air supply pipelines are respectively arranged in the temperature regulation areas and are respectively provided with branch valves;

[0046] A temperature adjustment module generates a branch valve opening degree control value according to a difference in the temperature regulation areas, so as to adjust a temperature difference between the plurality of temperature regulation areas;

[0047] A segmentation module segments the plane heat map into a plurality of temperature units by using a clustering algorithm;

[0048] A calculation module respectively calculates an average temperature of each temperature unit to generate a real-time temperature matrix;

[0049] An adjustment module generates an air outlet valve opening degree control value according to the temperature matrix, so as to adjust a temperature difference between the plurality of temperature units.

[0050] According to one embodiment of the present disclosure, the present application provides a new technical solution of a temperature balance control method for a production area of a cigarette factory. The method has higher precision through area division and twice temperature regulation, and ensures temperature uniformity and stability of the entire production area.

[0051] Other features and advantages of the present application will become clear from the following detailed description of exemplary embodiments thereof with reference made to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings incorporated in and forming a part of the specification illustrate embodiments of the present application and, together with the description, serve to explain the principles of the application.

[0053] Figure 1 A flowchart of a temperature balance control method for a production area of a cigarette factory according to the present application.

[0054] Figure 2 A process structure diagram of a production area of a cigarette factory according to the present application.

[0055] Figure 3 A control flowchart of adjusting a temperature difference between a plurality of temperature regulation areas according to the present application.

[0056] Figure 4 A control flowchart of segmenting a plane heat map into a plurality of temperature units by using a clustering algorithm according to the present application.

[0057] Figure 5 A flowchart of adjusting a temperature difference between a plurality of temperature units according to the present application.

[0058] Figure 6 A flowchart of a temperature balance system for a production area of a cigarette factory according to the present application.

[0059] Figure 7 Flowchart of the temperature balance control device for the production area of a cigarette factory according to the present application. DETAILED DESCRIPTION

[0060] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of the components and steps set forth in the examples, as well as the numerical expressions and values, are not limiting to the scope of the present application unless otherwise specifically stated.

[0061] The following description of at least one exemplary embodiment is merely exemplary in nature and is in no way intended to limit the scope of the application its application or uses.

[0062] Techniques, methods, and devices known to those of ordinary skill in the relevant art can not be discussed in detail herein. However, the techniques, methods, and devices are sufficiently described in the disclosure.

[0063] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as a limitation. Thus, other examples of the exemplary embodiments can have different values.

[0064] As shown in Figure 1 The present application discloses a temperature balance control method for a production area of a cigarette factory, comprising:

[0065] Step S110, dividing the production area into multiple temperature regulation areas;

[0066] Step S120, collecting temperature data of each temperature regulation area in real time, and generating a plane heat map;

[0067] Step S130, generating a branch valve opening control value according to the difference in the temperature regulation area, to adjust the temperature difference between the multiple temperature regulation areas;

[0068] Step S140, using a clustering algorithm to divide the plane heat map into multiple temperature units;

[0069] Step S150, calculating the average temperature of each temperature unit respectively, and generating a real-time temperature matrix;

[0070] Step S160, generating an air outlet valve opening control value according to the temperature matrix, to adjust the temperature difference between the multiple temperature units.

[0071] Specifically, the whole production area is taken as a first-level area, and is divided into respective second-level areas according to process production lines, and the air volume of the air supply pipe branch is controlled to balance the average temperature between the second-level areas; in the second-level areas, the temperature matrix collected above is used to adjust the opening of the corresponding air outlet valve according to the average difference of each temperature data.

[0072] It should be particularly pointed out that the new technical scheme of the temperature balancing control method for the production area of a cigarette factory provided in the embodiment has higher precision after area division and twice temperature adjustment, and ensures the temperature uniformity and stability of the whole production area.

[0073] As shown in Figure 3 , in step S130, a branch valve opening control value is generated according to the difference in the temperature adjustment area to adjust the temperature difference between the plurality of temperature adjustment areas, and specifically includes the following steps:

[0074] In step S131, a predetermined temperature value of each temperature adjustment area is obtained;

[0075] In step S132, temperature data of each temperature adjustment area is collected in real time, and is compared with the predetermined temperature value of the temperature adjustment area respectively to obtain a difference temperature of each temperature adjustment area;

[0076] In step S133, a plurality of deviation amounts are obtained by comparing the difference temperatures respectively;

[0077] In step S134, the deviation amounts are input into corresponding PID controllers respectively to obtain corresponding branch valve opening control values.

[0078] As shown in Figure 2 , 6 , in actual application, taking a process air conditioning action area as an example, according to the production distribution on site, the area is divided into three areas A, B and C, and three area average temperatures (T a , T b , T c ) are set. Air valve devices are installed on the air supply branch pipes between the three areas, and the openings (V ab , V bc ) are controlled by PLC.

[0079] The air valve devices are controlled by PID, and the differences (ΔT a , ΔT b , ΔT c ) between the three area average temperatures and the total average temperature T avg are calculated, the difference (ΔT a -ΔT b ) between the difference of the A area and the B area and the difference of the C area and the B area is calculated, and the branch valve opening control value of the A area is obtained by inputting the difference into the PID controller.c -ΔT b ) respectively as the input of two PID controllers, the output value of the controller is obtained by the PID algorithm as the air valve opening control value (V ab ,V bc ) on the air pipe between the AB area and the BC area. nm The above A area is taken as an example. The air outlet is distributed in a matrix at the top of the production workshop, and an air valve device is installed on each air outlet, and the opening V n,m is controlled by the PLC, wherein m and n are the row and column numbers of the air outlet.

[0080] It should be particularly pointed out that the air valve opening in the embodiment is controlled by the PID controller, which can adjust the air valve opening in real time, ensure real-time online adjustment of the opening when the temperature changes, and further ensure the temperature uniformity between the areas.

[0081] As shown in Figure 4 , the temperature data of each temperature regulation area is collected in real time, and a plane thermal map is generated; the plane thermal map is segmented into multiple temperature units by using a clustering algorithm; specifically including:

[0082] Step S141, pixel point segmentation is performed on the plane thermal map, and a multi-dimensional feature vector data point about the pixel point and the temperature value corresponding to the pixel point is generated;

[0083] Step S142, the number of categories is obtained, and a number of values in the temperature value are randomly selected as initial clustering center points;

[0084] Step S143, according to the initial clustering center points, the Manhattan distance corresponding to each data point is calculated respectively, and the data points are distributed into the corresponding categories according to the Manhattan distance;

[0085] Step S144, the data mean of each category is calculated and taken as a new clustering center point;

[0086] Step S145, repeat step S143 and step S144, and take them as iterative clustering center points, until the termination condition is met, and generate a stop clustering center point.

[0087] In a preferred embodiment, the termination condition is that the iterative clustering center point is less than a set threshold value; or the iterative clustering center point reaches a preset iteration number. It should be particularly pointed out that the Manhattan distance calculation formula is:

[0088] d n,m = |n1-m1|+|n2-m2|;

[0089] Wherein, d n,m represents the Manhattan distance, (n1, n2) is a data point, and (m1, m2) represents a stop clustering center point.

[0090] Clustering algorithms are used to segment planar heatmaps into temperature units. The K-means algorithm will be used as an example for further explanation:

[0091] First, the collected planar heat map data is converted into pixel coordinates and corresponding temperature values ​​T. xy This forms a P-dimensional feature vector, where P is the number of pixels;

[0092] Then, select an appropriate K value based on actual needs. The K value represents the number of categories to be divided into. Randomly select K values ​​from the above P data points as the initial cluster centers.

[0093] Then, the Manhattan distance d between each data point and each cluster center is calculated. n,m =|n1-m1|+|n2-m2|, which assigns each data point to the category of the nearest cluster center.

[0094] For each category, calculate the mean of all data points in that category and use it as the new cluster center for that category.

[0095] Repeat the above steps until the offset distance of each cluster center point is less than the set threshold or the maximum number of iterations is reached.

[0096] Output the cluster label to which each data point belongs, thereby dividing the planar heatmap into P small regions.

[0097] Calculate the average temperature of each of the small regions obtained above, and form a real-time temperature matrix (T1, T2, ..., T...). p Then, based on the effective range of each air outlet, a vertical projection is performed, and this is correlated with the temperature data mentioned above.

[0098] like Figure 5 As shown, step S160, generating the outlet valve opening control value based on the temperature matrix to adjust the temperature difference between the multiple temperature units, specifically includes:

[0099] Step S161: Establish an initial correlation model between the temperature matrix and the opening degree of the air outlet valve;

[0100] Step S162: Obtain the prediction model based on the current temperature matrix and the current opening value of the air outlet valve;

[0101] Step S163: Calculate the performance metrics of the prediction model;

[0102] Step S164: Based on the prediction model and the performance index, design a control input sequence to minimize the performance index and obtain the final association model.

[0103] In a preferred embodiment, the initial association model is a discrete-time linear time-invariant system model, specifically:

[0104] x(k+1) = Ax(k) + Bu(k);

[0105] where x(k) is the system state, u(k) is the control input, and A and B are system matrices.

[0106] In a preferred embodiment, the performance index of the prediction model is calculated as:

[0107]

[0108] where y i is the actual temperature value of the i-th temperature unit at the current time, r i is the target temperature value of the temperature unit, and N is the number of temperature units.

[0109] Based on the above real-time temperature matrix, the DMPC algorithm is used to control the opening size of the associated air supply outlet, and the specific steps are:

[0110] Modeling, modeling the system as a discrete-time linear time-invariant system, which can be represented as:

[0111] x(k+1) = Ax(k) + Bu(k)

[0112] where x(k) is the system state, u(k) is the control input, and A and B are system matrices.

[0113] Based on the current state and control input, the system response in the future is predicted, and the prediction model is obtained:

[0114]

[0115] where x(k+i+1|k) is the state prediction value at the i+1 step at time k.

[0116] Select an appropriate objective function or performance index to evaluate the control effect of the prediction model, commonly using the average deviation square sum:

[0117]

[0118] where y i is the actual temperature value of the i-th small space at the current time, r i is the target temperature value of the small space, and N is the number of small areas.

[0119] According to the prediction model and the performance index, a control input sequence u(k), u(k+1),..., u(k+M-1) is designed to minimize the performance index:

[0120]

[0121]

[0122] u min ≤u(k+j)≤u max ,j=0,1,...,M-1

[0123] where M is the length of the control input sequence, u min and u max are the upper and lower bounds of the control input.

[0124] The first control input value in the obtained control input sequence is applied to the system, and the feedback response of the system is obtained, and the above steps are repeated until the system is stable. Specifically, it includes the following steps:

[0125] The opening size of the actual output is obtained, and the temperature values of the temperature units in a unit time are collected in real time;

[0126] And calculate the stability coefficient according to the target temperature value;

[0127] When the stability coefficient meets the threshold condition, the final correlation model of the temperature matrix and the outlet valve opening is obtained.

[0128] Specifically, the calculation formula of the stability coefficient is:

[0129]

[0130] Where C m,n is the stability coefficient, T m,n represents the corresponding temperature of the temperature unit, expresses the target temperature, K m,n represents the actual output opening, expresses the average opening of the valve corresponding to all temperature units.

[0131] When the stability coefficient meets the threshold condition, the final correlation model of the temperature matrix and the outlet valve opening is obtained, and specifically, the threshold condition can be set according to the operation rule, or the extreme value of stability is set to ensure the stability of the final output model.

[0132] Specifically, the algorithm inputs all real-time temperature data, aims to reduce the difference between the average temperature of each small area and the overall average temperature, controls the size of the air supply opening that can affect the temperature value of each small area, and then optimizes the model and objective function through the feedback response of the system to adjust the control design, so as to realize the uniformity and stability of the temperature of the entire area environment.

[0133] In a preferred embodiment, the application also provides a temperature balancing control device for a production area of a cigarette factory, comprising: a division module 210, an acquisition module 220, a plurality of air supply pipelines 230, a temperature adjustment module 240, a segmentation module 250, a calculation module 260, and an adjustment module 270.

[0134] The division module 210 divides the production area into a plurality of temperature adjustment regions; the acquisition module 220 acquires temperature data of each temperature adjustment region in real time and generates a plane heat map; the plurality of air supply pipelines 230 are respectively arranged in the temperature adjustment regions and are respectively provided with branch valves; the temperature adjustment module 240 generates a branch valve opening control value according to the difference in the temperature adjustment region to adjust the temperature difference between the plurality of temperature adjustment regions; the segmentation module 250 divides the plane heat map into a plurality of temperature units by using a clustering algorithm; the calculation module 260 calculates the average temperature of each temperature unit respectively to generate a real-time temperature matrix; and the adjustment module 270 generates an air outlet valve opening control value according to the temperature matrix to adjust the temperature difference between the plurality of temperature units.

[0135] The branch air pipes and the air outlet opening size of the application are controllable, the air supply amount can be controlled by adjusting the opening size, and the temperature and humidity of the production area can be balanced;

[0136] According to the analysis of the space heat load distribution by the plane heat map, the opening size of each air supply valve can be accurately controlled by the established model calculation; the regional hierarchical control can better adapt to the production distribution on site, and can also better guarantee the overall stability of the temperature and humidity balancing control.

[0137] Although some specific embodiments of the application have been described in detail by examples, those skilled in the art should understand that the above examples are only for illustration, not for limiting the scope of the application. Those skilled in the art should understand that the above embodiments can be modified without departing from the scope and spirit of the application. The scope of the application is defined by the appended claims.

Claims

1. A method for temperature balance control in the production area of ​​a cigarette factory, characterized in that, include: The production area is divided into multiple temperature-controlled zones; Real-time temperature data is collected for each temperature regulation zone, and a planar heat map is generated; The branch valve opening control value is generated based on the difference in the temperature regulation area to adjust the temperature difference between the multiple temperature regulation areas. The planar heatmap is divided into multiple temperature units using a clustering algorithm; Calculate the average temperature of each temperature unit to generate a real-time temperature matrix; The outlet valve opening control value is generated based on the temperature matrix to adjust the temperature difference between the multiple temperature units, specifically including: An initial correlation model is established between the temperature matrix and the opening degree of the air outlet valve; the initial correlation model is a discrete-time linear time-invariant system model: x(k+1)=Ax(k)+Bu(k); Where x(k) is the system state, u(k) is the control input, and A and B are the system matrices; Based on the current temperature matrix and the current opening value of the air outlet valve, the prediction model is obtained: in, Let A be the predicted state value at time k at step i+1. i+1 It refers to the system matrix at step i+1, A i-j It refers to the system matrix at step ij, and u(k+j) represents the control input sequence at time k+j; Calculate the performance metrics of the prediction model; the formula for calculating the performance metrics of the prediction model is: Among them, y i Let r be the actual temperature value of the i-th temperature unit at the current moment. i The target temperature value for this temperature unit, where N is the number of temperature units; Based on the prediction model and performance indicators, design control input sequences u(k), u(k+1), ..., u(k+M-1) to minimize the performance indicators, thus obtaining the final association model: u min ≤u(k+j)≤u max ,j=0,1,...,M-1 Where M is the length of the control input sequence, u min and u max To control the upper and lower bounds of the input; Apply the first control input value in the obtained control input sequence to the system, and then obtain the system feedback response. Repeat the above steps until the system is stable.

2. The temperature balance control method for a cigarette factory production area according to claim 1, characterized in that, The adjustment of the temperature difference between the plurality of temperature adjustment zones specifically includes: Obtain the predetermined temperature value for each temperature regulation zone; The temperature data of each temperature regulation zone is collected in real time and compared with the predetermined temperature value of the temperature regulation zone to obtain the temperature difference of each temperature regulation zone. Multiple deviation values ​​are obtained by comparing the temperature differences. The deviation is input into the corresponding PID controller to obtain the corresponding branch valve opening control value.

3. The temperature balance control method for a cigarette factory production area according to claim 1 or 2, characterized in that, The step of using a clustering algorithm to divide the planar heatmap into multiple temperature units specifically includes: The planar heatmap is segmented into pixels, and multidimensional feature vector data points about the pixels and their corresponding temperature values ​​are generated. Obtain the number of categories, and randomly select several values ​​from the temperature values ​​as the initial cluster centers; Based on the initial cluster center, calculate the Manhattan distance for each data point, and assign the data points to the corresponding categories based on the Manhattan distance; Calculate the mean of the data for each category and use it as the new cluster center; The mean of the data for each category is repeatedly calculated and used as the iterative cluster center until the termination condition is met, generating the cutoff cluster center.

4. The temperature balance control method for a cigarette factory production area according to claim 3, characterized in that, The termination condition is that the iterative clustering center point is less than a set threshold; or the iterative clustering center point reaches a preset number of iterations.

5. The temperature balance control method for a cigarette factory production area according to claim 4, characterized in that, The formula for calculating Manhattan distance is: d n,m =|n1-m1|+|n2-m2|; Where, d n,m Let (n1, n2) represent the Manhattan distance, (n1, n2) be the data points, and (m1, m2) represent the cutoff cluster center points.

6. The temperature balance control method for a cigarette factory production area according to any one of claims 1 to 5, characterized in that, It also includes adjusting the stability of the final association model, specifically including: Obtain the actual output opening size and collect the temperature value of the temperature unit per unit time in real time at intervals; And calculate the stability coefficient based on the target temperature value; When the stability coefficient meets the threshold condition, the final correlation model between the temperature matrix and the opening degree of the air outlet valve is obtained.

Images