A synchronous control method for gypsum board production based on main line control model

Abstract

The present invention discloses a synchronous control method for gypsum board production based on a main line control model, comprising the following steps: setting a virtual servo axis in the main line synchronization model of the gypsum board production line, and setting an adjustment coefficient for each physical servo axis in the main line synchronization model according to the virtual servo axis and the expected main line speed of the gypsum board; monitoring the actual main line speed of the gypsum board controlled by the physical servo axis, and performing a time series prediction on the actual main line speed of the gypsum board to obtain the actual main line speed of the gypsum board at a future time series, and comparing the actual main line speed of the gypsum board at the future time series with the expected main line speed of the gypsum board to determine the control target unit. The present invention realizes the advanced control of the main line synchronization model to reduce the asynchronous operation time, and respectively constructs the mapping relationship between the virtual servo axis and the working condition time series, and the adjustment coefficient of each physical servo axis and the working condition time series, so as to maintain the measurement accuracy of the control target unit while reducing the measurement complexity of the control target unit.

Description

Technical Field

[0001] The present invention relates to the technical field of gypsum board production, and in particular to a gypsum board production synchronous control method based on a main line control model. Background Art

[0002] The main line system of the gypsum board production line mainly consists of two parts: the forming belt and the open roller. During actual production, in order to avoid phenomena such as breakage, board piling, and substandard quality, the various equipment in the main line system need to operate synchronously, including speed synchronization between belts, between belts and open rollers, and between open rollers.

[0003] The matching of the main line speed is mainly achieved by real-time feedback between PLC, servo control system and encoder. When the production starts and the upper computer sets the production specifications and main line speed, the PLC will distribute the read information to the unloading system in real time. After the main line system receives the command, it will be transmitted to the servo control system in real time. The servo controls the motor to run to the set speed. At the same time, the encoder can try to read the running speed of the molding system equipment and feed it back to the controller. Since the servo controller has a fast response speed and can reduce the motor's rotational inertia, it can quickly meet the production setting requirements.

[0004] In the existing synchronous control method for gypsum board production, the PLC of each motor establishes a virtual servo control axis. During production, the production speed set by the host computer is input into the virtual servo control axis after PLC calculation. The speed, position and other information of the virtual axis are synchronized to the actual servo control axis in real time. When a deviation between the speed and the set value is detected, the virtual servo axis can quickly make a correction and transmit it to each servo controller at the same time, thereby achieving speed synchronization of each motor. However, the correction of the virtual servo axis in the existing technology requires corrective calculation after detecting the speed deviation, and the corrected data can only be applied to the virtual servo axis. The correction process of the virtual servo axis is a delayed correction, and the time required for the correction calculation process becomes the time of asynchronous operation. The asynchronous operation of the gypsum board will directly affect the production effect of the gypsum board. Summary of the Invention

[0005] The object of the present invention is to provide a method for synchronous control of gypsum board production based on a main line control model, so as to solve the technical problem of reducing the asynchronous operation existing in the gypsum board production line in the prior art.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0007] A synchronous control method for gypsum board production based on a mainline control model comprises the following steps:

[0008] Step S1: Setting a virtual servo axis in a main line synchronization model of a gypsum board production line, and setting an adjustment coefficient for each physical servo axis in the main line synchronization model based on the virtual servo axis and the expected main line speed of the gypsum board. The main line synchronization model is a control model for the synchronous operation of the main line components of the production line in a master-multiple slave servo control mode composed of multiple physical servo axes;

[0009] Step S2: monitoring the actual main line speed of the gypsum board controlled by the physical servo axis, performing a time series prediction on the actual main line speed of the gypsum board to obtain the actual main line speed of the gypsum board at a future time series, and comparing the actual main line speed of the gypsum board at the future time series with the expected main line speed of the gypsum board to determine a control target unit, wherein the control target unit includes the adjustment coefficients of the virtual servo axis and each physical servo axis;

[0010] Step S3: determining in real time the value of the control target unit at a future time sequence, so as to achieve advanced control of the main line synchronization model and reduce the asynchronous operation time;

[0011] Step S4: respectively construct mapping relationships between the virtual servo axis and the working condition timing, and between the adjustment coefficient of each physical servo axis and the working condition timing, so as to maintain the accuracy of the calculation of the control target unit while reducing the complexity of the calculation of the control target unit.

[0012] As a preferred solution of the present invention, the physical servo axis includes a first physical servo axis, a second physical servo axis, a third physical servo axis, a fourth physical servo axis and a fifth physical servo axis, and the main line components of the production line correspond one-to-one to the physical servo axes. The main line components of the production line include a first belt controlled by the first physical servo axis, a second belt controlled by the second physical servo axis, a third belt controlled by the third physical servo axis, a first open roller controlled by the fourth physical servo axis and a second open roller controlled by the fifth physical servo axis.

[0013] As a preferred solution of the present invention, the actual main line speed of the gypsum board includes the actual speed of the main line components of the production line, and the expected main line speed of the gypsum board includes the expected speed of the main line components of the production line.

[0014] As a preferred solution of the present invention, setting an adjustment coefficient for each physical servo axis in the main line synchronization model according to the virtual servo axis and the expected main line speed of the gypsum board includes:

[0015] Setting multiple selected values for the desired main line speed of the gypsum board, and sequentially setting the speed of the virtual servo axis to the multiple selected values;

[0016] Trial production was carried out at each speed of the virtual servo axis. During each trial production process, the adjustment coefficient of each physical servo axis was measured based on the actual main line speed of the gypsum board and the speed of the virtual servo axis. Based on the speed of the virtual servo axis and the adjustment coefficient of each physical servo axis during each trial production process, an adjustment coefficient setting function was constructed to implement the setting of the adjustment coefficient in the main line synchronization model.

[0017] The function expression of the adjustment coefficient setting function is:

[0018] (A, B, C, D, E) = BP(V);

[0019] Where A, B, C, D, and E are the adjustment coefficients of the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis, respectively; V is the speed of the virtual servo axis; and BP is the function body of the adjustment coefficient setting function.

[0020] As a preferred solution of the present invention, the adjustment coefficient setting function is obtained by training a BP neural network using the speed of the virtual servo axis and the adjustment coefficients of each physical servo axis in each trial production process as data samples.

[0021] As a preferred solution of the present invention, the monitoring of the actual main line speed of the gypsum board controlled by the physical servo axis includes:

[0022] Real-time monitoring of the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, and using the monitoring sequence as a sequence attribute of the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller;

[0023] The actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller are arranged according to the monitoring time sequence to obtain the actual speed time series of the first belt, the second belt, the third belt, the first open roller and the second open roller.

[0024] As a preferred solution of the present invention, the method of performing time series prediction on the actual main line speed of the gypsum board to obtain the actual main line speed of the gypsum board at a future time series includes:

[0025] The actual speed prediction models of the first belt, the second belt, the third belt, the first open roller and the second open roller are obtained by using an LSTM neural network to perform network training based on the actual speed time series of the first belt, the second belt, the third belt, the first open roller and the second open roller;

[0026] The actual speed prediction model of the first belt, the second belt, the third belt, the first open roller and the second open roller is used to predict the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller at future time sequences.

[0027] As a preferred solution of the present invention, the step of comparing the actual main line speed of the gypsum board at a future time sequence with the expected main line speed of the gypsum board to determine the control target unit includes:

[0028] Compare the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller at the future time sequence with the expected speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, respectively, wherein,

[0029] If the actual speed of the first belt is inconsistent with the expected speed of the first belt, the control target unit at the future timing is the first physical servo axis;

[0030] If the actual speed of the second belt is inconsistent with the expected speed of the second belt, the control target unit at the future timing is the second physical servo axis;

[0031] If the actual speed of the third belt is inconsistent with the expected speed of the third belt, the control target unit at the future timing is the third physical servo axis;

[0032] If the actual speed of the first open roller conveyor is inconsistent with the desired speed of the first open roller conveyor, the target unit to be controlled at the future timing is the fourth physical servo axis;

[0033] If the actual speed of the second open roller table is inconsistent with the expected speed of the second open roller table, the control target unit at the future timing is the fifth physical servo axis;

[0034] If the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller are inconsistent with the expected speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, the control target unit at the future timing is the virtual servo axis.

[0035] As a preferred solution of the present invention, the real-time determination of the value of the control target unit at a future time sequence includes:

[0036] When the control target units are the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis, the adjustment coefficients of the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis at the future time sequence are determined using the actual speeds of the first belt, the second belt, the third belt, the first open roller, and the second open roller at the future time sequence and the speeds of the virtual servo axis;

[0037] When the control target unit is a virtual servo axis, the speed of the virtual servo axis at the future time sequence is determined by using the deviation between the actual speed of the first belt, the second belt, the third belt, the first open roller and the second open roller at the future time sequence and the expected speed of the first belt, the second belt, the third belt, the first open roller and the second open roller at the future time sequence.

[0038] As a preferred solution of the present invention, the mapping relationships between the virtual servo axis and the working condition timing, and the adjustment coefficient of each physical servo axis and the working condition timing are respectively constructed, including:

[0039] The speed of the virtual servo axis at each working condition time sequence is obtained and arranged in time sequence to obtain a virtual servo axis speed sequence. The LSTM neural network is used to perform network training based on the virtual servo axis speed sequence to obtain a virtual servo axis speed prediction model that characterizes the mapping relationship between the virtual servo axis and the working condition time sequence.

[0040] The speed of each physical servo axis at each working condition timing is obtained and arranged in time series to construct a physical servo axis adjustment coefficient sequence. The LSTM neural network is used to perform network training based on the physical servo axis adjustment coefficient sequence to obtain a prediction model for each physical servo axis adjustment coefficient that characterizes the mapping relationship between the adjustment coefficient of each physical servo axis and the working condition timing.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention performs time-series prediction on the actual main line speed of the gypsum board to obtain the actual main line speed of the gypsum board at the future time sequence, compares the actual main line speed of the gypsum board at the future time sequence with the expected main line speed of the gypsum board to determine the control target unit, and determines the value of the control target unit at the future time sequence in real time to achieve advanced control of the main line synchronization model to achieve a reduction in the asynchronous operation time, and constructs the mapping relationship between the virtual servo axis and the working condition time sequence, and the adjustment coefficient of each physical servo axis and the working condition time sequence, respectively, to maintain the measurement accuracy of the control target unit while reducing the measurement complexity of the control target unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0044] Figure 1 A flow chart of a synchronous control method for gypsum board production provided by an embodiment of the present invention;

[0045] Figure 2 This is a structural block diagram of the mainline synchronization model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0047] like Figure 1 and Figure 2 As shown, the present invention provides a synchronous control method for gypsum board production based on a main line control model, comprising the following steps:

[0048] Step S1: Setting a virtual servo axis in the main line synchronization model of the gypsum board production line, and setting an adjustment coefficient for each physical servo axis in the main line synchronization model based on the virtual servo axis and the expected main line speed of the gypsum board. The main line synchronization model is a control model for the synchronous operation of the main line components of the production line in a one-master-multiple-slave servo control mode composed of multiple physical servo axes;

[0049] The physical servo axis includes a first physical servo axis, a second physical servo axis, a third physical servo axis, a fourth physical servo axis and a fifth physical servo axis. The main line components of the production line correspond one-to-one to the physical servo axes. The main line components of the production line include a first belt controlled by the first physical servo axis, a second belt controlled by the second physical servo axis, a third belt controlled by the third physical servo axis, a first open roller controlled by the fourth physical servo axis and a second open roller controlled by the fifth physical servo axis.

[0050] The actual main line speed of the gypsum board includes the actual speed of the main line components of the production line, and the expected main line speed of the gypsum board includes the expected speed of the main line components of the production line.

[0051] According to the virtual servo axis and the expected main line speed of the gypsum board, the adjustment coefficients are set for each physical servo axis in the main line synchronization model, including:

[0052] Setting multiple selected values for the desired main line speed of the gypsum board, and sequentially setting the speed of the virtual servo axis to the multiple selected values;

[0053] Trial production was carried out at each speed of the virtual servo axis. During each trial production process, the adjustment coefficient of each physical servo axis was measured based on the actual main line speed of the gypsum board and the speed of the virtual servo axis. Based on the speed of the virtual servo axis and the adjustment coefficient of each physical servo axis during each trial production process, an adjustment coefficient setting function was constructed to implement the setting of the adjustment coefficient in the main line synchronization model.

[0054] The function expression of the adjustment coefficient setting function is:

[0055] (A, B, C, D, E) = BP(V);

[0056] Where A, B, C, D, and E are the adjustment coefficients of the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis, respectively; V is the speed of the virtual servo axis; and BP is the function body of the adjustment coefficient setting function.

[0057] The adjustment coefficient setting function is obtained by training the BP neural network using the speed of the virtual servo axis and the adjustment coefficient of each physical servo axis in each trial production process as data samples.

[0058] Due to the different tension levels between the belts, the belts are subjected to different tensions. Therefore, in the actual production process, when the speed given by the servo controller is the same, the actual running speed of the gypsum board on the belt and the open roller is different. Therefore, an adjustment coefficient is added to the main line synchronization model. The servo control axis corresponding to each belt and open roller corresponds to a different adjustment coefficient. The actual adjustment coefficient can be generated from the actual value of the actual test of the belt operation at different main line speeds during trial production. The following is a table representing the relationship between the adjustment coefficient of the physical servo axis and the main line speed and the virtual servo axis speed.

[0059] Table 1 Characterization of the adjustment coefficient of the physical servo axis

[0060]

[0061] The mainline equipment is speed-controlled using a single-master, multi-slave servo control system. The actual operating speed is determined by multiplying the virtual servo by the adjustment factor for the corresponding belt tension. Each actual servo controller outputs the final speed, which is calculated by the virtual servo based on the actual production settings of the host computer. This single-master, multi-slave servo control system, combined with real-time feedback from external encoders, ensures that the speeds of the various mainline devices remain synchronized with the actual production speeds.

[0062] An adjustment coefficient setting function is constructed to characterize the correlation between the adjustment coefficients of the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis and the fifth physical servo axis and the speed of the virtual servo axis. Therefore, the speed of the virtual servo axis is measured through the adjustment coefficient setting function, and the measurement is highly automated.

[0063] Step S2: monitoring the actual main line speed of the gypsum board controlled by the physical servo axis, performing a time series prediction on the actual main line speed of the gypsum board to obtain the actual main line speed of the gypsum board at a future time series, and comparing the actual main line speed of the gypsum board at the future time series with the expected main line speed of the gypsum board to determine a control target unit. The control target unit includes the adjustment coefficients of the virtual servo axis and each physical servo axis;

[0064] Monitor the actual main line speed of the gypsum board controlled by the physical servo axis, including:

[0065] Real-time monitoring of the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, and using the monitoring sequence as a sequence attribute of the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller;

[0066] The actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller are arranged according to the monitoring time sequence to obtain the actual speed time series of the first belt, the second belt, the third belt, the first open roller and the second open roller.

[0067] The actual main line speed of the gypsum board is predicted in time series to obtain the actual main line speed of the gypsum board at the future time series, including:

[0068] The actual speed prediction models of the first belt, the second belt, the third belt, the first open roller and the second open roller are obtained by using an LSTM neural network to perform network training based on the actual speed time series of the first belt, the second belt, the third belt, the first open roller and the second open roller;

[0069] The actual speed prediction model of the first belt, the second belt, the third belt, the first open roller and the second open roller is used to predict the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller at future time sequences.

[0070] Comparing the actual main line speed of the gypsum board at the future time sequence with the expected main line speed of the gypsum board to determine the control target unit, including:

[0071] Compare the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller at the future time sequence with the expected speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, respectively, wherein,

[0072] If the actual speed of the first belt is inconsistent with the expected speed of the first belt, the control target unit at the future timing is the first physical servo axis;

[0073] If the actual speed of the second belt is inconsistent with the expected speed of the second belt, the control target unit at the future timing is the second physical servo axis;

[0074] If the actual speed of the third belt is inconsistent with the expected speed of the third belt, the control target unit at the future timing is the third physical servo axis;

[0075] If the actual speed of the first open roller conveyor is inconsistent with the desired speed of the first open roller conveyor, the target unit to be controlled at the future timing is the fourth physical servo axis;

[0076] If the actual speed of the second open roller table is inconsistent with the expected speed of the second open roller table, the control target unit at the future timing is the fifth physical servo axis;

[0077] If the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller are inconsistent with the expected speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, the control target unit at the future timing is the virtual servo axis.

[0078] A neural network is used to construct the timing law of the actual main line speed of the gypsum board to predict the actual main line speed of the gypsum board at the future time sequence, and the control target unit at the future time sequence is determined according to the actual main line speed of the gypsum board. The control target unit at the future time sequence can be grasped in advance, and then the control target unit can be corrected in advance, realizing the advanced control of the main line synchronization model to reduce the asynchronous operation time, ensure the stability of the synchronous operation of the production line, and improve the production effect.

[0079] When determining the control target unit, it is divided into the correction of the adjustment coefficient of the physical servo axis and the correction of the speed of the virtual servo axis. For example, when in the future time sequence, only the actual speed of the first belt does not match the expected speed, it means that the adjustment coefficient of the physical servo axis corresponding to the first belt is inaccurate and needs to be corrected. Therefore, the control target unit is determined to be the physical servo axis that drives the first belt. The second belt, the third belt, the first open roller and the second open roller are the same as the above situation and will not be repeated here. When in the future time sequence, the actual speed of the first belt, the second belt, the third belt, the first open roller and the second open roller do not match the expected speed, it means that the speed and position information of the virtual servo axis followed by the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis and the fifth physical servo axis have deviations, so the deviation of the virtual servo axis is corrected. Correction is carried out by category, and the correction is more detailed.

[0080] Step S3: determining in real time the value of the control target unit at a future time sequence, so as to achieve advanced control of the main line synchronization model and reduce the asynchronous operation time;

[0081] Determine the value of the control target unit at the future time in real time, including:

[0082] When the control target units are the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis, the adjustment coefficients of the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis at the future time sequence are determined using the actual speeds of the first belt, the second belt, the third belt, the first open roller, and the second open roller at the future time sequence and the speeds of the virtual servo axis;

[0083] When the control target unit is a virtual servo axis, the speed of the virtual servo axis at the future time sequence is determined by using the deviation between the actual speed of the first belt, the second belt, the third belt, the first open roller and the second open roller at the future time sequence and the expected speed of the first belt, the second belt, the third belt, the first open roller and the second open roller at the future time sequence.

[0084] Step S4: respectively construct mapping relationships between the virtual servo axis and the working condition timing, and between the adjustment coefficient of each physical servo axis and the working condition timing, so as to maintain the accuracy of the calculation of the control target unit while reducing the complexity of the calculation of the control target unit.

[0085] Construct the mapping relationship between the virtual servo axis and the working condition timing, and the adjustment coefficient of each physical servo axis and the working condition timing, including:

[0086] The speed of the virtual servo axis at each working condition time sequence is obtained and arranged in time sequence to obtain a virtual servo axis speed sequence. The LSTM neural network is used to perform network training based on the virtual servo axis speed sequence to obtain a virtual servo axis speed prediction model that characterizes the mapping relationship between the virtual servo axis and the working condition time sequence.

[0087] The speed of each physical servo axis at each working condition timing is obtained and arranged in time series to construct a physical servo axis adjustment coefficient sequence. The LSTM neural network is used to perform network training based on the physical servo axis adjustment coefficient sequence to obtain a prediction model for each physical servo axis adjustment coefficient that characterizes the mapping relationship between the adjustment coefficient of each physical servo axis and the working condition timing.

[0088] A virtual servo axis speed prediction model and a physical servo axis adjustment coefficient prediction model are constructed to characterize the mapping relationship between the virtual servo axis and the working condition timing, and the adjustment coefficient of each physical servo axis and the working condition timing. According to the working condition timing, the speed and position information of the virtual servo axis and the adjustment coefficient of each physical servo axis can be obtained to maintain the measurement accuracy of the control target unit while reducing the measurement complexity of the control target unit.

[0089] The present invention performs time-series prediction on the actual main line speed of the gypsum board to obtain the actual main line speed of the gypsum board at the future time sequence, compares the actual main line speed of the gypsum board at the future time sequence with the expected main line speed of the gypsum board to determine the control target unit, and determines the value of the control target unit at the future time sequence in real time to achieve advanced control of the main line synchronization model to achieve a reduction in the asynchronous operation time, and constructs the mapping relationship between the virtual servo axis and the working condition time sequence, and the adjustment coefficient of each physical servo axis and the working condition time sequence, respectively, to maintain the measurement accuracy of the control target unit while reducing the measurement complexity of the control target unit.

[0090] The above embodiments are merely exemplary embodiments of the present application and are not intended to limit the scope of the present application. The scope of protection of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and scope of protection of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present application.

Claims

1. A synchronous control method for gypsum board production based on a mainline control model, characterized by: The following steps are involved: Step S1: Setting a virtual servo axis in a main line synchronization model of a gypsum board production line, and setting an adjustment coefficient for each physical servo axis in the main line synchronization model based on the virtual servo axis and the expected main line speed of the gypsum board. The main line synchronization model is a control model for the synchronous operation of the main line components of the production line in a master-multiple slave servo control mode composed of multiple physical servo axes; Step S2: monitoring the actual main line speed of the gypsum board controlled by the physical servo axis, performing a time series prediction on the actual main line speed of the gypsum board to obtain the actual main line speed of the gypsum board at a future time series, and comparing the actual main line speed of the gypsum board at the future time series with the expected main line speed of the gypsum board to determine a control target unit, wherein the control target unit includes the adjustment coefficients of the virtual servo axis and each physical servo axis; Step S3: determining in real time the value of the control target unit at a future time sequence, so as to achieve advanced control of the main line synchronization model and reduce the asynchronous operation time; Step S4: constructing mapping relationships between the virtual servo axis and the working condition sequence, and between the adjustment coefficient of each physical servo axis and the working condition sequence, respectively, so as to maintain the accuracy of the calculation of the control target unit while reducing the complexity of the calculation of the control target unit; The actual main line speed of the gypsum board is predicted in time series to obtain the actual main line speed of the gypsum board at the future time series, including: The actual speed prediction models of the first belt, the second belt, the third belt, the first open roller and the second open roller are obtained by using an LSTM neural network to perform network training based on the actual speed time series of the first belt, the second belt, the third belt, the first open roller and the second open roller; The actual speed prediction model of the first belt, the second belt, the third belt, the first open roller and the second open roller is used to predict the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller at a future time sequence; Comparing the actual main line speed of the gypsum board at the future time sequence with the expected main line speed of the gypsum board to determine the control target unit, including: Compare the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller at the future time sequence with the expected speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, respectively, wherein, If the actual speed of the first belt is inconsistent with the expected speed of the first belt, the control target unit at the future timing is the first physical servo axis; If the actual speed of the second belt is inconsistent with the expected speed of the second belt, the control target unit at the future timing is the second physical servo axis; If the actual speed of the third belt is inconsistent with the expected speed of the third belt, the control target unit at the future timing is the third physical servo axis; If the actual speed of the first open roller conveyor is inconsistent with the desired speed of the first open roller conveyor, the target unit to be controlled at the future timing is the fourth physical servo axis; If the actual speed of the second open roller table is inconsistent with the expected speed of the second open roller table, the control target unit at the future timing is the fifth physical servo axis; If the actual speeds of the first belt, the second belt, the third belt, the first open roller conveyor, and the second open roller conveyor are inconsistent with the desired speeds of the first belt, the second belt, the third belt, the first open roller conveyor, and the second open roller conveyor, then the control target unit at the future time sequence is the virtual servo axis; Determine the value of the control target unit at the future time in real time, including: When the control target units are the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis, the adjustment coefficients of the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis at the future time sequence are determined using the actual speeds of the first belt, the second belt, the third belt, the first open roller, and the second open roller at the future time sequence and the speeds of the virtual servo axis; When the control target unit is a virtual servo axis, the speed of the virtual servo axis at the future time sequence is determined by using the deviation between the actual speeds of the first belt, the second belt, the third belt, the first open roller conveyor, and the second open roller conveyor at the future time sequence and the expected speeds of the first belt, the second belt, the third belt, the first open roller conveyor, and the second open roller conveyor at the future time sequence; The physical servo axis includes a first physical servo axis, a second physical servo axis, a third physical servo axis, a fourth physical servo axis and a fifth physical servo axis. The main line components of the production line correspond to the physical servo axes one by one. The main line components of the production line include a first belt controlled by the first physical servo axis, a second belt controlled by the second physical servo axis, a third belt controlled by the third physical servo axis, a first open roller controlled by the fourth physical servo axis and a second open roller controlled by the fifth physical servo axis. The actual main line speed of the gypsum board includes the actual speed of the main line components of the production line, and the desired main line speed of the gypsum board includes the desired speed of the main line components of the production line.

2. The method for synchronous control of gypsum board production based on the main line control model according to claim 1, characterized in that: The step of setting an adjustment coefficient for each physical servo axis in the main line synchronization model based on the virtual servo axis and the expected main line speed of the gypsum board includes: Setting multiple selected values for the desired main line speed of the gypsum board, and sequentially setting the speed of the virtual servo axis to the multiple selected values; Trial production was carried out at each speed of the virtual servo axis. During each trial production process, the adjustment coefficient of each physical servo axis was measured based on the actual main line speed of the gypsum board and the speed of the virtual servo axis. Based on the speed of the virtual servo axis and the adjustment coefficient of each physical servo axis during each trial production process, an adjustment coefficient setting function was constructed to implement the setting of the adjustment coefficient in the main line synchronization model. The function expression of the adjustment coefficient setting function is: (A,B,C,D,E)=BP(V); Where A, B, C, D, and E are the adjustment coefficients of the first physical servo axis, the second physical servo axis, the third physical servo axis, the fourth physical servo axis, and the fifth physical servo axis, respectively; V is the speed of the virtual servo axis; and BP is the function body of the adjustment coefficient setting function.

3. The method for synchronous control of gypsum board production based on the main line control model according to claim 2, characterized in that: The adjustment coefficient setting function is obtained by training a BP neural network using the speed of the virtual servo axis and the adjustment coefficients of each physical servo axis in each trial production process as data samples.

4. The method for synchronous control of gypsum board production based on a mainline control model according to claim 3, characterized in that: The monitoring of the actual main line speed of the gypsum board controlled by the physical servo axis includes: Real-time monitoring of the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller, and using the monitoring sequence as a sequence attribute of the actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller; The actual speeds of the first belt, the second belt, the third belt, the first open roller and the second open roller are arranged according to the monitoring time sequence to obtain the actual speed time series of the first belt, the second belt, the third belt, the first open roller and the second open roller.

5. The method for synchronous control of gypsum board production based on the main line control model according to claim 1, characterized in that: The mapping relationships between the virtual servo axis and the working condition timing, and the adjustment coefficients of each physical servo axis and the working condition timing are respectively constructed, including: The speed of the virtual servo axis at each working condition time sequence is obtained and arranged in time sequence to obtain a virtual servo axis speed sequence. The LSTM neural network is used to perform network training based on the virtual servo axis speed sequence to obtain a virtual servo axis speed prediction model that characterizes the mapping relationship between the virtual servo axis and the working condition time sequence. The speed of each physical servo axis at each working condition timing is obtained and arranged in time series to construct a physical servo axis adjustment coefficient sequence. The LSTM neural network is used to perform network training based on the physical servo axis adjustment coefficient sequence to obtain a prediction model for each physical servo axis adjustment coefficient that characterizes the mapping relationship between the adjustment coefficient of each physical servo axis and the working condition timing.

Images