A method and system for controlling the wrap angle of a roller coating machine

Abstract

This invention discloses a wrap angle control method for a roll coating machine. The method inputs the measured tension of the strip during production and the user-set target coating thickness into a trained neural network model. Then, the target wrap angle between the coating roller and the strip, output by the neural network model, is transmitted to the roll coating machine. The roll coating machine adjusts the wrap angle between the coating roller and the strip according to the target wrap angle. This control method can adjust the wrap angle between the coating roller and the strip based on fluctuating tension, thereby improving the consistency of the coating thickness.

Description

Technical Field

[0001] This invention relates to the field of cold rolling coating, and more specifically to a method and system for controlling the wrap angle of a roll coating machine. Background Technology

[0002] On cold-rolled strip steel production lines, to ensure the strip steel's corrosion and contamination resistance and extend its service life, surface coating is required to form a corrosion- and contamination-resistant film on both sides of the strip steel. The main factors affecting the strip steel's performance and use are coating thickness and uniformity. Currently, strip steel production lines use roller coaters to coat the strip steel, with a pair of rollers applying coating to both sides of the strip. Factors affecting coating uniformity include roller cylindricity and installation position deviations, which need to be controlled during design and manufacturing. Factors affecting coating thickness include material viscosity, the speed ratio between the rollers and the strip steel, and the wrap angle and pressure between the rollers and the strip steel. However, due to tension fluctuations in the strip steel, especially during coil changing and welding, and the tension difference during uncoiling and coiling, the pressure between the rollers and the strip steel also changes, affecting the consistency of the coating thickness.

[0003] Currently, operators mainly adjust the wrap angle based on their experience by observing the coating condition, without taking into account the impact of tension fluctuations. Furthermore, the effects of wrap angle and pressure on coating thickness are not linear but interrelated, which can lead to different average thicknesses for different steel coils and different coating thicknesses for different sections of the same steel coil. Summary of the Invention

[0004] The purpose of this invention is to provide a wrap angle control method for a roller coating machine, which adjusts the wrap angle between the coating roller and the strip when the tension of the strip fluctuates, thereby reducing the change in coating thickness caused by tension fluctuations and improving the consistency of coating thickness.

[0005] To solve the above-mentioned technical problems, the present invention provides a technical solution: a wrap angle control method for a roll coating machine, wherein the measured tension of the strip steel during production and the target coating thickness set by the user are input into a trained neural network model, and then the target wrap angle between the coating roller and the strip steel output by the neural network model is transmitted to the roll coating machine; the roll coating machine adjusts the wrap angle between the coating roller and the strip steel according to the target wrap angle.

[0006] According to the above scheme, the input layer of the neural network model is tension and coating thickness, and the output layer is wrap angle; the training data of the neural network model includes measured tension, measured coating thickness, and actual wrap angle.

[0007] According to the above scheme, the measured tension is obtained by a tension meter, and the measured coating thickness is obtained by a film thickness gauge. When the film thickness gauge measures the coating thickness, the measurement position of the film thickness gauge is recorded at the same time. The time taken for the strip steel to be transported to that position is calculated based on the measurement position of the film thickness gauge, and then the measured tension and the measured coating thickness at the same position are synchronously correlated.

[0008] According to the above scheme, when the strip steel is uncoiled, rolled, or welded, the coating thickness is measured by a film thickness gauge to obtain the actual coating thickness.

[0009] According to the above scheme, when the film thickness gauge measures the coating thickness, at the same position along the length of the strip, several points dispersed along the width are taken to measure the coating thickness and calculate the average value. The average value obtained is the coating thickness at that length position.

[0010] According to the above scheme, the measured tension, measured coating thickness, measurement position of the film thickness gauge, and actual wrap angle are read by the PLC and packaged as training data for the neural network model.

[0011] According to the above scheme, the packaged data is synchronized to the OPC service via the OPC protocol so that it can be subsequently input into the neural network model.

[0012] According to the above scheme, the training data should include at least ten measured coating thicknesses at each of the ten measurement positions when the strip is uncoiled and when it is coiled.

[0013] According to the above scheme, when the strip is uncoiled or coiled, points are taken at intervals of 50 to 150m along the length of the strip as measurement positions for the coating thickness, so as to obtain the measured coating thickness.

[0014] A wrap angle control system for a roller coater used to implement the wrap angle control method of the roller coater described above, comprising:

[0015] The strip tension measurement module is used to obtain the tension of the strip.

[0016] The target coating thickness input module allows users to input the target coating thickness.

[0017] The server is equipped with a neural network model, which is used to obtain the target wrap angle based on the measured tension and the target coating thickness.

[0018] The beneficial effects of this invention are: compared with the prior art, it eliminates the need for operators to observe the coating condition and adjust the wrap angle between the roller and the strip, reducing manual working time and labor costs; moreover, this invention reduces the impact of tension fluctuations on coating thickness. Under different tensions, it uses a neural network model to calculate the wrap angle required for the target coating thickness, comprehensively considering the impact of changes in contact area and pressure on the film thickness during wrap angle adjustment, thereby improving the consistency of coating thickness. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a roller coating machine wrap angle control method according to an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the described embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0021] Before implementing the wrap angle control method for the roller coating machine described in this invention, it is first necessary to train the neural network model in the server. The neural network model is built using the Keras library, with tension and coating thickness as the input layers and wrap angle as the output layer.

[0022] The training parameters for the neural network model are obtained as follows: The measured tension is obtained by a tension meter, read by the PLC, and recorded in the DB block. The measured coating thickness of the strip during uncoiling, coiling, and welding is obtained by a film thickness gauge, and the corresponding positions are recorded. The measured tension, measured coating thickness, and actual wrap angle between the coating roller and the strip are sent to the server (in this embodiment, the OPC protocol is used to synchronize the data to the OPC service, and the program on the server periodically retrieves the data; this is written in Python).

[0023] Furthermore, coating thickness measurements were preferentially taken at points of tension fluctuation, namely during strip uncoiling, coiling, and welding. To provide sufficient training data, coating thickness was measured at multiple locations during uncoiling and coiling, with points taken at 100-meter intervals, resulting in at least ten points each during uncoiling and coiling. Simultaneously, to ensure measurement accuracy, five points dispersed along the width direction were taken at the same location along the strip's length to calculate the average coating thickness value, which was then used as the coating thickness at that length location.

[0024] The training data is imported into the neural network model to train the node parameters, resulting in a usable neural network model. Since the contact area and pressure between the roller and the strip change simultaneously during the wrap angle adjustment, and the relationship between the wrap angle, tension, and coating thickness is non-linear, this embodiment preferentially selects a neural network to construct a model of the relationship between these three factors. Other models may be used in other embodiments of this invention.

[0025] In actual production, the measured tension of the strip coating is obtained from the OPC service at regular intervals. The measured tension and the target coating thickness set by the user (through the user interface) are input into the neural network model. The neural network outputs the target wrap angle, and then the server transmits the target wrap angle to the roller coater. The roller coater controls the action of the coating roller according to the target wrap angle so that the coating roller and the strip achieve the target wrap angle.

[0026] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for controlling the wrap angle of a roller coating machine, characterized in that: The measured tension of the strip steel during production and the target coating thickness set by the user are input into the trained neural network model. Then, the target wrap angle between the coating roller and the strip steel output by the neural network model is transmitted to the roll coating machine. The roll coating machine adjusts the wrap angle between the coating roller and the strip steel according to the target wrap angle.

2. The wrap angle control method for a roller coating machine according to claim 1, characterized in that: The input layer of the neural network model consists of tension and coating thickness, and the output layer consists of wrap angle. The training data of the neural network model includes measured tension, measured coating thickness, and actual wrap angle.

3. The wrap angle control method for a roller coating machine according to claim 2, characterized in that: The measured tension is obtained by a tension meter, and the measured coating thickness is obtained by a film thickness gauge. When the film thickness gauge measures the coating thickness, the measurement position of the film thickness gauge is recorded at the same time. The time taken for the strip steel to be transported to that position is calculated based on the measurement position of the film thickness gauge, and then the measured tension and the measured coating thickness at the same position are synchronously correlated.

4. The wrap angle control method for a roller coating machine according to claim 3, characterized in that: When the strip steel is uncoiled, rolled, or welded, the coating thickness gauge is used to measure the coating thickness to obtain the actual coating thickness.

5. The wrap angle control method for a roller coating machine according to claim 3, characterized in that: When a film thickness gauge measures the coating thickness, several points are taken at the same position along the length of the strip, and the coating thickness is measured at a distance along the width. The average value is then calculated, and the average value is the coating thickness at that length position.

6. The wrap angle control method for a roller coating machine according to claim 2, characterized in that: The measured tension, measured coating thickness, measurement position of the film thickness gauge, and actual wrap angle are read by the PLC and packaged as training data for the neural network model.

7. The wrap angle control method for a roller coating machine according to claim 6, characterized in that: The packaged data is synchronized to the OPC service via the OPC protocol for subsequent input into the neural network model.

8. The wrap angle control method for a roller coating machine according to claim 4, characterized in that: The training data should include at least ten measured coating thicknesses at each of the ten measurement locations during strip uncoiling and coiling.

9. The wrap angle control method for a roller coating machine according to claim 8, characterized in that: When the strip is uncoiling or coiling, points are taken at intervals of 50 to 150 m along the length of the strip as measurement positions for the coating thickness to obtain the actual coating thickness.

10. A wrap angle control system for implementing the wrap angle control method of the roll coating machine according to any one of claims 1-9, characterized in that: include; The strip tension measurement module is used to obtain the tension of the strip. The target coating thickness input module allows users to input the target coating thickness. The server is equipped with a neural network model, which is used to obtain the target wrap angle based on the measured tension and the target coating thickness.

Images