Ai spray chamber, ai spraying production line, and ai airflow control method

Abstract

A spray chamber, spraying production line, and airflow control method with artificial intelligence airflow control are disclosed for spraying a substrate. The chamber includes: a cavity; a nozzle disposed within the internal space for spraying paint onto the substrate; an airflow guiding device that provides and guides two opposing airflows within the internal space to form a downward laminar airflow, allowing the paint leaving the nozzle to move stably downwards; and an intelligent control module comprising a sensing unit and a computing unit. The computing unit stores a differential pressure control decision model, receives various parameters, and inputs them into the differential pressure control decision model to automatically adjust the control parameters (Kp, Ki, Kd), negative pressure ratio (α), and gap leakage coefficient (kleak) of the PID controller, thereby controlling the rotational speeds of the air supply unit and the air extraction unit.

Claims

1. A spray chamber equipped with artificial intelligence airflow control for spraying a substrate, comprising: A cavity having an internal space, an inlet on one side of the cavity, and an outlet on the other side of the cavity opposite to the inlet, wherein the substrate enters the internal space of the cavity through the inlet and exits the cavity through the outlet; and a spray head disposed in the internal space for spraying a coating onto the substrate. An airflow guiding device provides and guides two opposing airflows within an internal space to form a downward laminar airflow, causing the coating material leaving the nozzle to move stably downwards. The airflow guiding device includes an air supply unit, an air extraction unit, two air supply ports, and an air extraction port. The two air supply ports are positioned opposite each other on both sides of the nozzle to provide the two opposing airflows. The air extraction port is located at the bottom of the internal space, between the two air supply ports, to extract gas from the internal space, thereby guiding the two opposing airflows to form the laminar airflow. An intelligent control module is electrically connected to the air supply unit and the air extraction unit. The air supply unit controls the intake airflow of the two air supply ports, and the air extraction unit controls the exhaust airflow of the air extraction port. The intelligent control module includes a sensing unit and a computing unit. The computing unit stores a differential pressure control decision model. The computing unit receives at least one of the following parameters: door opening status, spraying amount, volatile organic compound generation amount, outdoor wind pressure fluctuation value, temperature change value, humidity change value, and production mode, and inputs them into the differential pressure control decision model to automatically adjust the three control parameters (Kp, Ki, Kd), a negative pressure ratio (α), and a gap leakage coefficient () of the PID controller, and controls the speed of the intake fan of the air supply unit and the speed of the exhaust fan of the air extraction unit accordingly.

2. The spray chamber with artificial intelligence airflow control as described in claim 1, wherein the height of the lower edge of the two air supply ports is not lower than the height of the lower end of the nozzle.

3. The spray chamber with artificial intelligence airflow control as described in claim 1, wherein the airflow guiding device draws air through the air intake port at a rate greater than the air supply to the two air supply ports.

4. The spray chamber with artificial intelligence airflow control as described in claim 1, wherein the laminar airflow velocity is greater than or equal to 0.1 m / s and less than or equal to 0.3 m / s.

5. The spray chamber with artificial intelligence airflow control as described in claim 1, wherein the sensing unit includes a differential pressure sensor for detecting a measured pressure difference between the internal space and an external environment.

6. The nozzle with artificial intelligence airflow control as described in claim 1, wherein the differential pressure control decision model is based on minimizing a loss function: ; where, is the measured pressure difference, is the target pressure difference, is the intake air volume, is the exhaust air volume, is the change in control quantity, and is the weighting coefficient.

7. The spray chamber with artificial intelligence airflow control as described in claim 1, wherein the computing unit performs calculations under the constraint of a mass conservation equation, which is: ; where, This refers to the leakage air volume.

8. The spray chamber with artificial intelligence airflow control as described in claim 1, wherein when the sensing unit detects a change in the door opening state or a switch in the production mode, the computing unit dynamically updates the optimal parameter combination of the three control parameters (Kp, Ki, Kd) of the PID controller, the negative pressure ratio (α), and the gap leakage coefficient ().

9. A spraying production line for spraying a substrate, comprising: A spray chamber with artificial intelligence airflow control, as described in any of the requests 1 to 8; A first conveying device for conveying the substrate so that the substrate enters the internal space through the inlet; and a second conveying device for moving the substrate away from the spray chamber through the outlet.

10. An airflow control method applicable to the spraying production line as described in claim 9, the airflow control method comprising the following steps: providing two opposing airflows using the two air supply ports of the airflow guiding device, and drawing air from the bottom of the internal space using the air extraction port to form a downward laminar airflow; detecting a measured pressure difference in the internal space using the sensing unit; receiving at least one parameter of door opening status, spraying volume, volatile organic compound generation, outdoor wind pressure fluctuation value, temperature change value, humidity change value, and production mode using the computing unit; and executing the pressure difference control decision model using the computing unit, and automatically adjusting the speed of the intake fan of the air supply unit and the speed of the exhaust fan of the air extraction unit through reinforcement learning or deep learning algorithms; wherein, The adjustment aims to minimize a loss function, causing the measured pressure difference to converge to a target pressure difference, and to achieve a stable balance between the intake and exhaust airflow in the internal space.

Images