Shibaura Mechatronics Used Model-Based Design to Develop a Temperature Controller for Organic Film Deposition

“Using Model-Based Design, we were able to immediately implement model predictive control into the PLC through automatic code generation.”

Key Outcomes

  • Utilizing simulation reduced control parameter adjustment time by half a day to 10 seconds
  • Using Model Predictive Control Toolbox reduced the number of control parameters from seven PID controllers to one MPC controller
  • Enabling auto ST code generation of a model predictive control algorithm with Simulink PLC Coder allowed deployment to the PLC
Video length is 26:25

Shibaura Mechatronics Corporation is a manufacturer of semiconductor/FPD manufacturing equipment and vacuum application devices. During the development of a temperature controller for organic film deposition equipment, several challenges were encountered. These included the need to reduce commissioning time, simplify control parameter rules, and perform real-time verification without relying on the actual equipment.

In terms of control performance, challenges included overshoot suppression, the influence of heat transfer characteristics, and complex multi-input multi-output control. To address these difficulties, the team at Shibaura Mechatronics implemented model predictive control (MPC) and Model-Based Design methodologies, which proved effective in saving time and labor during development.

To expedite the development process and enable simulation-based development without relying on physical equipment, the company adopted grey-box modeling to estimate unknown parameters of the plant model from experimental data. The development of the MPC algorithm utilized Model Predictive Control Toolbox™, while Simulink Control Design™ was employed to linearize the nonlinear model and use it as a predictive model for MPC.

Additionally, Simulink PLC Coder™ was used to automatically generate ST code, enabling the team to rapidly implement the advanced MPC controller on the PLC.