CYBER PHYSICAL CONTROL SYSTEM OF THREE INTERACTING-TANK

• Main Author: Sukma Pratiwi, Citrawati
• Format: Final Project
• Language: Indonesia
• Subjects: Karya Umum
• Online Access: https://digilib.itb.ac.id/gdl/view/39433
• Institution: Institut Teknologi Bandung

In the era of Industry 4.0, the industrial production is driven by global competition and it needs a rapid production adaptation to fulfill the changing market demand. A decentralized control system and an ease of connectivity in the exchange of information is needed to identify, to monitor and to optimize production processes in real-time so that production adaptations can be carried out quickly. The information is then stored in cloud storage to monitor the production process. The implementation of Cyber-Physical System (CPS) is one of the right solutions in optimizing industrial process control systems automatically. In this final project, an implementation of CPS for the control system of fluid level in the third tank of three interacting-tank system is done. The control process starts with optimizing the model of the open-loop system to be identical with the dynamic response of the real plant. Then PID tuning is done to control the dynamic response of system to obtain the expected set point. When system in the steady state condition, it will be given a disturbance in the form of the opening change of manual valves between the three tanks randomly. This makes the former system model and the PID controller no longer optimal to be implemented. For this reason, the CPS is then be implemented with MATLAB as the cyber system and is connected to the physical system through OPC UA in real-time and automatically to update the suitable system model and value of PID controller. So the optimal system dynamics responses for each set point can be obtained. The Industrial Internet of Things System is built for monitoring and controlling set points remotely via the Ubidot IIoT platform with the MQTT protocol. Modeling of the three interacting-tank system is done by approaching first-order plus dead time system model equation because the obtained ISE from real system response is the smallest. The three interacting-tank cyber physical control system can run real-time to monitor the performances of the risetime, maximum overshoot and RMSE response systems. Furthermore, CPS also can generate new model and control parameters after being given disturbance automatically. System data monitoring and storage is built online and set point settings can be done via website with a minimum average data latency of 892,587 ms for the QoS 0 value.