PID ROBUST Hinf INTEGRAL-BACKSTEPPING CONTROL WITH OPTIMAL CONTROL STRATEGY FOR AUTOMATED PEOPLE MOVERS (APM)

• Main Author: Elbert Suryana, Lucas
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/36478
• Institution: Institut Teknologi Bandung

Nowadays, train signaling system technology in Indonesia is experiencing rapid development with the construction of rail-based public transport infrastructure for urban areas, known as urban transit. This signaling system technology is known as Communication-Based Train Control (CBTC). There is three urban transit construction in Indonesia. They are Light Rail Transit (LRT) which connects Jakarta-Bogor-Depok-Bekasi, Mass Rapid Transit (MRT) in the middle of Jakarta, and Automated People Movers (APM) which connects terminals 1,2 and 3 at Soekarno-Hatta International Airport, Tangerang. By implementing CBTC signaling technology, the train can be operated automatically without a driver. This was made possible by the Automatic Train Operation (ATO) system in CBTC. This automatic mode operation can be done by providing a speed profile to ATO which serves as a reference value that must be followed by the train. The APM Soekarno-Hatta International Airport is planned to be operated automatically. Speed profiles are formulated for every trip between stations by considering the travel time and speed of the train. Problems in determining the optimal speed profile will be solved using one of the methods in dynamic optimization, namely Interior Point Method (IPM). Where the speed limit, acceleration limit, and distance that must be taken become the limitation of the system. The optimal speed profile design is designed using minimal time objective functions. After a speed profile is formed, it will then be used as a reference from the control system. The control system is designed using the Proportional-Integral-Derivative (PID) Robust H? Integral-Backstepping Control. PID Robust Hinf Integral-Backstepping control will be represented in the state feedback augmented to the state of the state space of the train system. Robust controllers are chosen because, in operation, there is uncertainty in the form of problems within the train system, especially at peak hours. The simulation results show the value of the Root Mean Square Error (RMSE) and the difference in the final position of the train with references in a row are 1,14 x 10^-7 dan 2,86 x 10^-4 m.