Data-driven PID tuning for liquid slosh-free motion using memory-based SPSA algorithm

• Main Authors: Mohd Ashraf, Ahmad, Nik Mohd Zaitul, Akmal Mustapha, Mohd Zaidi, Mohd Tumari, Mohd Helmi, Suid, Raja Mohd Taufika, Raja Ismail
• Format: Book Section
• Language: English; English
• Published: Springer Singapore 2018
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://umpir.ump.edu.my/id/eprint/22979/13/37.%20Data-driven%20PID%20tuning%20for%20liquid%20slosh-free%20motion%20using%20memory-based%20SPSA%20algorithm.pdf; http://umpir.ump.edu.my/id/eprint/22979/14/37.1%20Data-driven%20PID%20tuning%20for%20liquid%20slosh-free%20motion%20using%20memory-based%20SPSA%20algorithm.pdf; http://umpir.ump.edu.my/id/eprint/22979/; https://link.springer.com/chapter/10.1007/978-981-13-3708-6_17

This study proposes a data-driven PID tuning for liquid slosh suppression based on enhanced stochastic approximation. In particular, a new version of Simultaneous Perturbation Stochastic Approximation (SPSA) based on memory type function is introduced. Tis memory-based SPSA (M-SPSA) algorithm has a capability to obtain better optimization accuracy than the conventional SPSA, since it is able to keep the best design parameter during the tuning process. The effectiveness of this algorithm is tested to data-drive PID tuning for liquid slosh problem. The achievement of the M-SPSA based algorithm is assessed in terms of trajectory tracking of trolley position, slosh angle reduction and also computation time. The outcome of this study shows that the PID-tuned M-SPSA is able to provide better control performance accuracy than the other variant of SPSA based method.