TY - GEN
T1 - Development of Autonomous Control System using Self-Organizing Map and Autoregressive Self-Organizing Map
AU - Ayu, Aqila Dzikra
AU - Matthew, Hansel
AU - Suherman, Iman Herlambang
AU - Subiantoro, Aries
AU - Kusumoputro, Benyamin
N1 - Funding Information:
ACKNOWLEDGMENT This research is funded by the Ministry of Education, Culture, Research and Technology through PTUPT with contract number NKB-1029/UN2.RST/HKP.05.00/2022.
Publisher Copyright:
© 2022 Institute of Advanced Engineering and Science (IAES).
PY - 2022
Y1 - 2022
N2 - Autonomous control system is a controller that can make its own decision when performing control functions. An autonomous control system is needed because of the increasing complexity of dynamic system and other controller requirements. This research uses the pressure process rig® because of its dynamic and non-linear properties to develop an autonomous control system. The pressure process rig is equipment to demonstrate pressure measurement and control in the process industry. However, because of its non-linear characteristics, the system will have different responses under different operating conditions. The non-linear system cannot utilize conventional control methods such as the proportional-integral-derivative (PID). This research implements Self-Organizing Map (SOM) and Autoregressive Self-Organizing Map (ARSOM)-based control to overcome the nonlinearity problem. In addition, variations in the number of network parameters, such as the number of mapping neurons, alpha, and beta, are carried out to determine the effect of each parameter. The network performance is measured by the MSE training value, the MSE testing value, the number of epochs used, and the training duration. The results showed that the SOM and ARSOM models resulted in low MSE training and testing with fast training times. In addition, the performance of the ARSOM model is better than the SOM model with fewer mapping neurons. The MSE value for training and testing of the ARSOM model is better than the SOM model with fewer mapping neurons.
AB - Autonomous control system is a controller that can make its own decision when performing control functions. An autonomous control system is needed because of the increasing complexity of dynamic system and other controller requirements. This research uses the pressure process rig® because of its dynamic and non-linear properties to develop an autonomous control system. The pressure process rig is equipment to demonstrate pressure measurement and control in the process industry. However, because of its non-linear characteristics, the system will have different responses under different operating conditions. The non-linear system cannot utilize conventional control methods such as the proportional-integral-derivative (PID). This research implements Self-Organizing Map (SOM) and Autoregressive Self-Organizing Map (ARSOM)-based control to overcome the nonlinearity problem. In addition, variations in the number of network parameters, such as the number of mapping neurons, alpha, and beta, are carried out to determine the effect of each parameter. The network performance is measured by the MSE training value, the MSE testing value, the number of epochs used, and the training duration. The results showed that the SOM and ARSOM models resulted in low MSE training and testing with fast training times. In addition, the performance of the ARSOM model is better than the SOM model with fewer mapping neurons. The MSE value for training and testing of the ARSOM model is better than the SOM model with fewer mapping neurons.
KW - Artificial Neural Networks
KW - Autonomous Control System
KW - Autoregressive Self-Organizing Map
KW - Self-Organizing Map
UR - http://www.scopus.com/inward/record.url?scp=85142713540&partnerID=8YFLogxK
U2 - 10.23919/EECSI56542.2022.9946634
DO - 10.23919/EECSI56542.2022.9946634
M3 - Conference contribution
AN - SCOPUS:85142713540
T3 - International Conference on Electrical Engineering, Computer Science and Informatics (EECSI)
SP - 407
EP - 413
BT - Proceedings - 9th International Conference on Electrical Engineering, Computer Science and Informatics, EECSI 2022
A2 - Facta, Mochammad
A2 - Syafrullah, Mohammad
A2 - Riyadi, Munawar Agus
A2 - Subroto, Imam Much Ibnu
A2 - Irawan, Irawan
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 9th International Conference on Electrical Engineering, Computer Science and Informatics, EECSI 2022
Y2 - 6 October 2022 through 7 October 2022
ER -