Implementasi PID-Genetic Algorithm Untuk Pengaturan Cascade Level-Pressure Pada Tangki PCT-100 Berbasis Programmable Logic Controller

Nabila, Haura (2025) Implementasi PID-Genetic Algorithm Untuk Pengaturan Cascade Level-Pressure Pada Tangki PCT-100 Berbasis Programmable Logic Controller. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022211153-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Sistem kendali tangka PCT-100 memiliki karakteristik nonlinier akibat perubahan volume cairan, yang berpengaruh langsung pada tekanan, sehingga pemilihan parameter PID secara manual sering menimbulkan overshoot, ketidakstabilan, dan waktu respon yang panjang. Penelitian ini bertujuan untuk mengimplementasikan dan mengevaluasi kontroler PID pada plant PCT-100 menggunakan parameter dari metode optimisasi Genetic Algorithm agar respon yang didapatkan mencapai setpoint dan dapat menjaga performa saat ada gangguan dari luar, serta memperkecil waktu transien dengan respon dari auto-tuning sebagai acuannya. Hasil pengujian menunjukkan bahwa loop tunggal level dengan GA menghasilkan RMSE 3,21 sedangkan loop tunggal pressure GA sebesar 3,2518. Pada konfigurasi cascade, varian level-pressure dengan GA dapat menyerupai metode auto-tuning dengan perbedaan 0,0699 mempunyai RMSE 2,9688. Sementara itu, cascade pressure-level menampilkan RMSE lebih tinggi 3,0059–3,7789 dan respons lebih lambat. Dari analisis, dapat disimpulkan bahwa cascade level-pressure dipadu dengan GA dapat menyerupai respon kontroler dengan auto-tuning dari PID Compact. Memberikan keseimbangan optimal antara kecepatan respon dan akurasi. Penentuan konfigurasi loop berpengaruh besar terhadap dinamika sistem. Menempatkan variabel yang perubahannya lebih cepat (tekanan) pada inner loop mempercepat respon dan memperkecil nilai error hingga 2,9688 (pada GA) dan 2,8989 (pada auto-tuning). ========================================================================================================================
The PCT-100 tank control system exhibits nonlinear behavior due to changes in liquid volume, which directly affect pressure; consequently, manual PID-parameter selection often results in overshoot, instability, and long response times. This study implements and evaluates a PID controller on the PCT-100 plant using parameters optimized by a Genetic Algorithm (GA) so that the response reaches the setpoint, maintains performance under external disturbances, and reduces transient time by referencing the auto-tuning response. Test results show that the single-loop level control with GA achieves an RMSE of 3.21, whereas the single-loop pressure control with GA yields an RMSE of 3.2518. In the cascade configuration, the level-pressure variant tuned by GA closely matches the auto-tuning method—differing by only 0.0699—with an RMSE of 2.9688. Meanwhile, the pressure-level cascade exhibits higher RMSE values (3.0059–3.7789) and a slower response. From this analysis, it can be concluded that the GA-tuned level-pressure cascade can replicate the response of the Compact PID auto-tuning controller, offering an optimal balance between speed and accuracy. Loop configuration significantly influences system dynamics: placing the faster-changing variable (pressure) in the inner loop accelerates response and reduces error to as low as 2.9688 (GA) and 2.8989 (auto-tuning).

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Genetic Algorithm, Cascade Control, PLC Siemens S7-1500, RMSE. Genetic Algorithm, Cascade Control, PLC Siemens S7-1500, RMSE.
Subjects:	T Technology > TC Hydraulic engineering. Ocean engineering > TC424 Water levels T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Haura Nabila
Date Deposited:	25 Jul 2025 07:14
Last Modified:	25 Jul 2025 07:14
URI:	http://repository.its.ac.id/id/eprint/121610

Actions (login required)

View Item