Optimisasi Parameter PI Adaptif Pada Brine Blowdown System Berbasis Particle Swarm Optimization Di Desalination Plant PLTGU Priok

Razzaq, Rifqi Abid (2025) Optimisasi Parameter PI Adaptif Pada Brine Blowdown System Berbasis Particle Swarm Optimization Di Desalination Plant PLTGU Priok. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5009211120-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB)

Abstract

Penelitian ini mengembangkan dan menguji kontroler PI adaptif berbasis Particle Swarm Optimization (PSO) untuk menjaga brine level pada brine blowdown system di desalination plant Pembangkit Listrik Tenaga Gas dan Uap (PLTGU) agar stabil pada set‐point 50 %. Proses dimulai dengan identifikasi model Nonlinear ARX (NARX) berbasis data operasi aktual, yang mencapai akurasi model hingga 90.04 % setelah tahap optimisasi kedua. Algoritma PSO digunakan untuk menentukan nilai optimal konstanta proporsional (Kₚ) dan integral (Kᵢ) PI dengan meminimalkan integral kesalahan (IAE), serta memperhitungkan trade‐off antara overshoot, settling time, dan keausan aktuator (Δu RMS). Hasil pengujian menunjukkan bahwa kontroler PI–PSO mampu menurunkan IAE dari 835,72 %·s menjadi 325,84 %·s (efisiensi > 60 %). Selain itu, Δu RMS berhasil ditekan dari 0,143 % menjadi 0,061 %, sehingga gerakan valve menjadi lebih halus dan umur mekanis sistem meningkat. Analisis terhadap tiga konfigurasi (PI pabrik, optimisasi minggu pertama, dan minggu kedua) menunjukkan bahwa PSO secara konsisten meningkatkan performa sistem secara nyata. Dengan demikian, kontroler PI–PSO adaptif yang dikembangkan terbukti efektif, robust terhadap variasi kondisi operasi, dan mampu menjaga kestabilan brine level dalam batas aman.
============================================================
===========================================================================
This study develops and evaluates an adaptive PI controller based on Particle Swarm Optimization (PSO) to regulate brine level in the brine blowdown system of a desalination plant at a Combined Cycle Power Plant (PLTGU), aiming to maintain it near the 50% set-point. The process begins with system identification using a Nonlinear ARX (NARX) model trained on real operational data, achieving up to 95% model accuracy after the second optimization phase. PSO is employed to determine the optimal proportional (Kₚ) and integral (Kᵢ) gains of the PI controller by minimizing the Integral of Absolute Error (IAE), while balancing trade-offs involving overshoot, settling time, and actuator wear (Δu RMS). Simulation and field results show that the PI–PSO controller successfullydecreased IAE from 835.72 %·s to 325.84 %·s indicating an improvement of over 60%. Furthermore, the control effort (Δu RMS) dropped from 0.143 % to 0.061 %, enhancing actuator smoothness and mechanical lifespan. Comparative analysis across three configurations (factory PI, first optimization, and second optimization) confirms that PSO consistently enhances the system’s real-world performance. Therefore, the proposed adaptive PI–PSO controller proves to be robust, effective under varying operating conditions, and capable of maintaining brine level stability within safe limits.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	PI, Particle Swarm Optimization, Brine Level, NARX
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ164 Power plants--Design and construction T Technology > TJ Mechanical engineering and machinery > TJ217 Adaptive control systems T Technology > TJ Mechanical engineering and machinery > TJ217.2 Robust control
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Rifqi Abid Razzaq
Date Deposited:	05 Aug 2025 10:07
Last Modified:	05 Aug 2025 10:07
URI:	http://repository.its.ac.id/id/eprint/127606

Actions (login required)

View Item