Perancangan dan Implementasi BLDC Controller Berbasis Field-Oriented Control (FOC) dengan Algoritma Adaptif Gain Scheduling untuk Kondisi Beban Bervariasi

Rahman, Anis (2026) Perancangan dan Implementasi BLDC Controller Berbasis Field-Oriented Control (FOC) dengan Algoritma Adaptif Gain Scheduling untuk Kondisi Beban Bervariasi. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022221049-Undergraduate-Thesis.pdf - Accepted Version Restricted to Repository staff only Download (34MB) | Request a copy

Abstract

Motor Brushless DC (BLDC) banyak digunakan pada aplikasi penggerak karena efisiensi dan kerapatan dayanya yang tinggi, dan Field-Oriented Control (FOC) menjadi metode kendali yang umum untuk menghasilkan torsi yang halus. Namun, pengendali PI dengan gain tetap pada FOC kurang adaptif terhadap variasi beban, sehingga respons kecepatannya cenderung memburuk (overshoot, dip, dan waktu pemulihan membesar) ketika beban berubah. Penelitian ini merancang dan mengimplementasikan motor controller BLDC berbasis FOC yang dilengkapi algoritma Adaptive PI Gain Scheduling berbasis logika fuzzy Mamdani pada mikrokontroler STM32G474. FOC baseline mencakup transformasi Clarke–Park, CORDIC, SVPWM center-aligned 20 kHz melalui HRTIM, serta loop PI arus dan kecepatan. Modul fuzzy dirancang sebagai lapisan additive dengan dua masukan (error e dan laju perubahannya ∆e) dan dua keluaran (koreksi ∆Kp dan ∆Ki), 25 aturan Mamdani, serta defuzzifikasi weighted average; parameter motor (Kt = 0,0756 Nm/A, Ke = 0,0171 V·s/rad) dan kurva torsi rem elektromagnetik KEB005AA dikarakterisasi secara empiris sebagai dasar perancangan. Kinerja PI tetap dan adaptive fuzzy dibandingkan melalui pengujian step disturbance pada tiga level beban dan tiga titik operasi (200–600 RPM) dengan tiga repetisi (total 54 trial dengan gangguan yang identik antar konfigurasi). Hasilnya, skema adaptive fuzzy unggul pada seluruh 54 trial di setiap metrik tanpa satu pun sel yang kalah. Perbaikan terbesar muncul pada indeks error integral yang merangkum kualitas pemulihan sepanjang gangguan, yaitu ISE membaik 13–44% (rata-rata ∼33%) dan IAE 8–27% (rata-rata ∼23%); simpangan puncak juga membaik, yakni dip 5–22% dan overshoot 6–21%. Perbaikan tersebut konsisten di seluruh rentang beban, dan performa adaptive fuzzy meluruh lebih lambat terhadap kenaikan torsi (dip-versus-torque slope ∼17% lebih landai), menandakan ketahanan yang lebih baik terhadap variasi beban. Dengan demikian, adaptive PI scheduling berbasis fuzzy terbukti secara empiris efektif meningkatkan kualitas respons, kekokohan, dan konsistensi kendali kecepatan BLDC pada kondisi beban yang berubah-ubah.
=====================================================================================================================================
Brushless DC (BLDC) motors are widely used in drive applications owing to their high efficiency and power density, and Field-Oriented Control (FOC) is a common control method for producing smooth torque. However, a fixed-gain PI controller in FOC is not adaptive to load variations, so its speed response tends to degrade (larger overshoot, dip, and recovery time) as the load changes. This research designs and implements an FOC-based BLDC motor controller equipped with an Adaptive PI Gain Scheduling algorithm based on Mamdani fuzzy logic on an STM32G474 microcontroller. The baseline FOC comprises Clarke–Park transforms, CORDIC, 20 kHz center-aligned SVPWM via HRTIM, and current and speed PI loops. The fuzzy module is designed as an additive layer with two inputs (the error e and its rate of change ∆e) and two outputs (the gain corrections ∆Kp and ∆Ki), 25 Mamdani rules, and weighted-average defuzzification; the motor parameters (Kt = 0.0756 Nm/A, Ke = 0.0171 V·s/rad) and the torque curve of a KEB005AA electromagnetic brake were characterized empirically as the design basis. The fixed-PI and adaptive-fuzzy schemes were compared through step-disturbance tests at three load levels and three operating points (200–600 RPM), each repeated three times (54 trials under identical disturbances). The adaptive-fuzzy scheme outperformed the fixed-PI baseline in all 54 trials on every metric, with no losing cell. The largest gains appeared in the integral error indices that summarize recovery quality over the whole disturbance: ISE improved by 13–44% (about 33% on average) and IAE by 8–27% (about 23% on average); peak deviations also improved, with the dip reduced by 5–22% and the overshoot by 6–21%. These gains were consistent across the load range, and the adaptive-fuzzy performance degraded more slowly with increasing torque (a dip-versus-torque slope about 17% gentler), indicating better robustness to load variation, while the overshoot-versus-torque slope became slightly steeper (about 12%), suggesting a mild trade-off on speed overshoot under heavier loads where the overshoot itself remained low. In conclusion, the fuzzy-based adaptive PI scheduling was empirically proven to effectively improve the response quality, robustness, and consistency of BLDC speed control under varying load conditions.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Adaptive Gain Scheduling, Brushless DC (BLDC), Field-Oriented Control(FOC), Logika Fuzzy, Variasi Beban, Adaptive Gain Scheduling, Brushless DC (BLDC), Field-Oriented Control (FOC), Fuzzy Logic, Varying Load.
Subjects:	Q Science > QA Mathematics > QA9.64 Fuzzy logic T Technology > TJ Mechanical engineering and machinery > TJ217 Adaptive control systems T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL220 Electric vehicles and their batteries, etc.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Anis Rahman
Date Deposited:	21 Jul 2026 07:45
Last Modified:	21 Jul 2026 07:46
URI:	http://repository.its.ac.id/id/eprint/135997

Actions (login required)

View Item