Control of brushless DC motor using siingle-input fuzzy proportional-integral controller
Over the years, development in control industry has brought a hybrid controller, Fuzzy Proportional-Integral (PI) Controller (FPIC) as Brushless DC (BLDC) motor speed regulator with as good performance as PI controller. The FPIC suffers from lengthy design time due to the large number of rules and p...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/79366/1/NurNaqibahBaharudinMFKE2018.pdf http://eprints.utm.my/id/eprint/79366/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Over the years, development in control industry has brought a hybrid controller, Fuzzy Proportional-Integral (PI) Controller (FPIC) as Brushless DC (BLDC) motor speed regulator with as good performance as PI controller. The FPIC suffers from lengthy design time due to the large number of rules and parameter tuning. Thus, this thesis proposes a newly developed Single-Input Fuzzy PI Controller (SIFPIC) to be used as the BLDC motor speed controller. SIFPIC is a simplified version of FPIC with one input variable derived using signed distance method. SIFPIC gives a speed performance comparable to the FPIC but with much faster computing time and simpler tuning process. The motor performance with SIFPIC is evaluated through simulation and experimental approach in terms of speed, current and torque response under several test conditions. The performance is then compared with the motor performance with discrete PI and FPIC speed controller. FPIC is excluded from the comparison in the experiment due to the limitation of DS1104 Digital Signal Processor. From the simulation conducted, SIFPIC produced a comparable performance as FPIC in speed response where both controllers eliminated undershoot and oscillation problems. Under constant speed and changing speed conditions, SIFPIC also showed it superiority from discrete PI controller with average of 36.3% and 11.7% lower ripples than discrete PI controller, respectively. The simulation findings have been verified by the experimental results. |
|---|