Artificial Intelligence: A Guide to Intelligent Systems
Artificial Intelligence: A Guide to Intelligent Systems
| Hi-index | 0.00 |
A fuzzy logic controller (FLC) for a Boost power electronics converter has been designed. The controller parameters have been optimized using genetic algorithms. Controller design and simulations in Matlab SimPowerSystems are carried out in this paper. The designed fuzzy controller proved significantly better performance compared to two reference controllers; a classical PI controller and a PI controller tuned with FL. The novel topology is based on a literature study fuzzy controllers for this type of converter. Certain modifications have been made in the controller structure with respect to what have been found in the literature, and it is shown that genetic algorithms can be used for deciding optimal controller parameters. The disadvantages of the FLCs were also pointed out. One is that oscillations occur in the controller loop. To find the reason for these oscillations the system must be further investigated. Another disadvantage is the lack of transfer function theory for FLCs. Simulations can be carried out to check the controllers' sensibility towards parametric variations. However, the stability cannot be proven mathematically. Nevertheless, the Boost converter has better performance equipped with a FLC compared to a converter with a PI controller. Based on these experiences FLCs should be used for systems demanding a high accuracy and that follow a repeated reference like a robot arm or a charge/discharge curve of a super-capacitor bank.