Fuzzy Control of Continuous Stirred Tank Reactor
Continuous Stirred Tank Reactor (CSTR) involves complex reactions, highly nonlinear and very hard to control by conventional methods. Fuzzy controller is introduced in this research to control the CSTR. A powerful design and simulation tool which is MATLAB and SIMULINK has been used for evaluatio...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2006
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/212/3/549044_fk_2006_10_abstrak_je__dh_pdf_.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Continuous Stirred Tank Reactor (CSTR) involves complex reactions, highly nonlinear
and very hard to control by conventional methods. Fuzzy controller is
introduced in this research to control the CSTR. A powerful design and simulation
tool which is MATLAB and SIMULINK has been used for evaluation and test.
In this research, the process considered is the decomposition of hydrogen peroxide
(H2O2) using Fe3+ as the catalyst. The product is water (H20) and dissipated oxygen
(O2). The main task is to maintain the temperature and concentration in the CSTR
around the working point, in spite of disturbances by manipulating the flowrate of
inlet or outlet stream and coolant temperature.
Two fuzzy controllers have been designed and tested based on Mamdani and Sugeno
inference mechanisms. Performance comparisons have been made; first between the
two controllers and second between the best fuzzy controller and previous researches
on the similar topic but using different techniques. The overall performance of both Mamdani and Sugeno based fuzzy controller are excellent. The tests results showed
that Sugeno and Mamdani based fuzzy controllers have no significant difference in
performances due to slow process occurred in the CSTR. However, Sugeno based
fuzzy controller slightly improves the response time of Mamdani based fuzzy
controller by 20%. Steady state error is also reduced to less than 1%. Furthermore, the
simulation time is also reduced.
Fuzzy control provides a simple technique but yet the achievements are comparable
to other more complicated and time consuming techniques. Various tests have been
conducted throughout this research to show the robustness of both fuzzy controllers. |
|---|