Related Experiment Videos
Design of a neurofuzzy controller with simplified architecture
M E Bordon1, I N da Silva, E Avolio
1Department of Electrical Engineering, State University of São Paulo, Bauru, São Paulo 17033-360, Brazil. mebordon@bauru.unesp.br
International Journal of Neural Systems
|September 28, 2001
Summary
This study introduces a simplified neurofuzzy controller that reduces processing time for system modeling. This automated design enables efficient fluid flow control in industrial applications.
Area of Science:
- Control Engineering
- Artificial Intelligence
- Process Systems Engineering
Background:
- Traditional neurofuzzy controllers often involve complex architectures and lengthy processing times for system modeling.
- The configuration of control rules in existing systems can be time-consuming and requires manual intervention.
Purpose of the Study:
- To design a novel neurofuzzy controller with a simplified architecture.
- To minimize processing time in system and process modeling stages.
- To enable automatic generation of control rules for enhanced efficiency.
Main Methods:
- Development of a simplified neurofuzzy controller architecture.
- Streamlining fuzzification and defuzzification processes.
- Implementing direct computation for inference procedures.
- Automatic derivation of control rules based on the simplified architecture.
Main Results:
- Significant reduction in processing time for system and process modeling.
- Fast and straightforward configuration of the neurofuzzy controller.
- Successful automatic generation of control rules.
- Effective validation of the neurofuzzy system in an industrial fluid flow control application.
Conclusions:
- The proposed simplified neurofuzzy controller offers a computationally efficient alternative for process control.
- The automated rule generation simplifies controller design and implementation.
- The approach is validated for practical industrial applications, specifically in fluid flow control.