Neuro-Fuzzy-Based Control for Parallel Cascade Control
Author:
Karthikeyan R.,Manickavasagam K.,Tripathi Shikha,Murthy K.V.V.
Abstract
Abstract
This paper discusses the application of adaptive neuro-fuzzy inference system (ANFIS) control for a parallel cascade control system. Parallel cascade controllers have two controllers, primary and secondary controllers in cascade. In this paper the primary controller is designed based on neuro-fuzzy approach. The main idea of fuzzy controller is to imitate human reasoning process to control ill-defined and hard to model plants. But there is a lack of systematic methodology in designing fuzzy controllers. The neural network has powerful abilities for learning, optimization and adaptation. A combination of neural networks and fuzzy logic offers the possibility of solving tuning problems and design difficulties of fuzzy logic. Due to their complementary advantages, these two models are integrated together to form more robust learning systems, referred to as adaptive neuro-fuzzy inference system (ANFIS). The secondary controller is designed using the internal model control approach. The performance of the proposed ANFIS-based control is evaluated using different case studies and the simulated results reveal that the ANFIS control approach gives improved servo and regulatory control performances compared to the conventional proportional integral derivative controller.
Publisher
Walter de Gruyter GmbH
Subject
Modeling and Simulation,General Chemical Engineering
Reference38 articles.
1. Parallel cascade control;Luyben;Ind Eng Chem Fund,1973
2. Fuzzy DC motor control using particle swarm optimization Electronic of Practices and Technologies;Allaoua;Leonardo Journal,2009
3. Parallel control strategy abstracted from the bare receptor reflex;Pottmann;Chem Eng Sci,1996
4. Analytical method of PID controller design for parallel cascade control Process Control;Lee,2006
5. logic based control for parallel cascade control;Karthikeyan;Fuzzy,2010
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献