Expert Systems
1
Critical Review “Evolutionary Learning of Fuzzy Logic Controllers and Their Adaptation Through Perpetual Evolution” By Athula Rajapaksha, Kazuo Furuta, and Shunsuke Kondo
Gayan Rantharu Attanayake, Reading for M.Sc. in Industrial Automation, University of Moratuwa (128802 T) and Gas. The development of the controller begins with; 1. Identification of an artificial neural network (ANN) model of the process 2. Designing of an initial fuzzy controller through genetic learning using neural network model. 3. Applicability of evolutionary computing techniques for real-time adaptation of the fuzzy controller II. PROPOSED ADAPTIVE CONTROL ARCHITECTURE In the proposed adaptive control architecture consist of four main components 1. Fixed fuzzy logic controller 2. Variable fuzzy logic controller 3. Evolutionary tuner 4. Online neural network model of the controller process It emphasized that fixed fuzzy logic control is designed for offline and parameters are kept fixed. The variable controllers are allowed to change the parameters which can adopt for different working conditions. The evolutionary tuner executes the adaptive process. At the each step GA will apply to search for better parameter set. Fitness of the parameter will produces through genetic operation. The parameters of the variable fuzzy controller are replaces with those found by genetic search, if they better fit for the current control task than the existing parameters.
Abstract This paper presents a critical review of the article “ Evolutionary Learning of Fuzzy Logic Controllers and Their Adaptation Through Perpetual Evolution” which was written by Athula Rajapaksha, Kazui Furuta, and Shunsuke Kondo. The article presents an adaptive control architecture, where evolutionary learning is applied for initial earning and real-time tuning of a fuzzy logic controller. The proposed adaptive mechanism was based on the “Perpetual Evolution” where parameters of the fuzzy controller are updated at each time
Critical Review “Evolutionary Learning of Fuzzy Logic Controllers and Their Adaptation Through Perpetual Evolution” By Athula Rajapaksha, Kazuo Furuta, and Shunsuke Kondo
Gayan Rantharu Attanayake, Reading for M.Sc. in Industrial Automation, University of Moratuwa (128802 T) and Gas. The development of the controller begins with; 1. Identification of an artificial neural network (ANN) model of the process 2. Designing of an initial fuzzy controller through genetic learning using neural network model. 3. Applicability of evolutionary computing techniques for real-time adaptation of the fuzzy controller II. PROPOSED ADAPTIVE CONTROL ARCHITECTURE In the proposed adaptive control architecture consist of four main components 1. Fixed fuzzy logic controller 2. Variable fuzzy logic controller 3. Evolutionary tuner 4. Online neural network model of the controller process It emphasized that fixed fuzzy logic control is designed for offline and parameters are kept fixed. The variable controllers are allowed to change the parameters which can adopt for different working conditions. The evolutionary tuner executes the adaptive process. At the each step GA will apply to search for better parameter set. Fitness of the parameter will produces through genetic operation. The parameters of the variable fuzzy controller are replaces with those found by genetic search, if they better fit for the current control task than the existing parameters.
Abstract This paper presents a critical review of the article “ Evolutionary Learning of Fuzzy Logic Controllers and Their Adaptation Through Perpetual Evolution” which was written by Athula Rajapaksha, Kazui Furuta, and Shunsuke Kondo. The article presents an adaptive control architecture, where evolutionary learning is applied for initial earning and real-time tuning of a fuzzy logic controller. The proposed adaptive mechanism was based on the “Perpetual Evolution” where parameters of the fuzzy controller are updated at each time