<abstract xmlns="http://www.w3.org/1999/xhtml"><p>The article is a summary of previous work on the possibility of using Petri layers in adaptive neuro-fuzzy controllers. In the first part of the paper the controller and two types of Petri layer have been presented, competitive layer which resets certain signals and transition layer which causes omission of signals. Layer properties were described and comparison has been made. In the second part of the paper, the results of a simulation showing the advantages and disadvantages of proposed solutions have been presented. Both quality of reference signal tracking and energetic cost of control process have been calculated. In the last part, analysis and comments on the results were made. Main conclusions are that transition Petri layer can significantly reduce growth of numerical cost of the algorithm despite the increase of fuzzy rules count. Also both competitive Petri layer and transition Petri layer by changing some inner signals can affect output value of the fuzzy system and thus the control quality indicators change. Most positive solutions have been pointed out.</p></abstract>

The article is a summary of previous work on the possibility of using Petri layers in adaptive neuro-fuzzy controllers. In the first part of the paper the controller and two types of Petri layer have been presented, competitive layer which resets certain signals and transition layer which causes omission of signals. Layer properties were described and comparison has been made. In the second part of the paper, the results of a simulation showing the advantages and disadvantages of proposed solutions have been presented. Both quality of reference signal tracking and energetic cost of control process have been calculated. In the last part, analysis and comments on the results were made. Main conclusions are that transition Petri layer can significantly reduce growth of numerical cost of the algorithm despite the increase of fuzzy rules count. Also both competitive Petri layer and transition Petri layer by changing some inner signals can affect output value of the fuzzy system and thus the control quality indicators change. Most positive solutions have been pointed out.

Application of Competitive and Transition Petri Layers in Adaptive Neuro-Fuzzy Controller

Power Electronics and Drives

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

The article is a summary of previous work on the possibility of using Petri layers in adaptive neuro-fuzzy controllers. In the first part of the paper the...

Application of Competitive and Transition Petri Layers in Adaptive Neuro-Fuzzy Controller

Published Online: Oct 27, 2017

Page range: 103 - 115

Received: Apr 19, 2016

Accepted: Jul 04, 2016

DOI: https://doi.org/10.5277/ped160108

© 2017

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.