TY - GEN
T1 - Exact inversion of TSK fuzzy systems with linear consequents
AU - Ulu, Cenk
AU - Guzelkaya, Mujde
AU - Eksin, Ibrahim
PY - 2013
Y1 - 2013
N2 - In literature, there is no exact inversion method for TSK fuzzy systems with linear consequents. In this study, an analytical method is proposed for the exact inversion of TSK fuzzy systems with linear consequents of which input variables are described using strong triangular partitions. When strong triangular partitions are used, the universes of discourse of input variables are divided into specific regions. In the proposed method, linear equations of triangular membership functions of inversion variable and the rule consequents are directly used in the analytical formulation of TSK fuzzy system output. In this way, the output of the TSK fuzzy system can be expressed in a unique quadratic form in terms of the inversion variable for any region where only the parameters of the appropriate equations of triangular membership functions are embedded. Thus, the inverse solution is easily obtained for any region by using explicit solution of the quadratic equation. An illustrative example has been given to validate the proposed method.
AB - In literature, there is no exact inversion method for TSK fuzzy systems with linear consequents. In this study, an analytical method is proposed for the exact inversion of TSK fuzzy systems with linear consequents of which input variables are described using strong triangular partitions. When strong triangular partitions are used, the universes of discourse of input variables are divided into specific regions. In the proposed method, linear equations of triangular membership functions of inversion variable and the rule consequents are directly used in the analytical formulation of TSK fuzzy system output. In this way, the output of the TSK fuzzy system can be expressed in a unique quadratic form in terms of the inversion variable for any region where only the parameters of the appropriate equations of triangular membership functions are embedded. Thus, the inverse solution is easily obtained for any region by using explicit solution of the quadratic equation. An illustrative example has been given to validate the proposed method.
KW - TSK fuzzy systems
KW - fuzzy model inversion
KW - linear consequents
KW - strong triangular partitions
UR - http://www.scopus.com/inward/record.url?scp=84893366089&partnerID=8YFLogxK
U2 - 10.1109/ICAT.2013.6684090
DO - 10.1109/ICAT.2013.6684090
M3 - Conference contribution
AN - SCOPUS:84893366089
SN - 9781479904310
T3 - 2013 24th International Conference on Information, Communication and Automation Technologies, ICAT 2013
BT - 2013 24th International Conference on Information, Communication and Automation Technologies, ICAT 2013
PB - IEEE Computer Society
T2 - 2013 24th International Conference on Information, Communication and Automation Technologies, ICAT 2013
Y2 - 30 October 2013 through 1 November 2013
ER -