Method of fuzzy inference for one class of MISO-structure systems with non-singleton inputs

International Conference on Mechanical Engineering, Automation and Control Systems 2017

机械制造;无线电电子学;计算机科学

Sinuk, V.G.^1 ; Panchenko, M.V.^1

Belgorod State Technological University, Kostyukova St. 46, Belgorod

308012, Russia^1

关键词: Fuzzy implications;    Fuzzy input;    Fuzzy modeling;    Fuzzy truth values;    Polynomial complexity;    Simulated system;    Structure systems;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/327/4/042074/pdf
DOI  :  10.1088/1757-899X/327/4/042074

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

In fuzzy modeling, the inputs of the simulated systems can receive both crisp values and non-Singleton. Computational complexity of fuzzy inference with fuzzy non-Singleton inputs corresponds to an exponential. This paper describes a new method of inference, based on the theorem of decomposition of a multidimensional fuzzy implication and a fuzzy truth value. This method is considered for fuzzy inputs and has a polynomial complexity, which makes it possible to use it for modeling large-dimensional MISO-structure systems.