Inference methods for systems with many fuzzy inputs
Dmitry A. Kutsenko, V. G. Sinyuk · Journal of Computer and Systems Sciences International · 2015
In fuzzy simulation, both crisp and fuzzy information can be input into simulated fuzzy systems. The known methods of fuzzy inference for fuzzy input values either have low computational efficiency or do not allow using the whole diversity of logical operations. In this paper, a new method of logical inference is described based on a fuzzy degree of truth for systems with many inputs at which fuzzy input values are received. The method is compared with the initial Zadeh method and the popular Mamdani method. The computational efficiency of the proposed method is also shown. The method is generalized to systems with a block of rules.