A Study On The Property Of Algorithms Of Fuzzy Inference Based On Schweizer-Sklar T-norms

Since Zadeh proposed the concept of fuzzy sets in 1965, the fuzzy theory and method is widely used in mathematics and many applications field. Researching the fuzzy inference system with better properties is very important to the application for artificial intelligence and control.Therefore, this paper studies the properties of Schweizer-Sklar operators and the robustness of fuzzy inference algorithms based on Schweizer-Sklar t-norms, and then constructs a many fuzzy measurement spaces based on Schweizer-Sklar t-norms. The outline of this thesis is organised as follow:Fristly, we choose a fuzzy logic based on Schweizer-Sklar t-norms which is more flexible, confirming Minkowski distance to be the measurement standard to calculate the perturbation which is rasied by SchweizerSklar t-norms, its residuated implication, and then give the robustness of triple I aigorithms based on Schweizer-Sklar t-norms.Secondly, we continue with fuzzy logic based on Schweizer-Sklar tnorms, and the property of Schweizer-Sklar operators is studied, and then using the concept of maximum sensitivity to get the robustness of reverse triple I method based on Schweizer-Sklar t-norms. This result is very useful in control field when choosing the fuzzy inference.Lastly, we expand the concept of regular measurement in ref. [25]to Schweizer-Sklar t-norms(m<0), and then construct a bunch of fuzzy measurement spaces and prove every point in them is discrete.