Explore chapters and articles related to this topic
Handling Uncertainty: Probability and Fuzzy Logic
Published in Adrian A. Hopgood, Intelligent Systems for Engineers and Scientists, 2021
In the foregoing example, at a temperature of 180°C, the possibilities for the pressure being high and medium, μHP and μMP, are set to 0.75 and 0.25, respectively, by the fuzzy rules r3_6f and r3_7f. It is assumed that the possibility for the pressure being low, μLP, remains at 0. These values can be passed on to other rules that might have pressure is high or pressure is medium in their condition clauses without any further manipulation. However, if we want to interpret these membership values in terms of a numerical value of pressure, they would need to be defuzzified. Defuzzification is particularly important when the fuzzy variable is a control action such as “set current,” where a specific setting is required. The use of fuzzy logic in control systems is discussed further in Section 3.5. Defuzzification takes place in two stages, described in the following subsections.
AI-Based Scheme for the Protection of Power Systems Networks Due to Incorporation of Distributed Generations
Published in Almoataz Y. Abdelaziz, Shady Hossam Eldeen Abdel Aleem, Anamika Yadav, Artificial Intelligence Applications in Electrical Transmission and Distribution Systems Protection, 2021
Bhavesh Kumar R. Bhalja, Yogesh M. Makwana
The architecture of fuzzy logic contains several parts such as (i) rule base, (ii) fuzzification, (iii) inference engine, and (iv) defuzzification. In rule base, the expert develops certain rules based on linguistic information for smooth decision making with the help of IF and THEN conditions. However, in recent development, by the use of effective method of fuzzy logic design, the number of fuzzy rules could be minimalized. The fuzzification converts input into fuzzy sets. These inputs are called Crisps. The Crisps are the sensors measured output for control system processing. These sensors could be temperature, voltage, currents, pressure, speed, etc. Based on the input, it is a duty of Inference Engine to decide which fuzzy rule(s) is applicable to fire. With respect to the combined fire rules, the further control action would be taken. The defuzzification is the process of conversion of fuzzy sets obtained from inference into the output crisp value. For this process, different methods have been developed. Among these available methods, the best suited method with its expert system is used to minimize the error.
C
Published in Philip A. Laplante, Comprehensive Dictionary of Electrical Engineering, 2018
centripetal force force that is present during the robot motion. The force depends upon the square of the joint velocities of the robot and tends to reduce the power available from the actuators. centroid (1) the center of a mass. (2) description of the center of a particle beam profile. (3) a region in the pattern space to which a remarkable number of patterns belong. centroid defuzzification a defuzzification scheme that builds the weighted sum of the peak values of fuzzy subsets with respect to the firing degree of each fuzzy subset. Also called height defuzzification. centroid method a widely used method of defuzzification whereby the centroid of the membership function of the fuzzy set is used as the defuzzified or crisp value. It is also known as the center of gravity method or the composite moments method. centroidal profile a method for characterizing and analyzing the shape of an object having a well defined boundary. The centroid of the shape is first determined. Then a polar (r, ) plot of the boundary is computed relative to this origin: this plot is the centroidal profile, and has the advantage of permitting template matching for a 2-D shape to be performed relatively efficiently as a 1-D process. centrosymmetric medium a material that possesses a center of inversion symmetry. Of importance because, for example, second-order nonlinear optical processes are forbidden in such a material. cepstrum inverse Fourier transform of the logarithm of the Fourier power spectrum of a signal. The complex cepstrum is the inverse Fourier transform of the complex logarithm of the Fourier transform of the signal.
Pollutant monitoring in tail gas of sulfur recovery unit with statistical and soft computing models
Published in Chemical Engineering Communications, 2019
Akshay Morey, Soumyashis Pradhan, Rahul Anil Kumar, Ajaya Kumar Pani, Venkata Vijayan S., Varun Jain, Aayush Gupta
In general, design of fuzzy inference model involves four modules:Fuzzification: This module translates the crisp number inputs to linguistic variable fuzzy sets. This is known as fuzzification of inputs. This is accomplished by using a membership function stored in the knowledge base.Knowledge Base: This refers to a database that stores the set of if-then rules required for the processInference Engine: This is the heart of the system that mimics human reasoning by applying fuzzy inference to the inputs with the help of if-then rules and converting the fuzzy inputs to the fuzzy outputsDefuzzification module: The final obtained output is then defuzzified to crisp numbers by this module. The commonly used defuzzification methods include centroid of area (COA), mean of maximum (MOM) and bisector of area (BOA).
Investigating barriers to circular supply chain in the textile industry from Stakeholders’ perspective
Published in International Journal of Logistics Research and Applications, 2022
Ipek Kazancoglu, Yigit Kazancoglu, Aysun Kahraman, Emel Yarimoglu, Gunjan Soni
A successful approach to defuzzification must consider a fuzzy number identified by its form, spread, height, and relative position on the x-axis (Opricovic and Tzeng 2004). Centroid (Center-of-gravity) is regarded as the most widely used defuzzification method (Yager and Filev 1994), however, it cannot differentiate two fuzzy numbers that have the same crisp value in terms of different forms. Thus, Converting Fuzzy data into Crisp Scores (CFCS), is used to provide more crisp value than the Centroid method.
Medical image fusion using type-2 fuzzy and near-fuzzy set approach
Published in International Journal of Computers and Applications, 2020
Biswajit Biswas, Biplab Kanti Sen
Defuzzification is the translation of a fuzzy quantity to an accurate quantity, as fuzzification is the translation of an exact quantity to a fuzzy quantity. It is an inverse functional mapping from fuzzy plane to crisp plane[28,31]. In recent years, the several defuzzification methods have been proposed in the literature. Most common defuzzification techniques are centroid method, center of area, center of gravity, weighted average method, mean max membership, center of sums method, etc. [31].