Papers by Keyword: Fuzzy Model

Paper TitlePage

Authors: Jun Sheng Ren, Xian Ku Zhang
Abstract: State estimation is an important topic in controller design. H filtering problem is discussed for fuzzy dynamical systems with time delays by using Takagi-Sugeno (T-S) model. Fuzzy H filter is obtained such that the filtering error system is stable and guarantees a prescribed estimation error level. Delay-dependent Lyapunov functional approach is employed to lower the conservativeness of the filter design. Therefore, the results of fuzzy H filter are delay-dependent. An example is given to illustrate the proposed results.
Authors: Yong Fu Wang, Hua Long Cao, Yi Min Zhang, Bang Chun Wen
Abstract: Modeling of friction force has been a challenging task in mechanical engineering. Traditional way, such as mathematical modeling approaches, was found quite difficult to achieve satisfactory performances due to some immanent nonlinearity and uncertainties in systems. This paper aims to develop fuzzy modeling techniques to characterize the friction dynamics. The proposed fuzzy modeling approach has two folds, that is, extraction of fuzzy rules using data mining techniques; setup of static model based on the fuzzy rules. The results obtained demonstrate that our proposed method in this paper has good potential in many mechanical systems with unknown nonlinear friction.
Authors: Jiri Kocian, Stepan Ozana, Jiri Koziorek
Abstract: Many scientific papers deals with the usage of fuzzy rules to implement PID control. Fuzzy models, especially the Takagi-Sugeno-type, have received significant attention from various fields of interest. It is very often very difficult to determine all the parameters of the Takagi-Sugeno-type controller. In this paper we present optimization of Takagi-Sugeno-type fuzzy regulator parameters by genetic algorithm. Implementation of universal fuzzy P/PS/PD function block implemented to the PLC Simatic S7 300/400 is introduced. Mamdani model is used as comparative model. Parameters of Takagi-Sugeno-type fuzzy regulator are determined by genetic algorithm optimization from comparative regulation surface.
Authors: Jie Chen, Da Wei Qi, Xiu Juan Zhang
Abstract: In order to improve the accuracy of tiger skin texture recognition, a skin texture identification method based on fuzzy closeness is put forward in this paper: four skin texture stripes extracted from a tiger texture image are selected as a research object. The image treatment process is: firstly, connected their endpoints to form three quadrilaterals, then regarded the quadrilateral area, bottom length, and left bottom corner features as the characteristics index of the object, the fuzzy closeness degree derived from the characteristics is compared with the standard models, and the final closeness degree is obtained. Experimental results show that this method can improve the accuracy of tiger skin texture recognition, and identify the tiger skin texture effectively to achieve individual identification of tigers finally.
Authors: Zhong Chu Wang, Ran Bi, Xin Zhao
Abstract: This paper describes the Fuzzy control of circular cooling water in a twin-roll strip cast, joins a feed-forward compensation to solve the time lag in conveying water. Because of the large time lag, multivariable system and hard modeling, this paper presents a Fuzzy controller for them. Based on the synthesis reasoning rules and Fuzzy logic, the Fuzzy model of circular cooling water is established. Then the simulation results show that the strip can effectively keep a constant temperature. And it has a good tracking performance, robust, strong anti-interference ability, and the cooling water can in time exchange heat with the liquid metal to get high quality strip.
Authors: Jie Zhang, Ming Lv, Peng Fei Guo, Liang He, Yu Ming Bo
Abstract: Considering some robot control systems which employ wireless networks to transmit sensor signals between the controller and the nonlinear controlled object, the fault detection is carried out. Firstly, based on T-S fuzzy model, the object is linearized. The fuzzy observer is designed and the error equation of the observer is given by using the fuzzy dominant subsystem rule. Secondly, the error equation is equal to the discrete switched system related to the hop count of the wireless transmission, and the stability of the error system is proved. Finally, a simulation example is given to demonstrate the effectiveness of the proposed method in this paper.
Authors: Qin Li Zhang
Abstract: In this paper, the mathematical equivalence between the conditional mean of a Epanechnikov mixture model (EMM) and the defuzzified output of a rule-centered generalized fuzzy model (RCGFM) is derived theoretically. Our results provide a new perspective for fuzzy systems, i.e., interpreting them from a probabilistic viewpoint. Thus, instead of directly estimating the parameters of the fuzzy rules in a rule-centered generalized fuzzy model, we can first estimate the parameters of the corresponding EMM using any popular density estimation algorithm like the expectation maximization (EM) algorithm. Our experimental results clearly indicate that a rule-centered generalized fuzzy model trained in such a new way has higher approximation accuracy and generalization ability than other models.
Authors: Pavol Fedor, Daniela Perdukova
Abstract: The paper presents a methodology for designing a fuzzy model for the middle section of a continuous processing line for tension processing of various materials (sheet metal, tubes, foil, etc.), which is only described by input/output relations. The measured input/output data of the continuous line are the basis for creating its fuzzy model, which can be further applied in the design of a suitable controller and the verification of its properties by simulation. The first part of the paper describes the procedure of the fuzzy model construction; the second part presents the application of the model in the system of the middle section of a continuous processing line. The fuzzy model structure is based on the state space representation of the dynamic system in discrete form. The properties of the fuzzy model were verified by numerical simulation in Matlab. The obtained results have confirmed the rightness of the design method and its applicability to dynamic systems with multiple inputs and outputs.
Authors: Mohamed Laid Hadjili, Kamel Kara, Oussama Ait Sahed, Jamal Bouyanzar
Abstract: In this work a fuzzy model-based predictive control (FMPC) method that uses modified particle swarm optimization (PSO) is presented. The main objective of this work is the application of this method to the control of a Selective Compliant Assembly Robot Arm (SCARA) with four degrees of freedom (4-DOF).
Authors: Yuan Sheng Huang, Li Ming Yuan
Abstract: According to the national standard, this paper presents the evaluation indexes of power quality and the classifications of each index. The method integrates advantages of both G1 and entropy weight coefficient method. Also, it establishes an fuzzy synthetic evaluation for power quality evaluation by fuzzy theory. 5 observation points on the power quality was graded. The test shows that the combination weighting evaluation model based on fuzzy synthetic evaluation can evaluate the power quality comprehensively and effectively.
Showing 1 to 10 of 14 Paper Titles