TAILIEUCHUNG - Interpolation based on semantic distance weighting in hedge algebra and its application
In this paper, we propose the use of the interpolation to ensure the correctness mentioned above. It is interpolation based on the semantic distance weighting of the input semantic values versus the semantic value of the elements appearing in the rule. In the input semantic value vector, for each component we define its distance weight to the semantic value of the corresponding language class that appears in the rule. This distance-weighted value is used to join the output value. | Journal of Computer Science and Cybernetics, , (2017), 19–33 DOI INTERPOLATION BASED ON SEMANTIC DISTANCE WEIGHTING IN HEDGE ALGEBRA AND ITS APPLICATION NGUYEN TIEN DUY1 , VU NHU LAN2 1 Thai Nguyen University of Technology, Thai Nguyen University 2 Thang Long University 1 ; 2 vnlan@ Abstract. When solving the approximation problems proposed by a Linguistic Rule Base System (LRBS), the choice of reasoning method has a great influence on the result. The chosen reasoning method should ensure the correctness of the variable relationship between the outputs versus the inputs. According to the Hedge Algebra’s approach, we can convert the LRBS to a real “hyper surface” in semantic space, called the Quantified Semantic Rule Base System (QRBS), and use an interpolative method for reasoning. In this paper, we propose the use of the interpolation to ensure the correctness mentioned above. It is interpolation based on the semantic distance weighting of the input semantic values versus the semantic value of the elements appearing in the rule. In the input semantic value vector, for each component we define its distance weight to the semantic value of the corresponding language class that appears in the rule. This distance-weighted value is used to join the output value. With this proposal, we have designed and simulated the controller that control rules are given by LRBS for the resistance furnace. The results show that the controller works well under the control requirements. Keywords. Hedge algebras, approximate reasoning, bi-linear interpolation, linguistic rule base system, fuzzy control. 1. INTRODUCTION For the past few years, the theory of hedge algebras (HA) [9, 10] has been used to solve problems in many different fields, including the field of control [4, 5, 6, 12, 13, 16]. According to the HA approach, the design of approximately reasoning set working as a controller whose control rules .
đang nạp các trang xem trước
