Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: regulator rozmyty FLC typu TSK

Sortuj według:

Ogranicz wyniki do:

Determination of machining parameters in HSM through TSK-FLC

Nandi A. K.

Journal of Achievements in Materials and Manufacturing Engineering

2007

Vol. 21, nr 2

57-60

Purpose: The optimal setting of machining parameters that may be realized via a suitable model/controller is an important concern to fulfill the overall objectives in machining. Design/methodology/approach: The present paper proposes an approach for determination of optimal setting of machining parameters in high speed climb milling operation through an TSK-type fuzzy logic controller (TSK-FLC). A novel approach is proposed here which combines the techniques of linear regression (LR) and genetic algorithm (GA) to utilize the advantages of each other, in order to develop an efficient FLC for high-speed milling. Findings: Modeling of manufacturing process enables generating of manufacturing data and knowledge representation in machining process. Comparisons of results with real experimental data as well as those obtained by other common methods of modeling show the effectiveness of the FLC. Research limitations/implications: The design approach of fuzzy logic controller uses experimental data for learning. The shape fuzzy subsets as well as the structure(s) of rule consequent functions are the important concern for optimal knowledgebase (KB) of a FLC. Use of the advantages of both LR and GA makes it possible to achieve optimal KB of FLC. Practical implications: Use of developed FLC results in improved productivity and efficiency of machining process via the setting of optimal values of cutting parameters and the possibility to develop automatic manufacturing system by online determination of machining parameters. Originality/value: The paper describes a method for designing a FLC for manufacturing process by a combination of LR and GA, which leads to eliminate a long regression function as required in standard linear regression method.