fuzzy inference system

Pressing molds are widely used in the industry due to the high speed of parts production. Due to the very low clearance between the mandrel and the matrix and having complex shapes, the production of these molds is very expensive and costly.  The hardness of the mold and subsequently their fatigue life should be high. One of the processes that has the ability to create surfaces with high dimensional accuracy and complex shapes is the wirecut process. In this research, in order to improve the performance and increase the life of cutting dies, the hardness of Mo40 steel is optimized. For this purpose, based on the method of the response procedure, the design of the experiment has been designed for the three parameters of wire feed speed, wire tension and generator power. Next, the obtained data are clustered and then fuzzy rules with three inputs (wire speed, tension and power) and output (hardness) are extracted. The obtained rules were entered in the toolbox of fuzzy inference systems of MATLAB software. Based on the defined fuzzy inference system, it is possible to predict the hardness based on the parameters of wire speed, wire tension and generator power. In the next phase, based on this system and within the range of available variables, the optimal value of hardness and corresponding variables were extracted. In the final phase, these values ​​were experimentally tested and their agreement with the obtained value was observed.

The aerodynamic coefficients of airfoils are usually calculated through numerical and experimental methods causing time and cost consuming as well as depending on the airfoil surface cover. The estimation of aerodynamic coefficients of fabric covered airfoil through fuzzy logic, neural network and fuzzy-neural methods is the innovation of this research in order to determine a low-cost and fast method for optimal design as well as determining the aerodynamic coefficients of airfoils of vehicles having membrane wings. In the models, subsonic velocity was considered. Reynolds number and angle of attack were assumed as input and the values of lift and drag coefficients were assumed as output. Estimations were made on NACA2418 airfoil data, after which the final error of each method was compared. The mean squared error of lift and drag coefficients were 0.8023 and 4.3451e-04 for fuzzy model 0.0012 and 7.5767e-6 for neural network and 0.0697 and 0.0076 for network-fuzzy inference model respectively. The obtained results indicated good fitting of three studied models and best fitting for neural network model, which confirmed by the lift coefficients obtained from experimental tests done for validation.