İlhami Ilhan

Estimation of relative humidity based on artificial neural network approach in the Aegean Region of Turkey

The aim of this study is to estimate the monthly mean relative humidity (MRH) values in the Aegean Region of Turkey with the help of the topographical and meteorological parameters based on artificial neural network (ANN) approach. The monthly MRH values were calculated from the measurement in the meteorological observing stations established in Izmir, Mugla, Aydin, Denizli, Usak, Manisa, Kutahya and Afyonkarahisar provinces between 2000 and 2006. Latitude, longitude, altitude, precipitation and months of the year were used in the input layer of the ANN network, while the MRH was used in output layer of the network. The ANN model was developed using MATLAB software, and then actual values were compared with those obtained by ANN and multi-linear regression methods. It seemed that the obtained values were in the acceptable error limits. It is concluded that the determination of relative humidity values is possible at any target point of the region where the measurement cannot be performed.

Predicting the unevenness of polyester/viscose/acrylic-blended open-end rotor spun yarns

Mixture spinning aims to bring together the best properties of two or more fiber types into one single yarn. In mixture spinning, while some fiber types have impacts to increase the yarn strength, the other fiber types may have impacts to improve the yarn unevenness values. In this study, it is aimed to obtain a regression model that predicts the unevenness (CVm%) of polyester/viscose/acrylic (PES/CV/PAN)-blended OE-rotor spun yarns. For this purpose, the methods of statistical analysis as analysis of variance and Combined Analysis were used. In the context of experimental design, basic parameters affecting the yarn unevenness were selected as: the type of fiber mixture (by carding machine or drawframe), blend ratio (from 0 to 100% at five different rates), and the yarn count (24, 30 and 36 tex). As a result, a statistically significant (p = 0.05) model has been established to estimate the unevenness of PES/CV/PAN blended OE-rotor yarns. In addition, two-dimensional surface graphics have been created to view CVm% values according to the fiber ratios in ternary mixtures. For all of the yarn samples, the increase in acrylic ratio has been found to have positive impact on the CVm% values.

Predicting twist contraction in chenille yarn using mathematical and statistical approaches

In this study, the effective factors on twist contraction of chenille yarn that is a kind of fancy yarn were investigated with theoretical and statistical approaches. The aim of this investigation was to develop a model to predict the twist contraction in chenille yarn before production. In the theoretical approach, we developed a mathematical model by geometrical methods. Twenty-seven different chenille yarn samples were produced according to an experimental design containing different levels of effective factors. In statistical approach, the data sets were divided into two parts as 20 and 7. A stepwise regression analysis was applied to 20 of all and acquired a significant model. The remaining data sets were used for verification. Finally, the mathematical and statistical models were verified and compared their performance to estimate twist contraction. As a result, it has proved that the performance of regression model is slightly better than the mathematical model. However, both of the models can give acceptable predicted values for twist contraction in chenille yarns.