Browsing by Author "Bahsi, MM"

Now showing 1 - 4 of 4
  • No Thumbnail Available
    Item
    Machine learning for the prediction of problems in steel tube bending process
    Görüs, V; Bahsi, MM; Çevik, M
    Tube bending is a widely used process in marine, automotive, construction and other industries. Different methods are available for this process during which various problems and defects are encountered. This paper aims to develop the most relevant model for bending the tubes using a computer numerical controlled (CNC) bending machine on the first attempt without any defects. To achieve this goal, the parameters affecting the tube bending process on the bending machine are defined. Based on these parameters, 150 bending experiments are conducted in an industrial plant to collect data and their results are used as dataset for machine learning (ML) algorithms. Seven different state-of-the-art ML algorithms -logistic regression, decision tree, k-nearest neighbor, random forest, Na & iuml;ve Bayes, support vector machine and eXtreme gradient boosting (XGBoost) - are implemented using Python 's Scikit-Learn library. Their performance is compared using confusion matrix and classification metrics such as accuracy, precision, recall, F1-score, receiver operating characteristic (ROC) curve and area under the ROC curve (AUC) value. Considering all the performance metrics, logistic regression performed best in terms of prediction on our dataset. XGBoost and support vector machine were quiet successful based on F1 score and AUC, respectively. Overall, logistic regression, Na & iuml;ve Bayes, Support Vector Machine, and XGBoost showed performances above 90% across all metrics. Decision tree, k-nearest neighbor and random forest performed poorly compared to other algorithms for our data. The proposed ML method is expected to save costs by reducing material waste and labor time and to increase the process efficiency and the product quality.
  • No Thumbnail Available
    Item
    Improved Jacobi matrix method for the numerical solution of Fredholm integro-differential-difference equations
    Bahsi, MM; Bahsi, AK; Çevik, M; Sezer, M
    This study is aimed to develop a new matrix method, which is used an alternative numerical method to the other method for the high-order linear Fredholm integro-differential-difference equation with variable coefficients. This matrix method is based on orthogonal Jacobi polynomials and using collocation points. The improved Jacobi polynomial solution is obtained by summing up the basic Jacobi polynomial solution and the error estimation function. By comparing the results, it is shown that the improved Jacobi polynomial solution gives better results than the direct Jacobi polynomial solution, and also, than some other known methods. The advantage of this method is that Jacobi polynomials comprise all of the Legendre, Chebyshev, and Gegenbauer polynomials and, therefore, is the comprehensive polynomial solution technique
  • No Thumbnail Available
    Item
    Solution of the delayed single degree of freedom system equation by exponential matrix method
    Çevik, M; Bahsi, MM; Sezer, M
    In this paper, an exponential collocation method for the solution linear delay differential equations with constant delay is presented. The utility of this matrix based method is that it is very systematic and by writing a Maple program, any type of second order linear differential delay equation can be solved easily. The method is applied to three different types of delay equations; linear oscillator with delay (i) in the restoring force term, (ii) in the damping term, and (iii) in the acceleration term. Time response curves have been plotted for each type and the effect of the parameters of the delay terms has been shown. An error analysis based on residual function is carried out to show the accuracy of the results. (C) 2014 Elsevier Inc. All rights reserved.
  • No Thumbnail Available
    Item
    Orthoexponential polynomial solutions of delay pantograph differential equations with residual error estimation
    Bahsi, MM; Çevik, M; Sezer, M
    In this paper, a new matrix method based on orthogonal exponential (orthoexponential) polynomials and collocation points is proposed to solve the high-order linear delay differential equations with linear functional arguments under the mixed conditions. The convenience is that orthoexponential polynomials have shown to be effective in approximating a given function, fast and efficiently. An error analysis technique based on residual function is developed and applied to four problems to demonstrate the validity and applicability of the proposed method. It is confirmed that the present method yields quite acceptable results and the accuracy of the solution can significantly be increased by error correction and residual function. (C) 2015 Elsevier Inc. All rights reserved.