PREDIKSI UMUR PAHAT DENGAN METODE MESIN PENDUKUNG VEKTOR (SUPPORT VECTOR MACHINE)

WINOTO, AGUS (2011) PREDIKSI UMUR PAHAT DENGAN METODE MESIN PENDUKUNG VEKTOR (SUPPORT VECTOR MACHINE). Other thesis, Universitas Sebelas Maret.

[img] PDF - Published Version
Download (1117Kb)

    Abstract

    The aim of this research is to estimate the tool life using support vector machine method, that useful in machining process planning. The prediction used algorithm Perceptron Multylayer (MLP), Radial Basis Function (RBF) and polynomial. The input data on the program using the data collected and calculated from measurement data directly from conventional practice. The variations uses in this research are the rotations of shaft, the diameter of workpiece and cutting time. The number of variations of the experiment are as many as 50, 35 used for training data and the rest of 15 data will be used as test data. The analysis shows that that the results of training approached the actual calculation of prediction accuracy reached 90.03% (MLP), 98.17% (RBF) and 98.98% (polynomial). Based on these results can be concluded that the prediction of tool life with polynomial algorithm can be used appropriately and accurately to predict tool life. Keywords : support vector machine, tool life prediction, MLP algorithm, RBF algorithm, Polynomial algorithm.

    Item Type: Thesis (Other)
    Subjects: T Technology > TJ Mechanical engineering and machinery
    Divisions: Fakultas Teknik
    Fakultas Teknik > Teknik Mesin
    Depositing User: AB Kusuma
    Date Deposited: 18 Jul 2013 18:37
    Last Modified: 18 Jul 2013 18:37
    URI: https://eprints.uns.ac.id/id/eprint/5644

    Actions (login required)

    View Item