Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7148
Browse
10 results
Search Results
Now showing 1 - 10 of 10
Article Citation - WoS: 5Citation - Scopus: 6Prediction of Rainfall Runoff-Induced Sediment Load From Bare Land Surfaces by Generalized Regression Neural Network and Empirical Model(Wiley, 2020) Tayfur, Gökmen; Aksoy, Hafzullah; Eriş, EbruBased on three rainfall run-off-induced sediment transport data for bare surface experimental plots, the generalized regression neural network (GRNN) and empirical models were developed to predict sediment load. Rainfall intensity, slope, rainfall duration, soil particle median diameter, clay content of the soil, rill density and soil particle mass density constituted the input variables of the models while sediment load was the target output. The GRNN model was trained and tested. The GRNN model was found successful in predicting sediment load. Sensitivity analysis by the GRNN model revealed that slope and rainfall duration were the most sensitive parameters. In addition to the GRNN model, two empirical models were proposed: (1) in the first empirical model, all the input variables were related to the sediment load, and (2) in the second empirical model, only rainfall intensity, slope and rainfall duration were related to the sediment load. The empirical models were calibrated and validated. At the calibration stage, the coefficients and the exponents of the empirical models were obtained using the genetic algorithm optimization method. The validated empirical models were also applied to two more experimental data sets: (1) one data set was from a field experiment, and (2) one set was from a laboratory experiment. The results indicated the success of the empirical models in predicting sediment load from bare land surfaces.Article Citation - WoS: 37Citation - Scopus: 41Determination of Olive Oil Adulteration With Vegetable Oils by Near Infrared Spectroscopy Coupled With Multivariate Calibration(SAGE Publications, 2010) Öztürk, Betül; Yalçın, Ayşegül; Özdemir, DurmuşThere has been growing public awareness about the health benefits of olive oil throughout the world in recent years, resulting in a significant increase in its consumption as part of the daily diet This demand has attracted fraudulent attempts to market olive oil which has been adulterated with cheaper oils. This study focuses on the near infrared (NIR) spectroscopic determination of adulteration of olive oil by vegetable oils using multivariate calibration. The binary, ternary and quaternary mixtures of olive, soybean, cotton, corn, canola and sunflower oils were prepared using a random design. The absorbance spectra of these synthetic samples were measured by a near infrared (NIR) spectrometer. A genetic algorithm-based variable selection algorithm, coupled with an inverse least squares multivariate calibration method (GILS) was used to build calibration models for possible adulterants and olive oil in the adulterated mixtures The correlation coefficients of actual versus predicted concentrations resulting from multivariate calibration models for the different oils were between 0 90 and 0.99 The results demonstrated that NIR spectroscopy in conjunction with the GILS method makes it possible to determine the adulteration of olive oils regardless of adulterant vegetable oils over a wide range of concentrations.Article Citation - WoS: 1Citation - Scopus: 1Generalized Regression Neural Network and Empirical Models To Predict the Strength of Gypsum Pastes Containing Fly Ash and Blast Furnace Slag(Springer Verlag, 2020) Erdem, Tahir Kemal; Cengiz, Okan; Tayfur, GökmenGypsum is widely used in constructions owing to its easy application, zero shrinkage, and excellent fire resistance. Several parameters can affect the properties of gypsum pastes. To study the strength of the gypsum pastes experimentally by trying all these parameters is time-consuming and costly. Therefore, artificial intelligence methods can be very useful to predict the paste strength, which, in turn, can reduce the number of trial batches. Based on experimental data, the generalized regression neural network (GRNN) and empirical models were developed to predict strength of gypsum pastes containing fly ash (FA) and blast furnace slag (BFS). Gypsum content, pozzolan content, curing temperature, curing duration, and testing age constituted the input variables of the models while the paste strength was the target output. The trained and tested GRNN model was found to be successful in predicting strength. Sensitivity analysis by the GRNN model revealed that the curing duration and temperature were important sensitive parameters. In addition to the GRNN model, empirical models were proposed for the strength prediction. The same input variables formed the input vectors of the empirical models. The same dataset used for the calibration of the GRNN model was employed to establish the empirical models by employing genetic algorithm (GA) method. The empirical models were successfully validated. The GRNN and GA_based empirical models were also tested against the multi-linear regression (MLR) and multi-nonlinear regression (MNLR) models. The results showed the outperformance of the GRNN and the GA_based empirical models over the others.Article Citation - WoS: 24Citation - Scopus: 24GA-optimized model predicts dispersion coefficient in natural channels(IWA Publishing, 2009) Tayfur, GökmenModels whose parameters were optimized by genetic algorithm (GA) were developed to predict the longitudinal dispersion coefficient in natural channels. Following the existing equations in the literature, ten different linear and nonlinear models were first constructed. The models relate the dispersion coefficient to flow and channel characteristics. The GA model was then employed to find the optimal values of the constructed model parameters by minimizing the mean absolute error function (objective function). The GA model utilized an 80% cross-over rate and 4% mutation rate. It started each computation with a population of 100 chromosomes in the gene pool. For each model, while minimizing the objective function, the values of the model parameters were constrained between [-10, +10] at each iteration. The optimal values of the model parameters were obtained using a calibration set of 54 out of 80 sets of measured data. The minimum error was obtained for the case where the model was a linear equation relating dispersion coefficient to flow discharge. The model performance was then satisfactorily tested against the remaining 26 measured validation datasets. It performed better than the existing equations. it yielded minimum errors of MAE = 21.4m2/s (mean absolute error) and RMSE = 28.5m2/s (root mean-squares error) and a maximum accuracy rate of 81%. © IWA Publishing 2009.Article Citation - WoS: 18Citation - Scopus: 22Prediction of Lignin and Extractive Content of Pinus Nigra Arnold. Var. Pallasiana Tree Using Near Infrared Spectroscopy and Multivariate Calibration(Taylor and Francis Ltd., 2009) Üner, Birol; Karaman, İbrahim; Tanrıverdi, H.; Özdemir, DurmuşDetermination of quality parameters such as lignin and extractive content of wood samples by wet chemistry analyses takes a long time. Near infrared (NIR) spectroscopy coupled with multivariate calibration offers a fast and nondestructive alternative to obtain reliable results. However, due to the complexity of the spectra obtained from NIR, some wavelength selection is generally required to improve the predictive ability of multivariate calibration methods. Pinus nigra Arnold. Var. pallasiana is the second most growing pine species in Turkey. Even though its rotation period is very high, around 120 years, the forest products industry has widely accepted the use of Pinus nigra because of its ability to grow on a wide range of sites and its suitability to produce desirable products. In this study, 51 samples of Pinus nigra trees were collected and their lignin and extractive content were determined with standard reference (TAPPI) methods. Then, the same samples were scanned with near infrared spectrometer between 1000 and 2500 nm in diffuse reflectance mode. Multivariate calibration models were built with genetic inverse least squares method for both lignin and extractive content using the concentration information obtained from wet standard reference method. Overall, standard error of calibration (SEC) and standard error of prediction (SEP) were ranged between 0.35% (w/w) and 2.4% (w/w).Article Citation - WoS: 11Citation - Scopus: 13Near Infrared Spectroscopic Determination of Diesel Fuel Parameters Using Genetic Multivariate Calibration(Taylor and Francis Ltd., 2008) Özdemir, DurmuşThe use of full spectral region from near infrared spectroscopic analysis does not always end up with a good multivariate calibration model as many of the wavelengths do not contain necessary information. Due to the complexity of the spectra, some of the wavelengths or regions may, in fact, disturb the model-building step. Genetic algorithms are one of the useful tools for solving wavelength selection problems and may improve the predictive ability of conventional multivariate calibration methods. This study demonstrates application of genetic algorithm-based multivariate calibration to near infrared spectroscopic determination of several diesel fuel parameters. The parameters studied are cetane number, boiling and freezing point, total aromatic content, viscosity, and density. Multivariate calibration models were generated using genetic inverse least squares (GILS) method and used to predict the diesel fuel parameters based on their near infrared spectra. For each property, a different data set was used and in all cases the number of samples was around 250. Overall, percent standard error of prediction (%SEP) values ranged between 2.48 and 4.84% for boiling point, total aromatics, viscosity, and density. However, %SEP results for cetane number and freezing point were 11.00% and 14.86%, respectively.Article Citation - WoS: 63Citation - Scopus: 73Near Infrared Spectroscopic Determination of Olive Oil Adulteration With Sunflower and Corn Oil(Taiwan Food and Drug Administration, 2007) Özdemir, Durmuş; Öztürk, BetülDetermination of authenticity of extra virgin olive oils has become very important in recent years due to the increasing public concerns about possible adulterations with relatively cheap vegetable oils such as sunflower oil. This study was focused on the application of near infrared (NIR) spectroscopy in conjunction with multivariate calibration to identify the adulteration of olive oils. NIR transmittance measurements were made on pure olive oil and olive oil adulterated with varying concentrations (4-96%, v/v) of sunflower and corn oil in two sets of 26 binary and ternary mixtures. Multivariate calibration models were generated using genetic inverse least squares (GILS) method and used to predict the concentration of adulterants along with the concentration of olive oil in the samples. Over all, standard error of predictions ranged between 2.49 and 2.88% (v/v) for the binary mixtures of olive and sunflower oil and between 1.42 and 6.38% (v/v) for the ternary mixtures of olive, sunflower and corn oil.Article Citation - WoS: 8Citation - Scopus: 8Genetic Multivariate Calibration for Near Infrared Spectroscopic Determination of Protein, Moisture, Dry Mass, Hardness and Other Residues of Wheat(John Wiley and Sons Inc., 2006) Özdemir, DurmuşDetermination of wheat flour quality parameters, such as protein, moisture, dry mass by wet chemistry analyses takes long time. Near infrared spectroscopy (NIR) coupled with multivariate calibration offers a fast and nondestructive alternative to obtain reliable results. However, due to the complexity of the spectra obtained from NIR, some wavelength selection is generally required to improve the predictive ability of multivariate calibration methods. In this study, two different wheat data sets are investigated with the aim of establishing successful calibration models using NIR spectra of wheat samples. The first data set (material 1) was obtained from the ftp address (ftp://ftp.clarkson.edu/pub/hopkepk/Chemdata/) and contained 100 NIR spectra of wheat of which wet chemical analysis of protein and moisture content were done with reference methods. The second data set (material 2) contained 176 spectra and was downloaded from http://www.spectroscopynow.com/Spy/basehtml/SpyH/1,1181, 2-1-2-0-0-newsdetail-0-74,00.html. This wheat data set was given with the quality parameters, such as protein content, moisture content, other residues, dry mass, protein content in dry mass and hardness that were determined previously. Multivariate calibration models generated with genetic inverse least squares method demonstrated very good prediction results for the parameter mentioned here. Overall, the average per cent recoveries (APR) ranged between 99.23% and 100.34% with a standard deviation (SD) ranging from 0.34 to 3.15 for all the parameters investigated, except hardness. The APR value of hardness was 103.32 with the SD of 14.97.Article Citation - WoS: 21Citation - Scopus: 23Determination of Octane Number of Gasoline Using Near Infrared Spectroscopy and Genetic Multivariate Calibration Methods(Taylor and Francis Ltd., 2005) Özdemir, DurmuşThe feasibility of rating the octane number of gasoline using near infrared (NIR) spectroscopy and three different genetic algorithm-based multivariate calibration methods was demonstrated. The three genetic multivariate calibration methods are genetic regression (GR), genetic classical least squares (GCLS), and genetic inverse least squares (GILS). The sample data set was obtained from the ftp address (ftp://ftp.clarkson.edu/pub/hopkepk/Chemdata/) with the permission of Professor. J. H. Kalivas. This data set contains the NIR spectra of 60 gasoline samples collected using diffuse reflectance as log (I / R) with known octane numbers and covers the range from 900 to 1700 nm in 2 nm intervals. Of these 60 spectra, 20 were used as the calibration set, 20 were used as the prediction set, and 20 were reserved for the validation purposes. Several calibration models were built with the three genetic algorithm-based methods, and the results were compared with the partial least squares (PLS) prediction errors reported in the literature. Overall, the standard error of calibration (SEC), standard error of prediction (SEP), and standard error of validation (SEV) values were in the range of 0.15-0.32 (in the units of motor octane number) for the GR and GILS, which are comparable with the literature. However, GCLS produced relatively large results (0.36 for SEC, 0.39 for SEP and 0.52 for SEV) when compared with the other two methods.Article Citation - WoS: 3Citation - Scopus: 11Determination of Benazepril Hcl and Hydrochlorothiazide in Pharmaceutical Preparations Using Uv-Visible Spectrophotometry and Genetic Multivariate Calibration Method(Taiwan Food and Drug Administration, 2005) Özdemir, Durmuş; Dinç, ErdalSimultaneous determination of binary mixtures of benazepril and hydrochlorothiazide in pharmaceutical tablets using UV-visible spectrophotometry, classical least squares (CLS) and three genetic algorithms (GA) based multivariate calibration methods was demonstrated. The three genetic multivariate calibration methods are Genetic Classical Least Squares (GCLS), Genetic Inverse Least Squares (GILS) and Genetic Regression (GR). The sample data set contains the UV- spectra of 28 synthetic mixtures of benazepril (12∼36 μg/mL) and hydrochlorothiazide (10∼22 μg/mL) and 16 tablets containing both compounds. The spectra cover the range from 210 to 360 nm in 0.1 mn intervals. Several calibration models were built with the four methods. The root mean square error of calibration (RMSEC) and validation (RMSEV) for the synthetic data were in the range of 0.19 and 0.34 μg/mL for all the genetic algorithm based methods. The root mean square error of Prediction (RMSEP) values for the tablets were in the range of 0.04∼0.20 mg/tablets. A comparison of genetic algorithm selected wavelengths for each component was also included.