Browsing by Author "Nacar, Sinan"

Now showing 1 - 14 of 14

A downscaling application for local meteorological variables of Eastern Black Sea Basin and scenario based predictions
(Turkish Chamber Civil Engineers, 2022-11-01) Nacar, Sinan; Okkan, Umut; KANKAL, MURAT; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0003-0897-4742; AAZ-6851-2020
Climate change has become one of the most important problems discussed around the world due to its environmental, economic and social impacts. In order to determine the magnitude of the impact of climate change and possible adaptation studies, it is necessary to determine the changes in the future periods of temperature and precipitation being the most important variables of the climate. For this purpose, scenario outputs of general circulation models (GCM) with coarse spatial resolution are used. The low resolution of these outputs limits their direct use in determining the effects of climate change on a local scale. Therefore, GCM outputs should be downscaled into finer scale. The aim of this study is to determine the possible effects of climate change on precipitation and temperature values of Eastern Black Sea Basin (EBSB), which is one of the most important hydrological basin in Turkey. For this purpose, the coarse resolution outputs of the GFDL-ESM2M model under RCP4.5 and RCP8.5 scenarios were reduced to local scale using the multivariate adaptive regression splines (MARS) and classical regression analysis (CRA) methods. Various performance statistics were used to compare the downscaling capabilities of MARS and CRA based models, and the method with the highest performance was determined according to these statistics. Within the scope of the study, the monthly average temperature and total precipitation values for the next period (2021-2050, 2051-2080, 2081-2100) of 12 meteorology stations located in and around the basin were produced by using MARS-based models that give the best performance values. Mann-Kendall trend analysis was also applied to the scenario data obtained. According to the results, it is expected that the temperature values in the southern part of the basin with terrestrial climate characteristics (Bayburt, Gumushane, Susehri ve Sebinkarahisar stations) will increase by an average of 1 degrees C according to the RCP4.5 scenario and 1.5 degrees C according to the RCP8.5 scenario in the period 2021-2050. In addition, temperature increases of up to 2.5 degrees C are foreseen on the coastline of the basin where the Pazar, Rize and Hopa stations are located. According to the outputs of both scenarios, is expected in long-term precipitation average values in almost all of the basin. In the periods of 2051-2080 and 2081-2100, it is foreseen that the increases and decreases in temperature and precipitation values will be more than the 2021-2050 period. According to the results of the trend analysis, the RCP4.5 scenario for temperature and precipitation, no trend is expected in the future, while according to the RCP8.5 scenario, an increase trend for temperature and a decrease trend for precipitation have been determined.
Akarsularda çözünmüş oksijen konsantrasyonunun regresyon tabanlı yöntemlerle modellenmesi: Harşit Çayı örneği
(Bursa Uludağ Üniversitesi, 2022-02-22) Nacar, Sinan; Baki, Osman Tuğrul; Bayram, Adem
Bu çalışmada çok değişkenli uyarlanabilir regresyon eğrileri (MARS) ve TreeNet gradyan arttırma makinesi (TreeNet) isimli regresyon tabanlı yöntemler kullanılarak çözünmüş oksijen (ÇO) konsantrasyonu modellemesi amaçlanmıştır. Modelleme çalışmasında kentsel atıksuları bünyesine alarak yer yer kirlenen Harşit Çayı (Gümüşhane) üzerinde belirlenmiş altı su kalitesi gözlem istasyonunda, 15 gün aralıklarla ve 24 kez yerinde gerçekleştirilen ÇO konsantrasyonu (mg/L), sıcaklık (°C), pH ve elektriksel iletkenlik (mS/cm) ölçümleri yanı sıra akarsudan alınan su örneklerinde laboratuvarda gerçekleştirilen sertlik (°dH) tayinleri neticesinde elde edilen veriler kullanılmıştır. Elde edilen veri setinin % 80’i kurulan modellerin eğitilmesinde geriye kalan % 20’si ise söz konusu modellerin test edilmesinde kullanılmıştır. Kurulan modellerin eğitim ve test veri seti performanslarını değerlendirmek amacıyla ortalama karesel hatanın karekökü (OKHK), ortalama mutlak hata (OMH), ortalama rölatif hata (ORH) ve determinasyon katsayısı (R2 ) performans istatistikleri kullanılmıştır. En düşük OKHK, OMH ve ORH ile en yüksek R2 değerleri eğitim veri seti için sırasıyla 0,2247 mg/L, 0,0666 mg/L, % 0,66 ve 0,9995 olarak TreeNet yönteminden, test veri seti için ise 0,2911 mg/L, 0,2336 mg/L, % 2,27 ve 0,9992 olarak MARS yönteminden elde edilmiştir. Her iki veri seti için ortalamalar dikkate alındığında ise, MARS yönteminden elde edilen performans değerlerinin TreeNet yönteminden elde edilenlere kıyasla daha iyi olduğu sonucuna ulaşılmıştır.
Annual trends of precipitation and temperature in the northwestern part of Turkey using innovative approaches: A holistic and partial study
(Springer Basel Ag, 2023-07-03) San, Murat; Nacar, Sinan; Akçay, Fatma; Kankal, Murat; KANKAL, MURAT; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0003-0897-4742; AAZ-6851-2020
Trend analysis benefits collecting and analyzing reliable data in climate change studies. In this context, long-term temperature and precipitation data analysis, two variables sensitive to climate change, is essential. This study aims to holistically and partially determine the annual trends of precipitation and temperature time series for 1970-2019 at nine selected stations in the Susurluk Basin, Turkey. The innovative trend significance test (ITST), innovative crossing trend analysis (ICTA) method, and Mann-Kendall (MK) test were used to determine holistic trends. Also, partial trends were determined using successive average methodology (SAM). For the precipitation variable, while an increasing trend was determined for the ITST, there was mostly no trend for the other methods. While a strong increasing trend was detected for the temperature according to the ITST and MK methods, no trend was observed in any station according to the ICTA method. According to the SAM results, the maximum trend durations for the peak and trough change points were 4.9 (10.3) and 5.3 (8.4) years, respectively, for precipitation (temperature). The strong temperature trends in the basin will likely continue, requiring precautions against extreme events such as drought.
Daily precipitation performances of regression-based statistical downscaling models in a basin with mountain and semi-arid climates
(Springer, 2023-04-01) Şan, Murat; Nacar, Sinan; Bayram, Adem; Kankal, Murat; KANKAL, MURAT; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü; 0000-0003-0897-4742; AAC-6221-2021
The impacts of climate change on current and future water resources are important to study local scale. This study aims to investigate the prediction performances of daily precipitation using five regression-based statistical downscaling models (RBSDMs), for the first time, and the ERA-5 reanalysis dataset in the Susurluk Basin with mountain and semi-arid climates for 1979-2018. In addition, comparisons were also performed with an artificial neural network (ANN). Before achieving the aim, the effects of atmospheric variables, grid resolution, and long-distance grid on precipitation prediction were holistically investigated for the first time. Kling-Gupta efficiency was modified and used for holistic evaluation of statistical moments parameters at precipitation prediction comparison. The standard triangular diagram, quite new in the literature, was also modified and used for graphical evaluation. The results of the study revealed that near grids were more effective on precipitation than single or far grids, and 1.50 degrees x 1.50 degrees resolution showed similar performance to 0.25 degrees x 0.25 degrees resolution. When the polynomial multivariate adaptive regression splines model, which performed slightly higher than ANN, tended to capture skewness and standard deviation values of precipitations and to hit wet/dry occurrence than the other models, all models were quite well able to predict the mean value of precipitations. Therefore, RBSDMs can be used in different basins instead of black-box models. RBSDMs can also be established for mean precipitation values without dry/wet classification in the basin. A certain success was observed in the models; however, it was justified that bias correction was required to capture extreme values in the basin.
Estimates of hydroelectric generation using neural networks with the artificial bee colony algorithm for Turkey
(Pergamon-Elsevier, 2014-05-01) Uzlu, Ergun; Öztürk, Hasan; Nacar, Sinan; Kankal, Murat; Akpınar, Adem; Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0002-9042-6851; AAC-6763-2019; 23026855400
The primary objective of this study was to apply the ANN (artificial neural network) model with the ABC (artificial bee colony) algorithm to estimate annual hydraulic energy production of Turkey. GEED (gross electricity energy demand), population, AYT (average yearly temperature), and energy consumption were selected as independent variables in the model. The first part of the study compared ANN-ABC model performance with results of classical ANN models trained with the BP (back propagation) algorithm. Mean square and relative error were applied to evaluate model accuracy. The test set errors emphasized positive differences between the ANN-ABC and classical ANN models. After determining optimal configurations, three different scenarios were developed to predict future hydropower generation values for Turkey. Results showed the ANN-ABC method predicted hydroelectric generation better than the classical ANN trained with the BP algorithm. Furthermore, results indicated future hydroelectric generation in Turkey will range from 69.1 to 76.5 TWh in 2021, and the total annual electricity demand represented by hydropower supply rates will range from 14.8% to 18.0%. However, according to Vision 2023 agenda goals, the country plans to produce 30% of its electricity demand from renewable energy sources by 2023, and use 20% less energy than in 2010. This percentage renewable energy provision cannot be accomplished unless changes in energy policy and investments are not addressed and implemented. In order to achieve this goal, the Turkish government must reconsider and raise its own investments in hydropower, wind, solar, and geothermal energy, particularly hydropower.
Estimating suspended sediment load with multivariate adaptive regression spline, teaching-learning based optimization, and artificial bee colony models
(Elsevier, 2018-10-15) Yılmaz, Banu; Aras, Egemen; Nacar, Sinan; Kartal, Murat; Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0003-0897-4742; AAZ-6851-2020; 24471611900
The functional life of a dam is often determined by the rate of sediment delivery to its reservoir. Therefore, an accurate estimate of the sediment load in rivers with dams is essential for designing and predicting a dam's useful lifespan. The most credible method is direct measurements of sediment input, but this can be very costly and it cannot always be implemented at all gauging stations. In this study, we tested various regression models to estimate suspended sediment load (SSL) at two gauging stations on the Coruh River in Turkey, including artificial bee colony (ABC), teaching-learning-based optimization algorithm(TLBO), and multivariate adaptive regression splines (MARS). These models were also compared with one another and with classical regression analyses (CRA). Streamflow values and previously collected data of SSL were used as model inputs with predicted SSL data as output. Two different training and testing dataset configurations were used to reinforce the model accuracy. For the MARS method, the root mean square error value was found to range between 35% and 39% for the test two gauging stations, which was lower than errors for other models. Error values were even lower (7% to 15%) using another dataset. Our results indicate that simultaneous measurements of stream flow with SSL provide the most effective parameter for obtaining accurate predictive models and that MARS is the most accurate model for predicting SSL. (C) 2018 Elsevier B.V. All rights reserved.
Evaluation of the suitability of ncep/ncar, era-interim and, era5 reanalysis data sets for statistical downscaling in the eastern black sea basin, Turkey
(Springer Wien, 2022-04-01) Nacar, Sinan; Okkan, Umut; Kankal, Murat; KANKAL, MURAT; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0003-0897-4742; AAZ-6851-2020
Climate community frequently uses gridded reanalysis data sets in their climate change impact studies. However, these studies for a region yield more realistic results depending on the rigorous analysis of the reanalysis data sets for this region. This study aims to determine the most suitable reanalysis data set for the statistical downscaling method in the Eastern Black Sea Basin, one of Turkey's most important hydrological basins owing to the precipitation it receives throughout the year. For this purpose, the monthly mean temperature and total precipitation data measured from the 12 meteorological stations and 12 large-scale predictors of the NCEP/NCAR, ERA-Interim, and ERA5 reanalysis data sets were used. The multivariate adaptive regression splines (MARS) and conventional regression analysis with linear and exponential functions were used to create effective statistical downscaling models. For evaluating and comparing the performance of the downscaling models with three different reanalysis data set, four performance statistics (root means square error, scatter index, mean absolute error, and the Nash Sutcliffe coefficient of efficiency) were used. Besides, the relative importance of the input variables of the models was determined. The study revealed that the values obtained from the models of ERA5 were closer to the precipitation and temperature values measured from the meteorological stations. In addition, the model performances with three reanalysis data sets for the temperature variable were very close to each other. The study results have shown that the MARS method, which gives the highest performance values, can be used successfully as a statistical downscaling method in climate change impact studies.
Günlük çözünmüş oksijen konsantrasyonunun çok değişkenli uyarlanabilir regresyon eğrileri ile tahmin edilmesi
(Bursa Uludağ Üniversitesi, 2020-12-04) Nacar, Sinan; Mete, Betül; Bayram, Adem
Bu çalışmada su sıcaklığı (T), özgül iletkenlik (Öİ) verilerinden hesaplanmış elektriksel iletkenlik (Eİ), pH ve debi (Q) verileri kullanılarak çok değişkenli uyarlanabilir regresyon eğrileri (MARS) ve regresyon analizi (RA) yöntemleri ile ÇO konsnatrasyonunun tahmin edilmesi amaçlanmıştır. MARS yönteminde en iyi tahmin değerlerini üreten temel fonksiyonlar ve denklemler belirlenmiş, RA yöntemi doğrusal, üs, üstel ve kuadratik olmak üzere dört farklı fonksiyona uygulanmış ve bu fonksiyonlara ait katsayılar hesaplanmıştır. Modelleme çalışmalarında Amerika Birleşik Devletleri’nin Oregon eyaletinin kuzey batısında yer alan Willamette Nehri’nin yan kollarından biri olan ve yaklaşık 2435 km2 ’lik bir havza alanına sahip Clackamas Nehri’ne ait Eylül 2016 − Ağustos 2017 dönemi günlük ortalama verileri kullanılmıştır. Her bir su kalitesi değişkeninin ÇO konsantrasyonu tahmin performansına etkisini belirlemek amacıyla sekiz farklı model oluşturulmuştur. ÇO konsantrasyonu tahmininde kurulan modellerin ve kullanılan yöntemlerin performanslarının değerlendirilebilmesi için çeşitli istatistikler (ortalama karesel hatanın karekökü, ortalama mutlak hata, saçılım indeksi ve Nash Sutcliffe verimlilik katsayısı) kullanılmıştır. Modelleme çalışmalarından elde edilen sonuçlar irdelendiğinde, MARS yönteminin RA yönteminden daha iyi sonuçlar verdiği anlaşılmıştır. Regresyon fonksiyonları içerisinden ise en başarılı tahmin sonuçlarının kuadratik fonksiyondan elde edildikleri ve MARS yöntemi ile elde edilen değerlere de oldukça yakın oldukları görülmüştür. ÇO konsantrasyonu tahmininde en etkili değişkenlerin T ve Q oldukları dolayısıyla en etkisiz değişkenlerin ise Eİ ve pH oldukları anlaşılmıştır. Model 3, Model 5, Model 7 ve Model 8’den elde edilen sonuçların birbirine çok yakın olması sebebiyle daha az değişken ile güçlü tahminler yapması ve daha sade bir model olması bakımından ÇO tahmininde Model 3’ün kullanılmasının daha avantajlı olacağı sonucuna varılmıştır.
Performance evaluation of multiple adaptive regression splines, teaching–learning based optimization and conventional regression techniques in predicting mechanical properties of impregnated wood
(Springer, 2019-07) Tiryaki, Sebahattin; Tan, Hüseyin; Bardak, Selahattin; Nacar, Sinan; Peker, Hüseyin; Kankal, Murat; Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0003-0897-4742; AAZ-6851-2020; 24471611900
Understanding the mechanical behaviour of impregnated wood is crucial in making a preliminary decision on the usability of such woods for structural purposes. In this paper, by considering concentration (1, 3 and 5%), pressure (1, 1.5 and 2atm.), and time (30, 60, 90 and 120min), an experimental study was performed, and the mechanical behaviour of impregnated wood was determined as a result of the experimental process. Multiple adaptive regression splines (MARS), teaching-learning based optimization (TLBO) algorithms and conventional regression analysis (CRA) were applied to different regression functions by using experimentally obtained data. The functions were checked against each other to detect the best equation for each parameter and to assess performances of MARS, TLBO and CRA methods in the prediction of mechanical properties. The experimental results showed that higher values of mechanical properties were obtained when lower concentration, pressure and time were chosen. Overall, all the functions successfully predicted the mechanical properties. However, the MARS and TLBO provided better accuracy in predicting the mechanical properties. The modeling results indicated that the MARS and TLBO are promising new methods in predicting the mechanical properties of impregnated wood. With the use of these methods, the mechanical behavior of impregnated wood could be determined with high levels of accuracy. Thus, the proposed methods may facilitate a preliminary decision concerning the usability of such woods for areas where the mechanical properties are important. Finally, the employment of MARS and TLBO algorithms by practitioners in the wood industry is encouraged and recommended for future studies.
Prediction of maximum annual flood discharges using artificial neural network approaches
(Croatian Society of Civil Engineers, 2020-04-10) Anılan, Tuğçe; Nacar, Sinan; Yüksek, Ömer; Kankal, Murat; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0003-0897-4742; AAZ-6851-2020; 24471611900
The applicability of artificial neural network (ANN) approaches for estimation of maximum annual flows is investigated in the paper. The performance of three neural network models is compared: multi layer perceptron neural networks (MLP_NN), generalized feed forward neural networks (GFF_NN), and principal component analysis with neural networks (PCA_ NN). The proposed approaches were applied to 33 stream-gauging stations. It was found that the optimal 3-hidden layered PCA_NN method was more appropriate than the optimal MLP_NN and GFF_NN models for the estimation of maximum annual flows.
Prediction of parameters which affect beach nourishment performance using MARS, TLBO, and conventional regression techniques
(Springer, 2019-08-09) Karasu, Servet; Nacar, Sinan; Uzlu, Ergun; Yüksek, Ömer; Kankal, Murat; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği.; 0000-0003-0897-4742; AAZ-6851-2020; 24471611900
Artificial beach nourishment is one of the most important environmentally friendly coastal protection methods since it protects the aesthetic and recreational values of the beach and increases its protective properties. Therefore, the main aim of the current study is to assess the accuracy of multivariate adaptive regression splines (MARS) in predicting the parameters, namely sediment transport coefficients (K) and the diffusion rate (omega), which affect beach nourishment performance. The performance of the MARS was determined by comparison of the models using exponential, linear, and power regression equations trained by conventional regression analyses (CRA) and the teaching-learning based optimization (TLBO) algorithm. In all models, two different input data obtained from the experimental study were used, one dimensional and one non-dimensional. The results presented that the MARS models gave lower error values than the CRA and TLBO models according to the root mean square error, mean absolute error, and scattering index criteria. When the models were evaluated, it was revealed that dimensional and non-dimensional models gave approximate results. We proved that the dimensional and non-dimensional MARS models can be used to estimate the (K) and (omega) values.
Prediction of suspended sediment loading by means of hybrid artificial intelligence approaches
(Springer International Publishing Ag, 2019-12-01) Yılmaz, Banu; Aras, Egemen; Nacar, Sinan; Kankal, Murat; KANKAL, MURAT; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi; 0000-0003-0897-4742; AAZ-6851-2020
The main aim of the research is to use the artificial neural network (ANN) model with the artificial bee colony (ABC) and teaching-learning-based optimization (TLBO) algorithms for estimating suspended sediment loading. The stream flow per month and SSL data obtained from two stations, Inanli and Altinsu, in Coruh River Basin of Turkey were taken as precedent. While stream flow and previous SSL were used as input parameters, only SSL data were used as output parameters for all models. The successes of the ANN-ABC and ANN-TLBO models that were developed in the research were contrasted with performance of conventional ANN model trained by BP (back-propagation). In addition to these algorithms, linear regression method was applied and compared with others. Root-mean-square and mean absolute error were used as success assessing criteria for model accuracy. When the overall situation is evaluated according to errors of the testing datasets, it was found that ANN-ABC and ANN-TLBO algorithms are more outstanding than conventional ANN model trained by BP.
Spatial forecasting of dissolved oxygen concentration in the Eastern Black Sea Basin, Turkey
(MDPI, 2020-03-24) Nacar, Sinan; Bayram, Adem; Baki, Osman Tuğrul; Aras, Egemen; Kankal, Murat; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; AAZ-6851-2020; 24471611900
The aim of this study was to model, as well as monitor and assess the surface water quality in the Eastern Black Sea (EBS) Basin stream, Turkey. The water-quality indicators monitored monthly for the seven streams were water temperature (WT), pH, total dissolved solids (TDS), and electrical conductivity (EC), as well as luminescent dissolved oxygen (LDO) concentration and saturation. Based on an 18-month data monitoring, the surface water quality variation was spatially and temporally evaluated with reference to the Turkish Surface Water Quality Regulation. First, the teaching learning based optimization (TLBO) algorithm and conventional regression analysis (CRA) were applied to three different regression forms, i.e., exponential, power, and linear functions, to predict LDO concentrations. Then, the multivariate adaptive regression splines (MARS) method was employed and three performance measures, namely, mean absolute error (MAE), root means square error (RMSE), and Nash Sutcliffe coefficient of efficiency (NSCE) were used to evaluate the performances of the MARS, TLBO, and CRA methods. The monitoring results revealed that all streams showed the same trend in that lower WT values in the winter months resulted in higher LDO concentrations, while higher WT values in summer led to lower LDO concentrations. Similarly, autumn, which presented the higher TDS concentrations brought about higher EC values, while spring, which presented the lower TDS concentrations gave rise to lower EC values. It was concluded that the water quality of the streams in the EBS basin was high-quality water in terms of the parameters monitored in situ, of which the LDO concentration varied from 9.13 to 10.12 mg/L in summer and from 12.31 to 13.26 mg/L in winter. When the prediction accuracies of the three models were compared, it was seen that the MARS method provided more successful results than the other methods. The results of the TLBO and the CRA methods were very close to each other. The RMSE, MAE, and NSCE values were 0.2599 mg/L, 0.2125 mg/L, and 0.9645, respectively, for the best MARS model, while these values were 0.4167 mg/L, 0.3068 mg/L, and 0.9086, respectively, for the best TLBO and CRA models. In general, the LDO concentration could be successfully predicted using the MARS method with various input combinations of WT, EC, and pH variables.
Suspended sediment load prediction in rivers by using heuristic regression and hybrid artificial intelligence models
(Yıldız Teknik Üniversitesi, 2020-06-01) Yılmaz, Banu; Aras, Egemen; Kankal, Murat; Nacar, Sinan; KANKAL, MURAT; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/İnşaat Mühendisliği Bölümü.; 0000-0003-0897-4742; AAZ-6851-2020
Accurate prediction of amount of sediment load in rivers is extremely important for river hydraulics. The solution of the problem has been become complicated since the explanation of hydraulic phenomenon between the flow and the sediment on the river is dependent many parameters. The usage of different regression methods and artificial intelligence techniques allows the development of predictions as the traditional methods do not give enough accurate results. In this study, data of the flow and suspended sediment load (SSL) obtained from Karsikoy Gauging Station, located on Coruh River in the north-eastern of Turkey, modelled with different regression methods (multiple regression, multivariate adaptive regression splines) and artificial neural network (ANN) (ANN-back propagation, ANN teaching-learning-based optimization algorithm and ANN-artificial bee colony). When the results were evaluated, it was seen that the models of ANN method were close to each other and gave better results than the regression models. It is concluded that these models of ANN method can be used successfully in estimating the SSL.