Yazar "Yalpir, Sukran" seçeneğine göre listele

Listeleniyor 1 - 5 / 5
  • Küçük Resim Yok
    Öğe
    Adaptive network fuzzy inference system modeling for the input selection and prediction of anaerobic digestion effluent quality
    (ELSEVIER SCIENCE INC, 2011) Erdirencelebi, Dilek; Yalpir, Sukran
    This paper presents the development and evaluation of three adaptive network fuzzy inference system (ANFIS) models for a laboratory scale anaerobic digestion system outputs with varied input selection approaches. The aim was the investigation of feasibility of the approach-based-control system for the prediction of effluent quality from a sequential upflow anaerobic sludge bed reactor (UASBR) system that produced a strong nonlinearship between its inputs and outputs. As ANFIS demonstrated its ability to construct any nonlinear function with multiple inputs and outputs in many applications, its estimating performance was investigated for a complex wastewater treatment process at increasing organic loading rates from 1.1 to 5.5 g COD/L d. Approximation of the ANFIS models was validated using correlation coefficient. MAPE and RMSE. ANFIS was successful to model unsteady data for pH and acceptable for COD within anaerobic digestion limits with multiple input structure. The prediction performance showed a high feasibility of the model-based-control system on the anaerobic digester system to produce an effluent amenable for a consecutive aerobic treatment unit. (C) 2011 Elsevier Inc. All rights reserved.
  • Küçük Resim Yok
    Öğe
    Creating A Valuation Map In GIS Through Artificial Neural Network Methodology: A Case Study
    (BERG FAC TECHNICAL UNIV KOSICE, 2014) Yalpir, Sukran; Durduran, S. Savas; Unel, Fatma Bunyan; Yolcu, Melisa
    The present study compared models and market values by creating a model for valuation estimations with artificial neural networks (ANN), which is one of the most advanced methods in immovable valuation and making applications by using Multiple Regression Analysis (MRA). Estimation results obtained from the models with the help of geographical information systems (GIS) were analyzed for spatial analysis. Data sets consisted of the criteria as the values of 300 residential real estates in Bosna Hersek Neighborhood in Konya/Selcuklu Region, the ages that affect those values, the number of stories, the frontage of the apartments, the number of rooms, and the distance to social infrastructures. GIS based value maps, which were integrated to ArcGIS10.0 software, were produced in order to create an algorithm via ANN, which is one Pile modern immovable valuation methods, and MRA, which is used to determine the immovable valuations according to those results. Spatial information about parcel and buildings was transferred to the database of ArcGIS 10.0 software and then was associated with oral data. Value maps were obtained through integrating the values of immovable properties in market conditions and estimation values of ANN and MRA to the map of the studied area. The performances were calculated and the produced value maps were compared in order to ensure the success of the models.
  • Küçük Resim Yok
    Öğe
    Enhancement of parcel valuation with adaptive artificial neural network modeling
    (SPRINGER, 2018) Yalpir, Sukran
    This study targets a research on the application of artificial neural network (ANN) and multiple regression analysis (MRA) approaches in Geomatics Engineering science to land valuation process. The prediction capability was investigated and evaluated using three ANN models constructed with different activation functions (sigmoid, tangent hyperbolic and adaptive activation function) and MRA was used as a reference approach. These four methodologies were applied to land valuation in order to model the unit market value with various inputs based on essential criteria. All approaches were investigated with their estimation level in training and testing data. It was observed that adaptive ANN performed noticeably higher predicting the values with the highest accuracy and giving the smallest RMSE value in validation process, although other methodologies approximated to the raw data at a promising level for further valuation-based applications.
  • Küçük Resim Yok
    Öğe
    KNOWLEDGE-BASED FIS AND ANFIS MODELS DEVELOPMENT AND COMPARISON FOR RESIDENTIAL REAL ESTATE VALUATION
    (VILNIUS GEDIMINAS TECH UNIV, 2018) Yalpir, Sukran; Ozkan, Gulgun
    There has been an increasing concern on the development of alternative approaches to overcome the problems and deficiencies that occur during the application of real-estate valuation methods. This study was established to investigate the usability of the expert knowledge based fuzzy logic methodology in determining real-estates values. In addition, valuation with the Adaptive Neuro-Fuzzy Inference System (ANFIS) method provided model comparison. Samples were administered a questionnaire for the parameters planned for these models regarding the parameters that affect real estate values. To make value estimations for the Fuzzy Inference System (FIS) model by using the parameters obtained from the questionnaire analyses, the criteria that produced the best results were acquired from the various criteria alternatives. An algorithm was created and the valuation process for real estate was performed using the FIS in Konya/Turkey. As a result of poll studies the area, age, floor conditions, physical properties and location of the real-estate property were considered as the input variables and the market value as the output variable. The memberships were established with poll analysis and were rule based on expert knowledge. The model structure was formed by using the Mamdani structure in the MATLAB fuzzy toolbox. Model prediction performance was evaluated statistically with the Mean Absolute Percentage Error (MAPE) and a high accuracy of the model results to the market values indicated the reliability of the established model for residential real-estate valuation.
  • Küçük Resim Yok
    Öğe
    Prediction of primary treatment effluent parameters by Fuzzy Inference System (FIS) approach
    (ELSEVIER SCIENCE BV, 2011) Yel, Esra; Yalpir, Sukran
    A fuzzy-logic-based diagnosis system was developed to determine the primary treatment effluent quality in a municipal wastewater treatment plant (MWTP). The measured data of variables were implemented into the Fuzzy Inference System (FIS) with Mamdani's method. The fuzzy control rule base was shaped to define essential quality parameters monitored as pH, COD, BOD and SS outputs. The output approximations to real data remained in an acceptable range for a MWTP performance (89-96%). The averages and standard deviations of the model were also approximated closely as 93-98% and 89-97%, respectively. The resulting configuration proved a good modeling approach for MWTP effluent quality prediction. (C) 2010 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Guest Editor.