Bestamin Özkaya

Neural network prediction model for the methane fraction in biogas from field-scale landfill bioreactors

In this study we present a neural network model for predicting the methane fraction in landfill gas originating from field-scale landfill bioreactors. Landfill bioreactors were constructed at the Odayeri Sanitary Landfill, Istanbul, Turkey, and operated with (C2) and without (C1) leachate recirculation. The refuse height of the test cell was 5m, with a placement area of 1250m^2 (25mx50m). We monitored the leachate and landfill gas components for 34 months, after which we modeled the methane fraction in landfill gas from the bioreactors (C1 and C2) using artificial neural networks; leachate components were used as input parameters. To predict the methane fraction in landfill gas as a final product of anaerobic digestion, we used input parameters such as pH, alkalinity, Chemical Oxygen Demand, sulfate, conductivity, chloride and waste temperature. We evaluated the anaerobic conversion efficiencies based on leachate characteristics during different time periods. We determined the optimal architecture of the neural network, and advantages, disadvantages and further developments of the network are discussed.

NN-LEAP: A neural network-based model for controlling leachate flow-rate in a municipal solid waste landfill site

Abstract A method is proposed for modeling leachate flow-rate in a municipal solid waste (MSW) landfill site, based on a popular neural network – the backpropagation algorithm ( n eural n etwork-based lea chate p rediction method; NN-LEAP). After backpropagation training, the neural network model predicts flow-rates based on meteorological data. Depending on output value, relevant control strategies and actions are activated. To illustrate and validate the proposed method, a case study was carried out, based on the data obtained from the Istanbul Odayeri landfill site. As a critical model parameter (neural network outputs), daily flow-rate of leachate from the landfill site was considered. The Levenberg–Marquardt algorithm was selected as the best of 13 backpropagation algorithms. The optimal neural network architecture has been determined, and the advantages, disadvantages and further developments are discussed.

COD fractions of leachate from aerobic and anaerobic pilot scale landfill reactors

One of the most important problems with designing and maintaining a landfill is managing leachate that generated when water passes through the waste. In this study, leachate samples taken from aerobic and anaerobic landfill reactors operated with and without leachate recirculation are investigated in terms of biodegradable and non-biodegradable fractions of COD. The operation time is 600 days for anaerobic reactors and 250 days for aerobic reactors. Results of this study show that while the values of soluble inert COD to total COD in the leachate of aerobic landfill with leachate recirculation and aerobic dry reactors are determined around 40%, this rate was found around 30% in the leachate of anaerobic landfill with leachate recirculation and traditional landfill reactors. The reason for this difference is that the aerobic reactors generated much more microbial products. Because of this condition, it can be concluded that total inert COD/total COD ratios of the aerobic reactors were 60%, whereas those of anaerobic reactors were 50%. This study is important for modeling, design, and operation of landfill leachate treatment systems and determination of discharge limits.

Molecular weight distribution of a full-scale landfill leachate treatment by membrane bioreactor and nanofiltration membrane

In this study, Molecular weight (MW) distributions of a full-scale landfill leachate treatment plant consisting of membrane bioreactor (MBR) and nanofiltration (NF) membrane were investigated. The leachate was sampled from the equalization tank, and effluents of MBR and NF membrane in the landfill leachate treatment plant. Parameters of COD, TOC, TKN, NH4(+)-N and UV(254, 280 and 320) absorbance were analyzed to evaluate both the removal performance of the plant and MW distributions. MW distribution of samples were determined by ultrafiltration (UF) (100 kDa, 10 kDa, 5 kDa, 1 kDa and 500 Da) membranes. The results indicated that organic matter of one third percent is particulate or colloidal form and almost half of the organic fraction has a lower MW than 500 Da. In addition, organic matter had hydrophilic character. Most part of TKN was>500 Da with the corresponding rate of 92%. Further, UV absorbance of raw leachate (RW) decreased 85% after 500 Da.

Bioelectricity generation in continuously-fed microbial fuel cell: Effects of anode electrode material and hydraulic retention time

The main aim of this study is to investigate the bioelectricity production in continuously-fed dual chambered microbial fuel cell (MFC). Initially, MFC was operated with different anode electrode material at constant hydraulic retention time (HRT) of 2d to evaluate the effect of electrode material on electricity production. Pt electrode yielded about 642 mW/m(2) power density, which was 4 times higher than that of the MFC with the mixed metal oxide titanium (Ti-TiO2). Further, MFC equipped with Pt electrode was operated at varying HRT (2-0.5d). The power density generation increased with decreasing HRT, corresponding to 1313 mW/m(2) which was maximum value obtained during this study. Additionally, decreasing HRT from 2 to 0.5d resulted in increasing effluent dissolved organic carbon (DOC) concentration from 1.92 g/L to 2.23 g/L, corresponding to DOC removal efficiencies of 46% and 38%, respectively.

ARTIFICIAL INTELLIGENCE-BASED PREDICTION MODELS FOR ENVIRONMENTAL ENGINEERING

A literature survey was conducted to appraise the recent applications of artiflcal intelligence (AI)-based modeling studies in the environmental engineer- ing fleld. A number of studies on artiflcial neural networks (ANN), fuzzy logic and adaptive neuro-fuzzy systems (ANFIS) were reviewed and important aspects of these models were highlighted. The results of the extensive literature survey showed that most AI-based prediction models were implemented for the solution of water/wastewater (55.7%) and air pollution (30.8%) related environmental prob- lems compared to solid waste (13.5%) management studies. The present literature review indicated that among the many types of ANNs, the three-layer feed-forward and back-propagation (FFBP) networks were considered as one of the simplest and the most widely used network type. In general, the Levenberg-Marquardt algo- rithm (LMA) was found as the best-suited training algorithm for several complex and nonlinear real-life problems of environmental engineering. The literature sur- vey showed that for water and wastewater treatment processes, most of AI-based prediction models were introduced to estimate the performance of various biolog- ical and chemical treatment processes, and to control e†uent pollutant loads and ∞owrates from a speciflc system. In air polution related environmental problems, forecasting of ozone (O3) and nitrogen dioxide (NO2) levels, daily and/or hourly particulate matter (PM2:5 and PM10) emissions, and sulfur dioxide (SO2) and car- bon monoxide (CO) concentrations were found to be widely modeled. For solid waste management applications, reseachers conducted studies to model weight of waste generation, solid waste composition, and total rate of waste generation.

Effect of leachate recirculation on refuse decomposition rates at landfill site: a case study

In this study, a comparison of methane (CH4) generation rates for two test cells, one operated with (enhanced) and another without leachate recirculation at Odayeri Sanitary Landfill were compared using their bio-chemical methane potentials (BMP). Initial methane potential is approximately 34.5 m³ CH4/wet ton of solid waste. The remaining methane potential for the control (C1) and the enhanced (C2) cells are 32.6 m³ CH4/wet ton and 31.1 m³ CH4/wet ton of refuse after eight months of operation, respectively. The produced CH4 quantities for C1 and C2 after eight months of operation are 1.9 m³ CH4/wet ton and 3.4 m³ CH4/wet ton, respectively. On the other hand, 5.5% and 9.9% of the total potential are generated in eight months. However, the CH4 generation rates for the first year are determined as 2.85 and 5.10 m³/ton/year for C1 and C2 test cells, respectively. Due to the appropriate conditions such as moisture content, solid waste decomposition rate is enhanced at a rate of 79% at C2 test cell relative to C1 test cell. Hence, C2 test cell shows more decomposition relative to the C1 test cell.

Arsenic removal from acidic solutions with biogenic ferric precipitates

Treatment of acidic solution containing 5g/L of Fe(II) and 10mg/L of As(III) was studied in a system consisting of a biological fluidized-bed reactor (FBR) for iron oxidation, and a gravity settler for iron precipitation and separation of the ferric precipitates. At pH 3.0 and FBR retention time of 5.7h, 96-98% of the added Fe(II) precipitated (99.1% of which was jarosite). The highest iron oxidation and precipitation rates were 1070 and 28mg/L/h, respectively, and were achieved at pH 3.0. Subsequently, the effect of pH on arsenic removal through sorption and/or co-precipitation was examined by gradually decreasing solution pH from 3.0 to 1.6 (feed pH). At pH 3.0, 2.4 and 1.6, the highest arsenic removal efficiencies obtained were 99.5%, 80.1% and 7.1%, respectively. As the system had ferric precipitates in excess, decreased arsenic removal was likely due to reduced co-precipitation at pH<2.4. As(III) was partially oxidized to As(V) in the system. In shake flask experiments, As(V) sorbed onto jarosite better than As(III). Moreover, the sorption capacity of biogenic jarosite was significantly higher than that of synthetic jarosite. The developed bioprocess simultaneously and efficiently removes iron and arsenic from acidic solutions, indicating potential for mining wastewater treatment.

Bio-reduction of tetrachloroethen using a H2-based membrane biofilm reactor and community fingerprinting.

Chlorinated ethenes in drinking water could be reductively dechlorinated to non-toxic ethene by using a hydrogen based membrane biofilm reactor (H2-MBfR) under denitrifying conditions as it provides an appropriate environment for dechlorinating bacteria in biofilm communities. This study evaluates the reductive dechlorination of perchloroethene (PCE) to non-toxic ethene (ETH) and comparative community analysis of the biofilm grown on the gas permeable membrane fibers. For these purposes, three H2-MBfRs receiving three different chlorinated ethenes (PCE, TCE and DCE) were operated under different hydraulic retention times (HRTs) and H2 pressures. Among these reactors, the H2-MBfR fed with PCE (H2-MBfR 1) accomplished a complete dechlorination, whereas cis-DCE accumulated in the TCE receiving H2-MBfR 2 and no dechlorination was detected in the DCE receiving H2-MBfR 3. The results showed that 95% of PCE dechlorinated to ETH together with over 99.8% dechlorination efficiency. Nitrate was the preferred electron acceptor as the most of electrons generated from H2 oxidation used for denitrification and dechlorination started under nitrate deficient conditions at increased H2 pressures. PCR-DGGE analysis showed that Dehalococcoides were present in autotrophic biofilm community dechlorinating PCE to ethene, and RDase genes analysis revealed that pceA, tceA, bvcA and vcrA, responsible for complete dechlorination step, were available in Dehalococcoides strains.

Electricity production by a microbial fuel cell fueled by brewery wastewater and the factors in its membrane deterioration

Electricity production from brewery wastewater using dual-chamb er microbial fuel cells (MFCs) with a tin-coated copper mesh in the anode was investigated by changing the hydraulic retention time (HRT). The MFCs were fed with wastewater samples from the inlet (inflow, MFC-1) and outlet (outflow, MFC-2) of an anaerobic digester of a brewery wastewater treatment plant. Both chemical oxygen demand removal and current density were improved by decreasing HRT. The best MFC performance was with an HRT of 0.5 d. The maximum power densities of 8.001 and 1.843 µW/cm2 were obtained from reactors MFC-1 and MFC-2, respectively. Microbial diversity at different conditions was studied using PCR-DGGE profiling of 16S rRNA fragments of the microorganisms from the biofilm on the anode electrode. The MFC reactor had mainly Geobacter, Shewanella, and Clostridium species, and some bacteria were easily washed out at lower HRTs. The fouling characteristics of the MFC Nafion membrane and the resulting degradation of MFC performance were examined. The ion exchange capacity, conductivity, and diffusivity of the membrane decreased significantly after fouling. The morphology of the Nafion membrane and MFC degradation were studied using scanning electron microscopy and attenuated total reflection-Fourier transform infrared spectroscopy.