Mirjana Ristić

PM10 emission forecasting using artificial neural networks and genetic algorithm input variable optimization

This paper describes the development of an artificial neural network (ANN) model for the forecasting of annual PM(10) emissions at the national level, using widely available sustainability and economical/industrial parameters as inputs. The inputs for the model were selected and optimized using a genetic algorithm and the ANN was trained using the following variables: gross domestic product, gross inland energy consumption, incineration of wood, motorization rate, production of paper and paperboard, sawn wood production, production of refined copper, production of aluminum, production of pig iron and production of crude steel. The wide availability of the input parameters used in this model can overcome a lack of data and basic environmental indicators in many countries, which can prevent or seriously impede PM emission forecasting. The model was trained and validated with the data for 26 EU countries for the period from 1999 to 2006. PM(10) emission data, collected through the Convention on Long-range Transboundary Air Pollution - CLRTAP and the EMEP Programme or as emission estimations by the Regional Air Pollution Information and Simulation (RAINS) model, were obtained from Eurostat. The ANN model has shown very good performance and demonstrated that the forecast of PM(10) emission up to two years can be made successfully and accurately. The mean absolute error for two-year PM(10) emission prediction was only 10%, which is more than three times better than the predictions obtained from the conventional multi-linear regression and principal component regression models that were trained and tested using the same datasets and input variables.

Distribution of Heavy Metals and Arsenic in Soils of Belgrade (Serbia and Montenegro) 1

Soils of the urban and suburban area of Belgrade have been hardly studied, especially concerning their concentrations of potentially toxic metals. The present paper is aimed at determining the possible pollution in soils. The total acid soluble concentrations of heavy metals and As in the samples were determined. It was found that they were arranged in the order Zn > Ni > Pb > Cr > Cu > As > Hg > Cd in samples collected in the examined area (the order of the elements is based on their arithmetic mean concentrations). In all the samples collected at 0–10 and 40–50 cm depths from 46 selected sites, the contents of Pb and Zn were lower at the depth 40-50 cm. Using target values given by the Dutch Ministry of Housing, Spatial Planning and Environment, it may be concluded that Belgrade soil can, for the most part, be regarded as unpolluted. Traffic seems to be one of the main sources of these metals, but the influence of other factors cannot be excluded. 1Serbia and Montenegro was the name of the country at the time this paper was accepted; the name was changed to Serbia as of May 21, 2006.

The forecasting of municipal waste generation using artificial neural networks and sustainability indicators

The feasibility of modeling municipal waste generation (MWG) for countries at different levels of development using artificial neural networks (ANN) and selected generic indicators of sustainability was investigated. The main goals of this research were to develop ANN-based models for predicting MWG, to overcome the problem of incomplete MWG data, which is notable in developing countries, and to provide a new method for the planning of municipal solid waste management systems as well as for the simulation of various other scenarios. Data from 26 European countries was used in this study as training, test and validation datasets for the developing of ANN models. Since this kind of modeling is particularly important for developing countries where MWG data is missing or incomplete, emphasis was placed on modeling of MWG for Bulgaria and Serbia. Based on a comparison of actual MWG data with predictions given by the model, we show that ANNs can be applied successfully to modeling and forecasting MWG on a national scale. Moreover, the scope for possible application of the model is broad, since it uses generic indicators of sustainability such as gross domestic product, domestic material consumption and resource productivity, and performs well for countries with highly diversified levels of economic development, industrial structure, productivity and output.

Sorption of zinc by novel pH-sensitive hydrogels based on chitosan, itaconic acid and methacrylic acid

Novel pH-sensitive hydrogels based on chitosan, itaconic acid and methacrylic acid were applied as adsorbents for the removal of Zn(2+) ions from aqueous solution. In batch tests, the influence of solution pH, contact time, initial metal ion concentration and temperature was examined. The sorption was found pH dependent, pH 5.5 being the optimum value. The adsorption process was well described by the pseudo-second order kinetic. The hydrogels were characterized by spectral (Fourier transform infrared-FTIR) and structural (SEM/EDX and atomic force microscopy-AFM) analyses. The surface topography changes were observed by atomic force microscopy, while the changes in surface composition were detected using phase imaging AFM. The negative values of free energy and enthalpy indicated that the adsorption is spontaneous and exothermic one. The best fitting isotherms were Langmuir and Redlich-Peterson and it was found that both linear and nonlinear methods were appropriate for obtaining the isotherm parameters. However, the increase of temperature leads to higher adsorption capacity, since swelling degree increased with temperature.

Danube and Sava river sediment monitoring in Belgrade and its surroundings

Belgrade is the largest city in Serbia located at the confluence of river Sava to the Danube river. The quality of water and sediments of rivers which run through Belgrade is of a significant importance, since water from these rivers is a source of Belgrade drinking water supply system and probable anthropogenic contamination is related to industrialization and inputs of sewage water. In order to follow the sediment quality of river Sava (km 62-1) and river Danube (km 1193–1124) in Belgrade and its surroundings, the content of As, Cd, Cr, Cu, Zn, Ni, Pb and Hg were measured in the period 2001–2005. The content of 16 polycyclic aromatic hydrocarbons (PAHs) was measured in 2005. The results have shown that, due to the metal content, examined Danube sediment quality varies from class 1 to class 3, predominantly nickel being the class determining parameter. Elevated copper, zinc and mercury concentrations were measured at some profiles, as well. Typically due to the nickel content, Sava sediment quality belongs to class 3 in the period 2001–2004. Elevated concentrations of cadmium, zinc and mercury were observed in 2001, as well. Moreover, in 2005, sediments from three profiles were extremely polluted with nickel, leading the Sava sediment to class 4, when highest urgency measures are needed. Total PAH concentration in the sediments from Danube (213.1–575.4 μg kg− 1) was lower than total PAH concentration from Sava sediments (416.2–595.3 μg kg− 1). Nevertheless, according to the Dutch regulatory system, it has been concluded that river sediments in Belgrade and its surroundings were not polluted with PAHs in 2005.

Bioreactor validation and biocompatibility of Ag/poly(N-vinyl-2-pyrrolidone) hydrogel nanocomposites

Silver/poly(N-vinyl-2-pyrrolidone) (Ag/PVP) nanocomposites containing Ag nanoparticles at different concentrations were synthesized using γ-irradiation. Cytotoxicity of the obtained nanocomposites was determined by MTT assay in monolayer cultures of normal human immunocompetent peripheral blood mononuclear cells (PBMC) that were either non-stimulated or stimulated to proliferate by mitogen phytohemagglutinin (PHA), as well as in human cervix adenocarcinoma cell (HeLa) cultures. Silver release kinetics and mechanical properties of nanocomposites were investigated under bioreactor conditions in the simulated body fluid (SBF) at 37°C. The release of silver was monitored under static conditions, and in two types of bioreactors: perfusion bioreactors and a bioreactor with dynamic compression coupled with SBF perfusion simulating in vivo conditions in articular cartilage. Ag/PVP nanocomposites exhibited slight cytotoxic effects against PBMC at the estimated concentration of 0.4 μmol dm(-3), with negligible variations observed amongst different cell cultures investigated. Studies of the silver release kinetics indicated internal diffusion as the rate limiting step, determined by statistically comparable results obtained at all investigated conditions. However, silver release rate was slightly higher in the bioreactor with dynamic compression coupled with SBF perfusion as compared to the other two systems indicating the influence of dynamic compression. Modelling of silver release kinetics revealed potentials for optimization of Ag/PVP nanocomposites for particular applications as wound dressings or soft tissue implants.

Forecasting human exposure to PM10 at the national level using an artificial neural network approach

A neural network model for predicting country‐level concentrations of the fraction of particulates in the air with sizes less than 10 µm (PM10) has been developed using widely available sustainability and economical/industrial parameters as inputs. The model was trained and validated with the data for 23 European Union (EU) countries plus the EU27 as a group for the period from 2000 to 2008. The inputs for the model were selected using correlation analyses. Country‐level PM10 concentration data that were used as a model output were obtained from the World Bank. The artificial neural network (ANN) model, created with inputs chosen by correlation analyses, has shown very good performance in the forecast of country‐level PM10 concentrations. The mean absolute error for the ANN model prediction, in the case of most of the EU countries, was less than 13%, indicating stable and accurate predictions. The predictions obtained from the principal component regression model, which was trained and tested using the same datasets and input variables, had mean absolute errors from 20% to 150% for most of the countries. The wide availability of input parameters used in this model can overcome the problem of lack and scarcity of data in many countries, which can in turn prevent the determination of human exposure to PM10 at the national level. Copyright

Characterization of PCBs from computers and mobile phones, and the proposal of newly developed materials for substitution of gold, lead and arsenic.

In this paper, we have analyzed parts of printed circuit board (PCB) and liquid crystal display (LCD) screens of mobile phones and computers, quantitative and qualitative chemical compositions of individual components, and complete PCBs were determined. Differential thermal analysis (DTA) and differential scanning calorimetry (DSC) methods were used to determine the temperatures of phase transformations, whereas qualitative and quantitative compositions of the samples were determined by X-ray fluorescence spectrometry (XRF), inductively coupled plasma optical emission spectrometry (ICP-OES), and scanning electron microscopy (SEM)-energy dispersive X-ray spectrometry (EDS) analyses. The microstructure of samples was studied by optical microscopy. Based on results of the analysis, a procedure for recycling PCBs is proposed. The emphasis was on the effects that can be achieved in the recycling process by extraction of some parts before the melting process. In addition, newly developed materials can be an adequate substitute for some of the dangerous and harmful materials, such as lead and arsenic are proposed, which is in accordance with the European Union (EU) Restriction of the use of certain hazardous substances (RoHS) directive as well as some alternative materials for use in the electronics industry instead of gold and gold alloys.

Prediction of nitrogen oxides emissions at the national level based on optimized artificial neural network model

Nitrogen oxides (NOx) emissions into the atmosphere have multiple negative effects on the environment and effects directly and indirectly on human health. This paper describes the development of a model for NOx emission prediction at the national level based on artificial neural networks (ANNs) and on widely available sustainability, industrial, and economical parameters as input variables. In this study, 11 sustainability, industrial, and economical parameters were chosen as potential input variables. The ANN models were trained, validated, and tested with available data for 17 European countries, USA, China, Japan, Russia, and India for the years 2001 to 2008. The ANN modeling was performed using general regression neural network (GRNN), and correlation and variance inflation factor (VIF) analysis were applied to reduce the number of input variables. The best results were obtained using the selection of inputs based on the correlation between input variables, which provided a more accurate prediction than the GRNN model created with all initial selected input variables. Sensitivity analysis showed that the input variables with the largest influences on the GRNN model results were (in descending order) electricity production from oil sources, agricultural land, fossil fuel energy consumption, number of vehicles, gross domestic product, energy use, and electricity production from coal sources.

Chemometrics in biomonitoring: Distribution and correlation of trace elements in tree leaves

The concentrations of 15 elements were measured in the leaf samples of Aesculus hippocastanum, Tilia spp., Betula pendula and Acer platanoides collected in May and September of 2014 from four different locations in Belgrade, Serbia. The objective was to assess the chemical characterization of leaf surface and in-wax fractions, as well as the leaf tissue element content, by analyzing untreated, washed with water and washed with chloroform leaf samples, respectively. The combined approach of self-organizing networks (SON) and Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE) aided by Geometrical Analysis for Interactive Aid (GAIA) was used in the interpretation of multiple element loads on/in the tree leaves. The morphological characteristics of the leaf surfaces and the elemental composition of particulate matter (PM) deposited on tree leaves were studied by using scanning electron microscopy (SEM) with energy dispersive spectroscopy (EDS) detector. The results showed that the amounts of retained and accumulated element concentrations depend on several parameters, such as chemical properties of the element and morphological properties of the leaves. Among the studied species, Tilia spp. was found to be the most effective in the accumulation of elements in leaf tissue (70% of the total element concentration), while A. hippocastanum had the lowest accumulation (54%). After water and chloroform washing, the highest percentages of removal were observed for Al, V, Cr, Cu, Zn, As, Cd and Sb (>40%). The PROMETHEE/SON ranking/classifying results were in accordance with the results obtained from the GAIA clustering techniques. The combination of the techniques enabled extraction of additional information from datasets. Therefore, the use of both the ranking and clustering methods could be a useful tool to be applied in biomonitoring studies of trace elements.