Chemical characterization of PM₁₀ and PM₂.₅ mass concentrations emitted by heterogeneous traffic

In this paper, the chemical characterization of PM₁₀ and PM_2.5 mass concentrations emitted by heterogeneous traffic in Chennai city during monsoon, winter and summer seasons were analysed. The 24-h averages of PM₁₀ and PM_2.5 mass concentrations, showed higher concentrations during the winter season (PM₁₀ = 98 μg/m³; PM_2.5 = 74 μg/m³) followed by the monsoon (PM₁₀ = 87 μg/m³; PM_2.5 = 56 μg/m³) and summer (PM₁₀ = 77 μg/m³; PM_2.5 = 67 μg/m³) seasons. The assessment of 24-h average PM₁₀ and PM_2.5 concentrations was indicated as violation of the world health organization (WHO standard for PM₁₀ = 50 μg/m³ and PM_2.5 = 25 μg/m³) and Indian national ambient air quality standards (NAAQS for PM₁₀ = 100 μg/m³ and PM_2.5 = 60 μg/m³).The chemicals characterization of PM₁₀ and PM_2.5 samples (22 samples) for each season were made for water soluble ions using Ion Chromatography (IC) and trace metals by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) instrument. Results showed the dominance of crustal elements (Ca, Mg, Al, Fe and K), followed by marine aerosols (Na and K) and trace elements (Zn, Ba, Be, Ca, Cd, Co, Cr, Cu, Mn, Ni, Pb, Se, Sr and Te) emitted from road traffic in both PM₁₀ and PM_2.5 mass. The ionic species concentration in PM₁₀ and PM_2.5 mass consists of 47-65% of anions and 35-53% of cations with dominance of SO₄²⁻ ions. Comparison of the metallic and ionic species in PM₁₀ and PM_2.5 mass indicated the contributions from sea and crustal soil emissions to the coarse particles and traffic emissions to fine particles.     

Prediction of engine performance and exhaust emissions with different proportions of ethanol–gasoline blends using artificial neural networks

The main purpose of this study is to experimentally investigate the use of ANNs (artificial neural networks) modelling to predict engine power, torque and exhaust emissions of a spark ignition engine which operates with gasoline and methanol blends. For the ANN modelling, the standard back-propagation algorithm was found to be the optimal choice for training the model. Afterwards, the performance of the ANN predictions was evaluated with the experimental results by comparing the predictions. Fuel type and engine speed have been used as the input layer, while engine torque, power, exhaust emissions, Tex and BSFC have also been used separately as the output layer. It was found that the ANN model is able to predict the engine performance, exhaust emissions, Tex and BSFC with a correlation coefficient of 0.9991887425, 0.9990868573, 0.9986749623, 0.9988624137, 0.9976761492, 0.9992943894 and 0.9978899033 for the Power, Torque, CO, CO₂, HC, Tex and BSFC for testing data, respectively.     

Hourly predictive artificial neural network and multivariate regression trees models of Ganoderma spore concentrations in Rzeszów and Szczecin (Poland)

Ganoderma spores are one of the most airspora abundant taxa in many regions of the world, and are considered to be important allergens. The aerobiology of Ganoderma basidiospores in two cities in Poland was examined using the volumetric method, (Burkard and Lanzonii Spore Traps), from selected days in 2004, 2005 and 2006. Spores of Ganoderma were present in the atmosphere from June to November, with peak concentrations generally occurring from late July to mid-October. ANN (artificial neural network) and MRT (multivariate regression trees), models indicated that atmospheric phenomenon, hour and relative humidity were the most important variables influencing spore content. The remaining variables (air temperature, dew point, air pressure, wind speed and wind direction), also contributed to the high network performance, (ratio above 1), but their impact was less distinct. Those results are consistent with the Spearman's rank correlation analysis.