Asha B. Chelani

Prediction of sulphur dioxide concentration using artificial neural networks

Abstract A three-layer neural network model with a hidden recurrent layer is used to predict sulphur dioxide concentration and the predicted values are compared with the measured concentrations at three sites in Delhi. The Levenberg–Marquardt algorithm is used to train the network. The neural network is used to simulate the behaviour of the system. A multivariate regression model is also used for comparison with the results obtained by using the neural network model. The study results indicate that the neural network is able to give better predictions with less residual mean square error than those given by multivariate regression models.

Prediction of ambient PM10 and toxic metals using artificial neural networks.

Abstract In this study, an artificial neural network is employed to predict the concentration of ambient respirable particu-late matter (PM10) and toxic metals observed in the city of Jaipur, India. A feed-forward network with a back-propagation learning algorithm is used to train the neural network the behavior of the data patterns. The meteorological variables of wind speed, wind direction, relative humidity, temperature, and time are taken as input to the network. The results indicate that the network is able to predict concentrations of PM10 and toxic metals quite accurately.

Optimal selection of full scale tannery effluent treatment alternative using integrated AHP and GRA approach

The selection of optimal wastewater treatment alternative involves multiple objectives and/or criteria and hierarchy process. This study integrates analytical hierarchy process (AHP) and grey relation analysis (GRA) for optimal selection of full scale tannery effluent treatment plants. For this purpose, seven tanneries and their effluent treatment facilities are studied in detail in Southern India. The objective hierarchy criterion is considered based on three factors; economic, technical, and administrative, each ofwhich againinvolves hierarchy of indices. A realistic treatment alternative selection is obtained since all the data used is on actual basis.The biggest advantage of this approach is that it provides the information regarding the scope for further improvement in existing treatment options. The study indicates that the AHP and grey relation analysisare powerful tools that can be used for implementation of appropriate wastewater treatment technology.

Particle Size Distribution in Ambient Air of Delhi and Its Statistical Analysis

Particle size distribution in ambient air has been studied in an urban city, Delhi. Different activity sites namely; kerbside, industrial and residential were selected for the study. The statistical analysis was carried out to study the frequency distribution and sources of different particle size fractions. The dominance of coarse particles attributed to local activities was observed at all the sites. It was observed that at kerbside sites, up to 52% of the particles were lower respiratory tract and up to 47% of the particles were upper respiratory tract particles. At residential and industrial sites, up to 40% and 31% were lower and upper respiratory tract particles, respectively. Factor analysis results indicated auto-exhaust as the dominant source of particulate matter at two of the kerbside sites. Resuspended dust was dominant at remaining two kerbside and residential sites. It was inferred using geometric standard deviation of particle size fractions that these were from different sources at residential and industrial site and from similar sources at three of the kerbside sites.

Prediction of daily maximum ground ozone concentration using support vector machine.

The accurate predictions of ground ozone concentrations are required for proper management, control, and making public warning strategies. Due to the difficulties in handling phenomenological models that are based on complex chemical reactions of ozone production, neural network models gained popularity in the last decade. These models also have some limitations due to problems of overfitting, local minima, and tuning of network parameters. In this study, the predictions of daily maximum ozone concentrations are attempted using support vector machines (SVMs). The comparison between the accuracy of SVM and neural network predictions is performed to evaluate their performance. For this, the daily maximum ozone concentration data observed during 2002–2004 at a site in Delhi is utilized. The models are developed using the available meteorological parameters. The results indicated the promising performance of SVM over neural networks in predicting daily maximum ozone concentrations.

Statistical Modeling of Ambient Air Pollutants in Delhi

State space model coupled with Kalman filter is used toforecast metal concentrations observed at Delhi. Three metalsviz. Lead, Iron, Zinc along with Respirable SuspendedParticulate Matter (RSPM) Concentration observed during 1993to 1995 are selected for the study. The data for year 1996is selected to test the forecasting ability of the model.Wind speed is used as an external input. Autoregressive modelwith external input is also used to compare the results. Itis found that the State space model is giving betterpredictions for lead, iron and RSPM concentration than theARX model while for zinc concentration, both the models aregiving good performance results.

Irregularity analysis of CO, NO2 and O3 concentrations at traffic, commercial and low activity sites in Delhi

The irregularity analysis of exceedance time series of gaseous pollutants CO, NO2 and O3 is carried out using Shannon entropy and Fisher information measure. The data observed during 2007–2010 at three sites with different land-use activities in Delhi are analyzed. CO and NO2 showed irregular behavior at both, low anthropogenic activity and commercial activity sites, whereas at traffic site both the pollutant concentrations showed regular behavior. The irregularity is attributed to the multiplicity in emission sources at low activity and commercial site and regular behavior is observed due to the uniformity and well defined source characteristics at the traffic site. O3 at three sites showed irregular behavior owing to its secondary nature. Fisher–Shannon information plane showed the grouping of three pollutants except CO and NO2 at traffic and O3 at low activity site suggesting the similar temporal characteristics of the pollutants even at the sites with different land-use activities.

Nonlinear Analysis and Prediction of Coarse Particulate Matter Concentration in Ambient Air

Abstract This study attempts to characterize and predict coarse particulate matter (PM10) concentration in ambient air using the concepts of nonlinear dynamical theory. PM10 data observed daily from 1999 to 2002 at a site in Mumbai, India, was used to study the applicability of the chaos theory. First, the autocorrelation function and Fourier power spectrum were used to analyze the behavior of the time–series. The dynamics of the time–series was additionally studied through correlation integral analysis and phase space reconstruction. The nonlinear predictions were then obtained using local polynomial approximation based on the reconstructed phase space. The results were then compared with the autoregressive model. The results of nonlinear analysis indicated the presence of chaotic character in the PM10 time–series. It was also observed that the nonlinear local approximation outperforms the autoregressive model, because the observed relative error of prediction for the autoregressive model was greater than the local approximation model. The invariant measures of nonlinear dynamics computed for the predicted time–series using the two models also supported the same findings.