Prediction of ambient PM10 and toxic metals using artificial neural networks

In this study, an artificial neural network is employed to predict the concentration of ambient respirable particulate 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.     

人工神经网络和专家系统在污水生物处理系统中的应用

对近年来国内外污水生物处理系统中人工神经网络和专家系统的应用进行了简要的回顾。分析了废水生物处理工艺难于控制的原因及人工神经网络和专家系统的结构和特点。结果表明．国外智能控制发展迅速,并且应用领域遍及污水生物处理的各个方面,国内尚处于起步阶段。简要探讨了废水生物处理智能控制今后应深入研究的问题及方向。     

QSAR study on the toxicity of substituted benzenes to the algae (Scenedesmus obliquus).

50% effective inhibition concentration 48h-EC50 of 40 substituted benzenes to the algae (Scenedesmus obliquus) was determined. The energy of the lowest unoccupied molecular orbital (E(LUMO)) was calculated by the quantum chemical method MOPAC6.0-AM1. By using E(LUMO) and the hydrophobicity parameter log K(OW) the quantitative structure-activity relationship model (QSAR) was developed: log1/EC50=0.272 logK(OW) - 0.659E(LUMO) + 2.54, R2 = 0.793, S.E. = 0.316, F = 71.07, n = 40. A series of equations were obtained about the measured EC50 values of different subclasses of compounds. For those compounds containing double -NO2, their toxicity may be related chiefly to the intracellular reduction of -NO2 obtaining electron, while for anilines and phenols, K(OW) contributes most to the QSAR and E(LUMO) very little.     

Modelling of dissolved oxygen content using artificial neural networks: Danube River, North Serbia, case study

The aims of this study are to create an artificial neural network (ANN) model using non-specific water quality parameters and to examine the accuracy of three different ANN architectures: General Regression Neural Network (GRNN), Backpropagation Neural Network (BPNN) and Recurrent Neural Network (RNN), for prediction of dissolved oxygen (DO) concentration in the Danube River. The neural network model has been developed using measured data collected from the Bezdan monitoring station on the Danube River. The input variables used for the ANN model are water flow, temperature, pH and electrical conductivity. The model was trained and validated using available data from 2004 to 2008 and tested using the data from 2009. The order of performance for the created architectures based on their comparison with the test data is RNN > GRNN > BPNN. The ANN results are compared with multiple linear regression (MLR) model using multiple statistical indicators. The comparison of the RNN model with the MLR model indicates that the RNN model performs much better, since all predictions of the RNN model for the test data were within the error of less than ±10 %. In case of the MLR, only 55 % of predictions were within the error of less than ±10 %. The developed RNN model can be used as a tool for the prediction of DO in river waters.     

Application of artificial neural networks to modeling and prediction of ambient ozone concentrations

The deterministic modeling of ambient O3 concentrations is difficult because of the complexity of the atmospheric system in terms of the number of chemical species; the availability of accurate, time-resolved emissions data; and the required rate constants. However, other complex systems have been successfully approximated using artificial neural networks (ANNs). In this paper, ANNs are used to model and predict ambient O3 concentrations based on a limited number of measured hydrocarbon species, NOx compounds, temperature, and radiant energy. In order to examine the utility of these approaches, data from the Coastal Oxidant Assessment for Southeast Texas (COAST) program in Houston, TX, have been used. In this study, 53 hydrocarbon compounds, along with O3, nitrogen oxides, and meteorological data were continuously measured during summer 1993. Steady-state ANN models were developed to examine the ability of these models to predict current O3 concentrations from measured VOC and NOx concentrations. To predict the future concentrations of O3, dynamic models were also explored and were used for extraction of chemical information such as reactivity estimations for the VOC species. The steady-state model produced an approximation of O3 data and demonstrated the functional relationship between O3 and VOC-NOx concentrations. The dynamic models were able to the adequately predict the O3 concentration and behavior of VOC-NOx-O3 system a number of hourly intervals into the future. For 3 hr into the future, O3 concentration could be predicted with a root-mean squared error (RMSE) of 8.21 ppb. Extending the models further in time led to an RMSE of 11.46 ppb for 5-hr-ahead values. This prediction capability could be useful in determining when control actions are needed to maintain measured concentrations within acceptable value ranges.     

Biological treatment of whey by Tetrahymena pyriformis and impact study on laboratory-scale wastewater lagoon process

A procedure based on a biological treatment of whey was tested as part of research on waste treatment at the scale of small cheesemaking units. We studied the potential biodegradation of whey by a protozoan ciliate, Tetrahymena pyriformis, and evaluated the functional, microbiological and physiological disturbances caused by crude whey and the biodegraded whey in laboratory-scale pilots mimicking a natural lagoon treatment. The results show that T. pyriformis can strongly reduce the pollutant load of whey. In the lagoon pilots serving as example of receptor media, crude whey gradually but completely arrested operation, whereas with the biodegraded whey adverse effects were only temporary, and normal operation versus a control was gradually recovered in a few days.     

Quantitative structure-toxicity relationship study of lethal concentration to tadpole (Bufo vulgaris formosus) for organophosphorous pesticides

In the present study more than 1,000 structural parameters of 41 organophosphorus pesticides (OPs) were calculated using the software ChemOffice 8.03 and Dragon 2.1. Then, with multivariate linear regression and best subset regression analyses, different equations were derived to calculate the lethal toxicity, LC(50), for these 41 organophosphorous pesticides found in tadpoles (Bufo vulgaris formosus). An equation was developed for all selected OPs, especially those with relatively low toxicity levels (LC(50)>4.5mM) that accounted for 89.09% of the variability in the toxic effect. The equation indicated that the main contributions to OPs toxicity with tadpoles were the electrostatic contribution qH(+) (maximum net positive H atomic charge), spatial autocorrelation (MATS7 m) and hydrophobicity (lgK(ow)), with the two former being the most important parameters. For OPs with high toxicity, however, different structural parameters were introduced. The following equation was developed with LC(50)<4.5mM. These equations implied that with different levels of toxicity there could have different mechanisms in the tadpole. Furthermore, the results showed that molecular structural parameters had a particular value in modeling chemical reactivity within a homologous series of compounds.     

Estimation of COVID-19 patient numbers using artificial neural networks based on air pollutant concentration levels

Environmental Science and Pollution Research - The dilemma between health concerns and the economy is apparent in the context of strategic decision making during the pandemic. In particular,...     

Application of artificial neural networks to predict the heavy metal contamination in the Bartin River

Environmental Science and Pollution Research - In this study, copper (Cu), iron (Fe), zinc (Zn), manganese (Mn), nickel (Ni), and lead (Pb) analyses were performed, and the results were modelled by...     

Prediction of membrane fouling using artificial neural networks for wastewater treated by membrane bioreactor technologies: bottlenecks and possibilities

Membrane fouling is a major concern for the optimization of membrane bioreactor (MBR) technologies. Numerous studies have been led in the field of membrane fouling control in order to assess with precision the fouling mechanisms which affect membrane resistance to filtration, such as the wastewater characteristics, the mixed liquor constituents, or the operational conditions, for example. Worldwide applications of MBRs in wastewater treatment plants treating all kinds of influents require new methods to predict membrane fouling and thus optimize operating MBRs. That is why new models capable of simulating membrane fouling phenomenon were progressively developed, using mainly a mathematical or numerical approach. Faced with the limits of such models, artificial neural networks (ANNs) were progressively considered to predict membrane fouling in MBRs and showed great potential. This review summarizes fouling control methods used in MBRs and models built in order to predict membrane fouling. A critical study of the application of ANNs in the prediction of membrane fouling in MBRs was carried out with the aim of presenting the bottlenecks associated with this method and the possibilities for further investigation on the subject.     

Prediction of CO maximum ground level concentrations in the Bay of Algeciras, Spain using artificial neural networks

The region of the Bay of Algeciras is a very industrialized area where very few air pollution studies have been carried out. The main objective of this work has been the use of artificial neural networks (ANNs) as a predictive tool of high levels of ambient carbon monoxide (CO). Two approaches have been used: multilayer perceptron models (MLPs) with backpropagation learning rule and k-Nearest Neighbours (k-nn) classifiers, in order to predict future peaks of carbon monoxide. A resampling strategy with twofold cross-validation allowed the statistical comparison of the different topologies and models considered in the study. The procedure of random resampling permits an adequate and robust multiple comparisons of the tested models and allow us to select a group of best models.     

Pattern recognition of the movement tracks of medaka (Oryzias latipes) in response to sub-lethal treatments of an insecticide by using artificial neural networks

Specimens of medaka (Oryzias latipes) were observed continuously through an automatic image recognition system before and after treatments of an anti-cholinesterase insecticide, diazinon (0.1 mg/l), for 4 days in semi-natural conditions (2 days before treatment and 2 days after treatment). The "smooth" pattern was typically shown as a normal movement behavior, while the "shaking" pattern was frequently observed after treatments of diazinon. These smooth and shaking patterns were selected for training with an artificial neural network. Parameters characterizing the movement tracks, such as speed, degree of backward movements, stop duration, turning rate, meander, and maximum distance movements in the y-axis of 1-min duration, were given as input (six nodes) to a multi-layer perceptron with the back propagation algorithm. Binary information for the smooth and shaking patterns was separately given as the matching output (one node), while eight nodes were assigned to a single hidden layer. As new input data were given to the trained network, it was possible to recognize the smooth and shaking patterns of the new input data. Average recognition rates of the smooth pattern decreased significantly while those for the shaking pattern increased to a higher degree after treatments of diazinon. The trained network was able to reveal the difference in the shaking pattern in different light phases before treatments of diazinon. This study demonstrated that artificial neural networks could be useful for detecting the presence of toxic chemicals in the environment by serving as in-situ behavioral monitoring tools.     

Patterns of PCDDs and PCDFs in human milk and food and their characterization by artificial neural networks

Artificial neural network (ANN) has been recently introduced as a tool for data analysis. In this study, Kohonen's self-organizing maps (SOMs), a special type of neural network, were applied to a set of PCDD/PCDF concentrations found in 54 human milk and 83 food samples, which were collected in a number of countries all over the world. Data were obtained from the scientific literature. The purpose of the study was to find a potential relationship between PCDD/PCDF congener profiles in human milk and the dietary habits of the different countries in which samples were collected. The comparison of the SOM component planes for human milk and foodstuffs indicates that those countries with a greater fish consumption show also higher PCDD/PCDF concentrations in human milk. SOMs enable both the visualization of sample units and the visualization of congener distribution.     

Evaluation of artificial neural networks for fine particulate pollution (PM10 and PM2.5) forecasting

Multi-layer perceptron (MLP) artificial neural network (ANN) models are compared with traditional multiple regression (MLR) models for daily maximum and average O3 and particulate matter (PM10 and PM2.5) forecasting. MLP particulate forecasting models show little if any improvement over MLR models and exhibit less skill than do O3 forecasting models. Meteorological variables (precipitation, wind, and temperature), persistence, and co-pollutant data are shown to be useful PM predictors. If MLP approaches are adopted for PM forecasting, training methods that improve extreme value prediction are recommended.     

Acute toxicity of substituted phenols to Rana japonica tadpoles and mechanism-based quantitative structure-activity relationship (QSAR) study.

Acute 12 h and 24 h lethal toxicity (12 h-LC50 and 24 h-LC50) of 31 substituted phenols to Rana japonica tadpoles was determined. Results indicate that toxicity of phenols to tadpoles varied only slightly with length of exposure and the 12-h test could serve as surrogate of the 24-h test. A mechanism-based quantitative structure-activity relationship (QSAR) method was employed and 1-octanol/water partition coefficient (log K(ow))-dependent models were developed to study different modes of toxic action. Most phenols elicited their response via a polar narcotic mechanism and an excellent logK(ow)-dependent model was obtained. Soft electrophilicity and pro-electrophilicity were observed for some phenols and a good log K(ow)-dependent model was also achieved. Additionally, the significant dissociation of carboxyl on benzoic acid derivatives sharply reduced their toxicity. A statistically robust QSAR model was developed for all studied compounds with the combined application of log K(ow), energy of lowest unoccupied orbital (E(lumo)), heat of formation (HOF) and the first-order path molecular connectivity dices (1chi(p)).     

Development of models for predicting toxicity from sediment chemistry by partial least squares-discriminant analysis and counter-propagation artificial neural networks

There is strong interest in developing tools to link chemical concentrations of contaminants to the potential for observing sediment toxicity that can be used in initial screening-level sediment quality assessments. This paper presents new approaches for predicting toxicity in sediments, based on 10-day survival tests with marine amphipods, from sediment chemistry, by means of the application of Partial Least Squares-Discriminant Analysis (PLS-DA) and Counter-propagation Artificial Neural Networks (CP-ANNs) to large historical databases of chemical and toxicity data. The exploration of the internal structure of the developed models revealed inherent limitations of predicting toxicity from common chemical analyses of bulk contaminant concentrations. However, the results obtained in the validation of these models combined relevant values of non-error classification rate, sensitivity and specificity of, respectively, 76, 87 and 73% with PLS-DA and 92, 75 and 97% with CP-ANNs, outperforming the results reported for previous approaches.     

人工神经网络在强化一级处理系统过程模拟与优化中的应用

针对化学强化一级处理系统（CEPT）处理废水时影响因素多，难以进行适当的控制和处理效果的预测等问题，建立起基于BP人工神经网络的CEPT法处理猪场稳定塘废水预测模型，并应用该模型对烧杯试验进行了模拟。结果表明，预测值和实测值吻合较好，模型对COD、总磷、浊度去除率预测的平均相对误差分别为7．5％、4．8％和4．9％。通过对pH值和絮凝剂投药量等可控参数进行优化计算，得到CEPT系统的最佳操作条件和最合理操作条件。该模型的建立为CEPT法处理废水工艺系统实现自动化控制提供了一条简便实用的途径。