Modern studies in the field of environment science and engineering show that deterministic models struggle to capture the relationship between the concentration of atmospheric pollutants and their emission sources. The recent advances in statistical modeling based on machine learning approaches have emerged as solution to tackle these issues. It is a fact that, input variable type largely affect the performance of an algorithm, however, it is yet to be known why an algorithm is preferred over the other for a certain task. The work aims at highlighting the underlying principles of machine learning techniques and about their role in enhancing the prediction performance. The study adopts, 38 most relevant studies in the field of environmental science and engineering which have applied machine learning techniques during last 6 years. The review conducted explores several aspects of the studies such as: 1) the role of input predictors to improve the prediction accuracy; 2) geographically where these studies were conducted; 3) the major techniques applied for pollutant concentration estimation or forecasting; and 4) whether these techniques were based on Linear Regression, Neural Network, Support Vector Machine or Ensemble learning algorithms. The results obtained suggest that, machine learning techniques are mainly conducted in continent Europe and America. Furthermore a factorial analysis named multi-component analysis performed show that pollution estimation is generally performed by using ensemble learning and linear regression based approaches, whereas, forecasting tasks tend to implement neural networks and support vector machines based algorithms.

Worldwide, solar power has become the third biggest source of energy after hydro and wind, thus, catching a global attention. This study draws a comparison between the performance of Dual Axis Solar Tracker (DAST) and Static Solar System (SSS) or a solar system with fixed axis to assess the solar power potential of Ural region located in Russian Federation. By using various sun angles such as zenith, solar altitude etc., solar irradiance was calculated during a day either cloudy or sunny, whereas the cumulative energy generated during the whole month was also measured. Moreover, for a clearer understanding, efficiency differences during a day as well as during the whole month were considered too. The results obtained in the form of energy gain and efficiency of proposed tracking unit presented in the form of graphs and tables suggest that the performance of DAST is significantly better than SSS.

Due to serious health concerns, recently air quality monitoring and advance prediction model development have become a key focus for researchers. In view of prediction models, Ensemble data mining tool Random Forest was employed to predict the concentration of Nitrogen Dioxide (NO2) in air. For this paper, the monitoring dataset of emission inventory and meteorological parameters was analyzed as input predictors. While, the prediction performance of the model was calculated by using evaluation measures namely Correlation coefficient ( R ² ), Root Mean Squared Error (RMSE) and Relative Absolute Error (RAE). The model performance was validated by comparing the results with two well-known classification algorithms known as M5P and Support Vector Machines (SVM) for regression. The models were further assessed under different filtering techniques such as Principle Component Analysis and Relief Attribute Evaluator. The results obtained suggest that the performance of Random Forest to predict the atmospheric concentration of NO2 is better than other two data mining approaches.

One of the analytical methods of presenting solutions of nonlinear partial differential equations is the method of special series in powers of specially selected functions called basis functions. The coefficients of such series are found successively as solutions of linear differential equations. To find recurrence, the coefficient is achieved by the choice of basis functions, which may also contain arbitrary functions. By using such functional arbitrariness, it allows in some cases to prove the global convergence of the corresponding constructed series, as well as the solvability of the boundary value problem.