Air Pollution Modelling by Machine Learning Methods

Precise environmental modelling of pollutants distributions represents a key factor for addresing the issue of urban air pollution. Nowadays, urban air pollution monitoring is primarily carried out by employing sparse networks of spatially distributed fixed stations. The work in this paper aims at improving the situation by utilizing machine learning models to process the outputs of multi-sensor devices that are small, cheap, albeit less reliable, thus a massive urban deployment of those devices is possible. The main contribution of the paper is the design of a mathematical model providing sensor fusion to extract the information and transform it into the desired pollutant concentrations. Multi-sensor outputs are used as input information for a particular machine learning model trained to produce the CO, NO2, and NOx concentration estimates. Several state-of-the-art machine learning methods, including original algorithms proposed by the authors, are utilized in this study: kernel methods, regularization networks, regularization networks with composite kernels, and deep neural networks. All methods are augmented with a proper hyper-parameter search to achieve the optimal performance for each model. All the methods considered achieved vital results, deep neural networks exhibited the best generalization ability, and regularization networks with product kernels achieved the best fitting of the training set.

Reliable prediction of n-heptane isomerization over Pt/(CrOx/ZrO2)-HMS via comparative assessment of regularization networks and surface response methodologies

Novel Pt–Cr/Zr(x)-HMS catalysts with different molar ratios of Cr/Zr were synthesized.

Correction: Reliable prediction of n-heptane isomerization over Pt/(CrOx/ZrO2)-HMS via comparative assessment of regularization networks and surface response methodologies

Correction for ‘Reliable prediction of n-heptane isomerization over Pt/(CrOx/ZrO2)-HMS via comparative assessment of regularization networks and surface response methodologies’ by Nastaran Parsafard et al., RSC Adv., 2020, 10, 26034–26051, DOI: 10.1039/D0RA04313C.

Summability Techniques and Their Applications in Soft Computing

Summability methods are a useful tool in dealing with the problems in the soft computing like in filtering of the signals and for stabilizing the systems. Signals can be in the form of various types of series (Infinite Series, Fourier series, etc.) and hence, summability theory is applicable in finding the error of approximation and degree of approximation of such signals. In this chapter, the authors gave an introductory discussion on summability theory and approximation of the signals. Further, they explained about the stability of the frequency response of the system. Also, they used the Fourier approximation in the soft computing models (multilayer perceptrons; radial basis function (RBF) or regularization networks, and fuzzy logic models) and found the output data of requirement.