Forecasting nonstationary time series based on Hilbert-Huang transform and machine learning

We propose a modification of the adaptive approach to time series forecasting. On the first stage, the original signal is decomposed with respect to a special empirical adaptive orthogonal basis, and the Hilbert’s integral transform is applied. On the second stage, the resulting orthogonal functions and their instantaneous amplitudes are used as input variables for the machine learning unit that employs a hybrid genetic algorithm to train an artificial neural network and a regressive model based on support vector machines. The efficiency of the proposed approach is demonstrated on real data coming from Nord Pool Spot and Australian National Energy Market.     

The Concept of Constructing an Artificial Dispatcher Intelligent System Based on Deep Reinforcement Learning for the Automatic Control System of Electric Networks

Journal of Computer and Systems Sciences International - The concept of the Artificial Dispatcher autonomous intelligent control system for integrating into modern automatic control systems of...     

Production and characterization of porous magnetic biochar: before and after phosphate adsorption insights

Journal of Porous Materials - The modification process of biochars enables different advantages including enhanced adsorption properties for different pollutants. Herein, porous magnetic biochars...     

A Suite of Intelligent Tools for Early Detection and Prevention of Blackouts in Power Interconnections

We propose a suite of intelligent tools based on the integration of methods of agent modeling and machine learning for the improvement of protection systems and emergency automatics. We propose an online approach to the assessment and management of dynamic security of electric power systems (EPS) with the use of a streaming modification of the random forest algorithm. The suite allows to recognize dangerous modes of complex closed-loop EPS, preventing the risk of emergencies on early stages. We show results of experimental tests on IEEE test systems.     

Domestic Hydrotreating Units. Status and Modernization Methods

Data on use of domestic hydrotreating units are generalized. Ways of enhancing operation of the units to improve their technical and economic indexes are indicated. The results of operation of the L-24-6 unit at Angarsk Petrochemical Co. on AGKD-205A catalyst for manufacture of diesel fuel with a sulfur content of less than 0.05% are reported.__________Translated from Khimiya i Tekhnologiya Topliv i Masel, No. 3, pp. 15 – 17, May – June, 2005.     

Electrodeposition of nanocrystalline copper thin films from 1-ethyl-3-methylimidazolium ethylsulphate ionic liquid

Copper thin films are increasingly important as interconnectors for the creation of smaller and better performing integrated circuits and electrodeposition from ionic liquid-based electrolytes could provide a greener fabrication method for these films. The electrodeposition of copper from copper(I) and copper(II) salt solutions in a low cost, widely available ionic liquid, 1-ethyl-3-methylimidazolium ethylsulphate, was studied using a range of different deposition potentials and temperatures. Three different electrolytes containing ~0.1 M of copper(I) chloride(CuCl), copper(II) chloride (CuCl₂) and copper(II) sulphate (CuSO₄) were used. Under similar deposition conditions, the films obtained from CuCl and CuSO₄-based electrolytes presented better continuity than films obtained from CuCl₂-based electrolyte. Continuous films with a homogeneous structure were obtained by electrodeposition from CuCl and CuSO₄-based solutions at a constant potential of ?1.8 V and a temperature of 35 °C. Under similar deposition parameters, the films deposited from CuCl₂-based electrolyte presented the largest particle size, while those deposited from copper(I) chloride and CuSO₄-based solutions presented finer microstructures. X-ray diffraction analysis and energy dispersive X-ray spectroscopy showed that the deposits were crystalline and consisted mainly of copper, with traces of oxygen and sulphur resulting from residues of the ionic liquid. The films presented a nanocrystalline microstructure consisting of particles about 25 nm, aggregated in clusters.     

The study of relative phase-permeability functions for anisotropic media

Properties of relative phase-permeability functions for different characteristic configurations of a medium with anisotropic filtration properties are studied using the numerical-analytical method. The anisotropy may be related to fracturing created, e.g., as a result of a brittle rupture. Correlations between tensors of the phase and absolute permeabilities are analyzed on the assumption of capillary equilibrium. Examples of relative phase-permeability functions are obtained for media with orthotropic and monoclinic symmetry of filtration properties. The influence of the property of connectivity on the found functions is shown, and the dependence of the position of the principal axes of the phase-permeability tensor on saturation is also investigated. Thereby, the misalignment of the tensors of the phase and absolute permeabilities is shown. The results qualitatively agree with the experimental data.