PARTICLE FILTER OPTIMIZATION TOOLS FOR DYNAMICAL SYSTEMS PROBABILISTIC MODELS

В работе предложены математические инструменты на основе достаточных статистик и декомпозиции выборок в сочетании с алгоритмами распределенных вычислений, позволяющие существенно повысить эффективность процедуры фильтрации. Filtering algorithms are used to assess the state of dynamic systems when solving various practical problems, such as voice synthesis and determining the geo-position and monitoring the movement of objects. In the case of complex hierarchical dynamic systems with a large number of time slices, the process of calculating probabilistic characteristics becomes very time-consuming due to the need to generate a large number of samples. The essence of optimization is to reduce the number of samples generated by the filter, increase their consistency and speed up computational operations. The paper offers mathematical tools based on sufficient statistics and sample decomposition in combination with distributed computing algorithms that can significantly improve the efficiency of the filtering procedure.

Development and Validation of an ICP-AES Method for the Determination of Toxic and Nutrient Metals in Candies: Application for the Analysis of Different Samples from the Greek Market

In this study, we present the development and validation of an inductively coupled plasma-atomic emission spectrometric (ICP-AES) method for the determination of Ag, Ba, Bi, Ca, Cd, Co, Cr, Cu, Fe, Mg, Mn, Ni, Pb and Zn in different candies. Various wet digestion protocols were examined in order to ensure minimum consumption of chemicals and sample preparation time. Under optimized conditions, less than 10 min were required for complete sample decomposition. The ICP-AES method was validated in terms of linearity, accuracy, precision, limits of detection (LODs) and limits of quantification (LOQs). The relative recoveries for the proposed method ranged between 80.0% and 119.0%, while the relative standard deviation values were lower than 9.0%, indicating good method accuracy and precision, respectively. The LODs for the examined analytes were 0.04–2.25 mg kg−1. Finally, the proposed method was successfully employed for the analysis of hard candies, jellies and lollipops that are sold in the Greek market, which are highly likely to be consumed by children.

Analysis of influencing factors of data error in chemical analysis of iron and steel materials

This paper analyzes the influencing factors of data error in chemical analysis of iron and steel materials, including sample preparation factor, sample decomposition factor, analytical instrument factor, reagent factor, analysis method factor. The purpose is to reduce the error of data measurement results and improve the accuracy of data analysis results by studying the measures of eliminating instrument application error, doing a good job in reagent selection, appropriately increasing the number of experiments, strictly following the operation specifications and reasonably using the allowable deviation table.

Development of Methods for Mg, Sr and Pb Isotopic Analysis of Crude Oil by MC-ICP-MS: Addressing the Challenges of Sample Decomposition

New approaches in isotope geochemistry require the development of novel methods for the isotopic analysis of crude oil, a typically complex and very hard to digest organic geological matrix. In...

Thermal analysis: basic concept of differential scanning calorimetry and thermogravimetry for beginners

We present an overview for the basic fundamental of thermal analysis, which is applicable for educational purposes, especially for lecturers at the universities, who may refer to the articles as the references to “teach” or to “lecture” to final year project students or young researchers who are working on their postgraduate projects. Description of basic instrumentation [i.e. differential scanning calorimetry (DSC) and thermogravimetry (TGA)] covers from what we should know about the instrument, calibration, baseline and samples’ signal. We also provide the step-by-step guides for the estimation of the glass transition temperature after DSC as well as examples and exercises are included, which are applicable for teaching activities. Glass transition temperature is an important property for commercial application of a polymeric material, e.g. packaging, automotive, etc. TGA is also highlighted where the analysis gives important thermal degradation information of a material to avoid sample decomposition during the DSC measurement. The step-by-step guides of the estimation of the activation energy after TGA based on Hoffman’s Arrhenius-like relationship are also provided.

The use of differential calculation methods for the destruction of copolymers of polyethylene glycol fumarate with the acrylic acid

In this work, the thermal decomposition of copolymers based on polyethylene glycol fumarate with the acrylic acid using various ratios of initial monomers has been studied for the first time. The samples were studied in air and nitrogen. According to the thermograms analysis, it was found that the copolymer sample decomposition begins at higher temperatures for a copolymer with high content of polyester resin. The copolymer is vigorously oxidized by the oxygen when heated in air, and one can observe almost complete sample decomposition, whereas it decomposes with a residue of ~ 15% in an inert medium. The activation energies for copolymers with different compositions were estimated using the differential methods of Freeman-Carroll, Achar and Sharpe-Wentworth. The activation energy values found by the three methods demonstrated a good convergence. It was shown that, the activation energy values are higher (~ 200 kJ/mol in the inert medium, and ~ 95 kJ/mol in the oxygen atmosphere) for a copolymer with a lower composition of polyester resin, and the activation energy is ~180 and ~85 kJ/mol for a copolymer with a greater composition of p-EGF-AA. The copolymer is more thermostable in the nitrogen atmosphere according to the kinetic parameters. Additionally, there were determined the thermodynamic characteristics, such as the Gibbs energy (∆G) and the entropy (∆S). They also confirm the destruction process dependence on the components ratio in the synthesized copolymer.

Separation and quantification of 90Sr from ion-exchange resin radioactive waste: methods and techniques of analysis

Four methods for 90Sr separation from spent ion-exchange resin samples were carried out, offering a useful methodology to achieve interferences free 90Sr fractions. The four methods consist in resin sample decomposition, pre-treatment and selective separation of 90Sr by using: (a) a single chromatographic extraction process, (b) double chromatographic extraction, (c) a single chromatographic extraction process followed in sequence by two precipitations, and (d) ion-exchange chromatography, followed by extraction chromatography and precipitation. Mineralization by microwave acid digestion and the four 90Sr separation methods thoroughly presented are available. Data processing methods (adjustable modified efficiency tracing – a new improved approach for the efficiency tracing LSC technique, non-linear regression and α-β discrimination) to obtain the activities values of α, β-γ, pure β emitters and the evaluation of chemical recovery yield of strontium were presented. A discussion about activity assessment in 90Sr purified fractions, providing a convincing argument to support the accuracy of the 90Sr separation methods, is also offered.

4S-DT: Self Supervised Super Sample Decomposition for Transfer learning with application to COVID-19 detection

ABSTRACTDue to the high availability of large-scale annotated image datasets, knowledge transfer from pre-trained models showed outstanding performance in medical image classification. However, building a robust image classification model for datasets with data irregularity or imbalanced classes can be a very challenging task, especially in the medical imaging domain. In this paper, we propose a novel deep convolutional neural network, we called Self Supervised Super Sample Decomposition for Transfer learning (4S-DT) model. 4S-DT encourages a coarse-to-fine transfer learning from large-scale image recognition tasks to a specific chest X-ray image classification task using a generic self-supervised sample decomposition approach. Our main contribution is a novel self-supervised learning mechanism guided by a super sample decomposition of unlabelled chest X-ray images. 4S-DT helps in improving the robustness of knowledge transformation via a downstream learning strategy with a class-decomposition layer to simplify the local structure of the data. 4S-DT can deal with any irregularities in the image dataset by investigating its class boundaries using a downstream class-decomposition mechanism. We used 50,000 unlabelled chest X-ray images to achieve our coarse-to-fine transfer learning with an application to COVID-19 detection, as an exemplar. 4S-DT has achieved an accuracy of 97.54% in the detection of COVID-19 cases on an extended test set enriched by augmented images, out of which all real COVID-19 cases were detected, which was the highest accuracy obtained when compared to other methods.