The technological inspection of the electrolyte composition in aluminum production is performed using calibration X-ray quantitative phase analysis (QPA). For this purpose, the use of QPA by the Rietveld method, which does not require the creation of multiphase reference samples and is able to take into account the actual structure of the phases in the samples, could be promising. However, its limitations are in its low automation and in the problem of setting the correct initial values of profile and structural parameters. A possible solution to this problem is the application of the genetic algorithm we proposed earlier for finding suitable initial parameter values individually for each sample. However, the genetic algorithm also needs tuning. A self-configuring genetic algorithm that does not require tuning and provides a fully automatic analysis of the electrolyte composition by the Rietveld method was proposed, and successful testing results were presented.

The Tikhonov regularization of the reference intensity ratio (RIR) method integrated within the information retrieval system of X-ray phase analysis together with the method of cluster phase identification is reported. The possibilities of the methods are discussed by the example of the X-ray phase analysis of test mixtures from Round Robin on QPA held by the International Union of Crystallography Commission for Powder Diffraction (CPD).

A method of quantitative X-ray phase analysis (QXPA) based on multiobjective evolutionary optimization for automatization and increase in accuracy of QXPA that is based on full-profile analysis by the

Some possibilities of using convolutional artificial neural networks (ANN) for powder diffraction structural analysis of crystalline substances have been investigated. First, ANNs are used to classify crystalline systems and space groups according to calculated full-profile diffractograms calculated from the crystal structures of the ICSD database (2017 year). The ICSD database contains 192004 structures, of which 80% was used for in-depth network training, and 20% for independent testing of recognition accuracy. The accuracy of classification by a network of crystalline systems was 87.9%, and that of space groups was 77.2%. Secondly, the ANN is used for a similar classification of structural models generated by the stochastic genetic algorithm in the search processes for triclinic crystal structures of test compound K4SnO4 according to their full-profile diffraction patterns. The classification criterion was the entry of one or several atoms into their crystallographic positions in the structure of a substance. Independent deep network training was performed on 120 thousand structural models of the K4PbO4 triclinic structure generated in several runs of the genetic algorithm. The accuracy of the classification of K4SnO4 structural models exceeded 50%. The results show that deeply trained convolutional ANNs can be effective for classifying crystal structures according to the structural characteristics of their powder diffraction patterns

Analytical control of metals and product quality at every stage of the production flow – from raw materials to finished product — is a pre-requisite element in the production of non-ferrous metals. As a rule, the properties of ores and solid-phase middlings defining processing efficiency are, to a greater degree, characteristic of mineral phases and their aggregates rather than of the chemical elements. That’s why, in order to ensure proper process control, in most cases it is necessary to not only monitor the chemical composition but also the mineral phase composition. The most efficient type of control includes comprehensive in-process control, which needs to be highly automated, versatile and, if possible, non-destructive. X-ray spectral analysis for elemental monitoring in combination with X-ray phase analysis for material monitoring would probably be the best option to satisfy the above requirements. The method of X-ray phase analysis would be a limiting factor in implementing such integrated monitoring system, which is due to the fact that it is not easy to develop methods enabling automated in-process analysis of particular ores, middlings and finished products. This paper examines some new approaches to automation and integrated use of X-ray spectral and X-ray phase analyses by describing the techniques developed by the authors that enable to monitor the mineral phase composition of gold containing sulphide ores and appropriate concentrates, as well as the composition of aluminium cell electrolytes. Particular attention is given to the automation of the X-ray phase analysis method on the basis of Rietveld analysis as the most advanced technique for monitoring the mineral phase composition based on the actual crystalline structure of constituent phases. The paper also analyses the accuracy of the developed methods.