The aim of the study is to identify the influence of some parameters of the geological environment on the formation of the current hydrogeochemical profile in the Middle Jurassic hydrogeological complex (layers YU2-4) of the Lyaminsky oil and gas field in Western Siberia.
The factors controlling the changes in the chemical composition of the groundwater (geotemperatures, influence of neotectonics, etc.) after the process of sedimentation and water accumulation were analysed on the basis of the lithogenetic theory of the transformation of the composition of groundwater during geological evolution. In the course of the research, correlation coefficients were obtained for the value of groundwater mineralisation in the complex with the modern and paleotemperatures of the Tyumen suite, the porosity coefficient and the depth of the foundation. The obtained results allowed assuming that the current hydrogeochemical profile of the Middle Jurassic hydrogeological complex is the result of the consistent action of the whole complex of factors. The results of the analysis showed that the influence of neotectonic processes on the current hydrogeochemical profile is quite probable. The reasons for the heterogeneity of mineralisation, apart from neotectonic processes, may be different, but in any case, the identification of relationships such as "Mineralisation - a parameter of the state of the geological environment" indicates a change in the concentration of salts in groundwater after their accumulation together with sediment.

At the current stage of geological exploration, hydrocarbon deposits have been discovered in complex reservoirs that are "unconventional" in nature. In particular, at the Medvezhye field, according to the interpretation of seismic surveys and the drilling of exploratory wells, a gas deposit was discovered in the Senonian deposits of the Nizhneberezovskaya suite of Upper Cretaceous deposits. The gas deposit of the NB formation of the Medvezhye field is characterised by productive reservoirs, mostly represented by mudstones and opoka-like clays. These rocks are characterized by low filtration parameters. Commercial gas inflows from secondary reservoir openings indicate the presence of fractured rock, as confirmed by core testing, mechanical logging. A comprehensive analysis of hydrodynamic studies of field parameters, such as reservoir pressure, drawdown, flow rate, allows calculating the filtration characteristics of a productive formation. The performance of fractured and mixed pore-fractured reservoirs is determined using induction diagrams and pressure recovery curves, the reservoir hydrodynamic model and methods for improving development in complex "unconventional" reservoirs are refined.

The algorithm of analog-to-digital conversion of primary geological and geophysical information (on the example of identification of rock lithotypes based on the text description of the physical core) is presented.
As part of the work, a combination of three types of scientific research - prospecting, interdisciplinary and applied, in the formation of the initial base of qualitative data is implemented.
Common algorithms for textual information classification and mechanism of initial data preprocessing using tokenization are described.
The concept of text pattern recognition is implemented using artificial intelligence methods.
For creation of the neural network model of textual geological and geophysical information recognition the Python programming language is used in combination with the convolutional neural network technologies for text classification (TextCNN), bi-directional long-shortterm memory networks (BiLSTM) and bi-directional coder representation networks (BERT).
The stack of these technologies and the Python programming language, after developing and testing the basic version of the neural network model of qualitative information recognition, provided an acceptable level of performance of the algorithm of digital transformation of text data.
The best result (the current version of neural network model is 1.0; more than 3000 examples for training and testing) was achieved when using the algorithm of text data recognition based on BERT with an accuracy on the validation network (Validation Accuracy) ~0.830173 (25^th epoch), with Validation Loss ~0.244719, with Training Loss ~0.000984 and probability of recognition of the studied rock lithotypes more than 95 %.
The mechanisms of code modification for further improvement of textual prediction accuracy based on the created neural network were determined.

In the Izhma-Pechora syneclise, several fields have been discovered that are limited to sand deposits deposits of stream-bed genesis. However, it is difficult to find lithological traps using seismic data due to the complex nature of the distribution of sandy deposits and many discontinuous faults in the Permian terrigenous section. The article attempts to model the continental depositional environment that existed in the Izhma-Pechora syneclise and the Upper Pechora depression conjunction zone during the Ufimsky, Kazansky and Tatar ages of the Permian period. It aims to reconstruct the history of the development of the area and to trace the distribution of alluvial and lacustrine-alluvial sand deposits. DionisosFlow's unique industrial package helps to create a 3-dimensional model of the development of the area, to simulate the process of sedimentation of the target area, and to construct predictive maps of sandstone, siltstone and clay spread. The western part of the modeled area demonstrates the existence of slightly wandering north-easterly directed beds and their small offshoots. The south-eastern part of the area, dated to the Ufimsky age, shows deltaic deposits. The method of predicting the distribution of channel deposits based on sedimentological modeling is supposed to be used for territories uninvestigated by 3D survey. The aim is to locate the source of the reservoir and plan the seismic survey.

The article presents the process of forming measures based on digital technologies to improve the quality of well cementing at the fields of Western Siberia. The problem associated with the low quality of input information due to the use of several independent sources was identified and solved. The economic efficiency of the developed methods for reducing the labour costs of data collection for modelling using machine learning algorithms is demonstrated. If the solutions developed are implemented, there is a prospect of reducing the cost of repair and insulation work. Key information is provided about the hypotheses generated and their objectives. The authors of the article describe the method of using various mathematical algorithms to analyze the results of industrial experimental work. The efficiency of the developed solutions is evaluated by comparing the results of cementing experimental wells and wells built using the basic technology. The dynamics of cement quality growth in the fields of Western Siberia are summarised as a general result. As a result of the experience gained, the solutions have been adapted and are in the process of being re-implemented in order to make a final assessment of their effectiveness.

Under certain conditions, above-ground sections of pipelines are subject to vibration. Increasing the vibration amplitudes beyond the limits can cause the object to fail. In order to assess the vibration resistance of pipelines and to find the most effective ways of detuning from the resonant frequencies of disturbances, it is necessary to determine the natural frequencies of oscillations and the characteristics of the oscillation processes. The authors of the article developed a mathematical model to determine the frequency of free vibrations of above-ground pipeline sections, taking into account the influence of internal pressure and the velocity of the pumped fluid. As a result, a numerical solution based on the Bubnov-Galerkin method was obtained and a sufficient number of series elements in the approximate solution was determined. The authors performed numerical experiments. These showed that the frequencies should be determined for each section under specific conditions according to the pumping mode.

The article considers options to reduce heat losses accompanying the process of pipeline transportation of natural gas through long branch pipelines. Computational modeling of the gas cooling process in an extended off-take gas pipeline connecting the main gas pipeline and the gas distribution station has been performed. It is shown that when the length of the gas branch pipeline is long, especially in conditions of reduced flow, the gas temperature at the inlet of the gas distribution station will correspond to the ground temperature. We have developed and substantiated options to reduce the intensity of gas cooling in the off-take gas pipeline, involving an increase in its depth, the use of thermal insulation coatings, as well as the regulation of pressure at the inlet using special technical means. The calculation justification of the technical solution, which involves regulating the pressure at the inlet to the gas pipeline outlet, is performed. The results can be used by gas transportation companies. The implementation of the developed solutions to reduce heat losses will reduce the consumption of energy used for gas heating at gas distribution stations and reduce emissions of combustion products.