Abstract—The main purpose of the motor mode is to keep the turbine in hot standby (with the boiler off) without generating electricity to the grid. To do this, the generator is switched to the engine mode, which rotates the turbine rotor, and the cooling of the f low path is performed by supplying one or two intermediate steam extractions from the extractions of the neighboring turbine. The first stages of the high-pressure cylinder run on steam from the front seals. The noted specificity of the purpose and operating conditions of the turbine in low-steam modes imposes additional requirements on the mathematical model. When developing models of a turbine stage in a motor mode, the physical media included in the object were taken into account: the steam of the f low path passing through the nozzle and working discs, participating in the conversion of its thermodynamic and kinetic energy into mechanical energy of the rotor rotation; vapor leakage into the space between the discs; steam leakage through the rotor-guide vane seal; metal of rotating devices (working disks and turbine rotor); and metal of the stator and exhaust pipes. Multipoint models are obtained with lumped parameters distributed along the length of the f low channel with values that vary depending on the ordinal number of the stage. The main part of the step models are eight ordinary differential equations or partial differential equations of physical media. The developed models are necessary for computer modeling and investigating the temperature distribution of the turbine f low path. Based on the research results, it is proposed to create a real automatic control system that ensures the implementation of restrictions on metal heating and the generation of added power.

The relaxation algorithm of the selected coordinate descent has been developing to control the underground seismic structure beneath the industrial subjects. The distinctive feature of the algorithm is in the interdisciplinary approach to choose nonzero components of the starting point. Application of the algorithm to seismic data recorded in complex geological conditions has shown an improvement in statistics characterizing the accuracy of the inversion solution. Based on seismic images, we have studied the cause of the 1976 Kopasker earthquake in northeastern Iceland. We concluded that the drilling of production geothermal wells was hardly the main factor provoking this event in the region. For the planning of fuel energy subjects, the following result might be using. The location of the earthquake epicenter is at the boundary between two contrasts: high and low seismic velocity areas.

Visual odometry is a well-known technical problem, which is crucial for navigation of autonomous mobile robots in buildings or hard-to-reach areas where no precise external navigation is available. We propose an architecture of a neural network capable of solving the stated problem. Training of the network is performed based on data collected within a virtual environment. Additionally we suggest an algorithm for filtration of gathered data based on sensor fusion methods. The inputs of the whole system are two sequential videoframes and the subset of inertial sensor readings. The output of the proposed system is a vector of motion parameters containing information on robot's linear and angular velocities. The conducted experiments confirm adequacy and show possible application of the proposed network architecture and filtering algorithm for searching people lost in a forest.