Home – Home – Jornals – Avtometriya 2022 number 4

The paper considers the problem of control of movable objects with distrib-uted parameters, optimal according to standard quality criteria, concerning the actual goal of optimization of the operational temperature modes for continuous process instal-lations for preheating semi-finished products before subsequent pressure treatment. The solution to the problem provides the optimal energy consumption distributions of the heat sources' power along over the length of the heater in stationary modes of its operation, which act as the desired design solutions for the installation. Algorithms for automatic programmed and positional control of operational transient modes of a heating installa-tion are proposed that are optimal in terms of a standard quadratic quality criterion under conditions of a given accuracy of uniform approximation of the final spatial temperature distribution to the required one. The procedure for synthesizing an optimal controller is reduced to constructing a control system with linear feedback based on incomplete measurements of the object's state with non-stationary transmission coefficients determined by a preliminary calculation of programmed control.

To solve the problem of compensation of the influence of a perturbation on the angular position of the spacecraft, an engineering technique is proposed for the synthesis of an algorithm for controlling the flywheel orientation and stabilization system, based on ensuring selective invariance of the control system and the anti-accumulation signal scheme. The effectiveness of the method is confirmed by numerical modeling.

To solve the problem of estimating the orientation and position of an aircraft, a system based on series-connected Kalman filters is proposed: a linear Kalman filter in the attitude stabilization loop and an extended Kalman filter in the subsystem of maintaining the target location. The effectiveness of the developed control system is confirmed by experimental results of the quadrocopter indoor flight control.

The problem of monitoring the metal particles of friction pairs’ wear in lubrication systems of power plants is considered. The solution is proposed on the base of a multi-channel cluster eddy current sensor with single-coil sensitive elements. The sensor is installed in the oil pipeline break and provides the identification of ferromagnetic and non-ferromagnetic metal particles, as well as their ranking by several size groups. The features of the conversion of sensing elements’ inductances in differential measuring circuits during the movement of metal particles through the oil channel of the sensor are considered. The effect of the particles’ speed in the oil flow on the result of the conversion of sensor signals is analyzed. The algorithm for recognizing and assigning metal particles to specified size groups is given. Considering the possible effect of the particles’ speed on the result of the conversion of sensor signals, the operational correctness of the algorithm is evaluated.

The problem of estimation of the force loading channel gain of the electrohydraulic strength test bench is considered, in particular, on the basis of the frequency response analysis, the gradient method, the least squares method with sliding window, and the recursive least squares method with forgetting factor and based on the harmonic analysis. The calculation relations of the algorithms for obtaining an estimate of the force loading channel gain are given, the results of implementation of these methods on the experimental bench are presented, and a comparison of the estimates obtained by different methods is performed. As a result of the conducted research, recommendations are given for building an adaptive control system with the force loading channel coefficient identification loop on the aircraft structures strength test stand. The obtained results make it possible to automate the process of regulators’ adjustment for electric hydraulic actuators of the stand and to speed up the process of strength and fatigue strength tests of aircraft structures.

In modern robotics, the questions of local navigation of mobile robots (MRs) providing the purposeful movement of the robot in an environment with obstacles gain all the most urgency. At the same time, the most popular are the methods of local navigation of the MR based on the use of artificial potential fields (PFs) generating virtual forces of "attraction" to the target and "repulsion" from obstacles. However, their application faces the problem of potential pits, which are local "traps" for MRs. The new method of potential guidance of the MR that is based on the concept of a "fairway" and allows one not only to solve the problem of "traps," but also to choose an optimal route of the robot's movements to the target is stated in the paper. The constructive aspects of the "fairway" method are discussed, and a virtual particle method is proposed to find the desired "fairway" on the map of force lines of the PF.

Solving problems of environmental monitoring reasonably requires collecting, digitizing, storing, and analyzing a large amount of spatiotemporal data. Data processing often needs distributed computing. As a rule, there are no means to fully support such computing in the well-known geographic information systems. In the paper, we propose an approach to solving this problem based on integrating a geographic information system with tools for developing and using distributed scientific applications in a heterogeneous computing environment through representing applied software as web data processing services. We show the advantages of the proposed approach in reducing the complexity of preparing and carrying out experiments with geodata.

The problem of improving the synchronization accuracy of three-phase voltages of an autonomous four-wire inverter with a network is considered. Increasing the synchronization accuracy is achieved by applying a proportional-integral-differential (PID) resonant controller. A new structure of the PID resonant controller and a methodology for calculating its parameters are proposed. The peculiarity of the proposed approach consists in the application of the time-scale separation method, which makes it possible to calculate the PID-component of the controller regardless of the choice of the parameters of its resonant components. Application of this approach to the calculation of the controller of autonomous voltage inverters makes it possible to improve electromagnetic compatibility and reduce losses in distributed power generation systems. The results of numerical simulation are presented.

A constructive technology for solving parametrizable problems of two-channel time-optimal boundary control of two physically inhomogeneous objects with distributed parameters described by linear heat conduction equations under conditions of a given accuracy of uniform approximating the resultant spatial distribution of the controlled variable to the desired state is proposed. The two objects are considered under conditions of perfect thermal contact of interacting surfaces. The developed technology uses a special method for describing the studied temperature fields in a form containing uniformly converging series. The developed approach is based on the parametrization procedure of the sought control actions and subsequent reduction to a special form of the task of semi-infinite optimization, which is solved by the scheme of the alternance method proposed earlier with an additional restriction on the same duration of the control process for two control actions.

In this paper, the possibility of stabilizing algorithm application for a linearized system “Double inverted pendulum” by means of the localization method is discussed. In order to reduce the complexity of the controller calculation procedure, a preliminary transformation of the object definition to the canonical basis is provided. Moreover, the control algorithm formulation in the form of a function of the canonical variables is suggested. The controller definition in a primitive basis is obtained via an inverse transformation of the canonical variables. It provides a possibility of issuing recommendations for the physical implementation of the system. In addition, it is needed to estimate the entire list of variables for creating the controller, but only the part of internal variables of the object can be measured. With the purpose of the variables’ estimation, three similar differentiators are proposed, which provides the possibility of estimating the total state vector. The designed controller guarantees the pendulum’s stabilization in a vertical position with a small angle of deviation. The main features of the designed system are shown by simulating the model in the MatLab Simulink integrated environment.

A possibility of using different ways of adjusting parameters of a proportional-integral-differential (PID) regulator in a system with a second-order object with a transport delay is discussed. According to the experimentally obtained acceleration parameter, the transfer function of the object and its parameters are determined, the quality indicators of the transient process are evaluated, and the synthesis task is formulated. Because of a significant delay of the control object, it is planned to investigate some ways of regulator tuning in order to choose the most appropriate one: Ziegler-Nichols’ method (frequency method), Chyn-Chrones-Resvik’s method, Kun’s method (fast adjustment), Stogestad’s method, and method of numerical optimization with various cost functions are investigated in this project. Based on the calculation results, the simulation of synthesized systems in the integrated MatLab Simulink and VisSim environments is performed. A comparison of graphs of transients obtained by using different methods is carried out. It is shown that, among all the approaches studied, only the numerical optimization method allows providing a transient process with desired properties at the output of a closed system.

In a number of systems for non-contact tracking of the gaze direction based on video data generated by a video camera, the gaze direction is determined by the mutual position of the centers of the pupil and the cornea of the eye. In the camera coordinate system, the three-dimensional coordinates of the cornea center can be calculated from the positions of glares created on the cornea surface by point sources of IR radiation rigidly connected to the camera. The paper considers a new method for estimating the three-dimensional coordinates of the cornea center from images of glares, generated by a pair of point radiation sources. It is shown that the problem comes down to solving one nonlinear equation, where the distance between the eye and the camera is much greater than the radius of the cornea. We present an iterative algorithm for estimating the coordinates of the cornea center, which ensures low computational costs. The results of the numerical experiment confirm the preservation of the estimates accuracy.

The article considers a new method for reconstructing images from holograms with high spatial resolution, based on hologram shifts by a value smaller than the aperture size. The method is based on the discretization equation obtained using generalized functions. Mathematical modeling of the process of restoring a high-resolution image from low-resolution holograms is carried out. The proposed method can be used to obtain digital holograms using a photodetector array with a low resolution. The method does not require a direct solution of a system of algebraic equations of high dimension.