Search

Issues of design and research of energy-efficient and reliable semiconductor DC voltage converters for wide application in power supply devices of arc plasmatrons used in plasma metal cutting installations are considered. An estimated structural dynamic model of a soft-switching DC converter with a closed-loop control system combining the power part and the control system, intended for use as part of the latter as a digital module, was built. A technique is proposed and methods of controlling the converter are found, which provide the specified duration of transient processes, the permissible value of load current pulsations in a quasi-steady mode and the astatism of the output current, which confirms the correctness of the proposed technique. A prototype of a pulse stabilizer with digital control was made. The results of experimental studies of the sample confirm the effectiveness of the developed control device, namely the achievement of the specified duration of transient processes caused by a step change in the load current, close to 10-12 periods of conversion and astatism of the output current. It is shown that the application of a pulse stabilizer, in which a digital control loop is used, has significant advantages compared to analog options and has the advantages of a strategic plan. The use of combined control allows you to significantly reduce the requirements for the overall gain of the main control channel, which greatly facilitates the choice of sequential digital correction. Research results may be of interest to specialists in the field of power electronics, power supply systems of autonomous objects and control systems. References 14, figures 11. [ABSTRACT FROM AUTHOR]

the purpose of this paper is to substantiate the procedure for analyzing the quality of multidimensional closed-loop control systems. Quality indicators in each case may be different, and there are many such indicators. This is due to the variety of control systems and the various tasks they perform. The object of quality analysis is the Stewart platform working surface motion control system. The Stewart platform has six degrees of freedom: three translational degrees of mobility (displacement of a characteristic point along the axes of a fixed coordinate system associated with a fixed base) and three rotational degrees of mobility (rotation of the moving platform relative to its axes of the moving coordinate system associated with the moving platform). As part of the study of the dynamic design of optimal multidimensional stochastic stabilization systems, the problem of maximizing the quality of the Stewart platform working surface motion reproduction is formulated. One of the components of this problem is to determine the principles of selection and algorithm for calculating the quality indicators of the Stewart platform working surface motion control system under operating conditions, i.e., when deterministic components of useful signals, disturbances, and interference dominate. When the determining value of the quality assessment of the control system has a deterministic component, the quality indicator is the sum of the weighted integral quadratic indicators of the errors of maintaining the position of the Stewart platform working surface on the program trajectory and the values of the control vector, which characterizes the power required for control. [ABSTRACT FROM AUTHOR]

Context. The presence of time delays occurs in many complex dynamical systems, particularly in the areas of modern communication and information technologies, such as the problem of stabilizing networked control systems and high-speed communication networks. In many cases, time-delays lead to a decrease in the efficiency of such systems and even to the loss of stability. In the last decade, many interesting solutions using the Lyapunov-Krasovskii functional have been proposed for stability analysis and synthesis of a stabilizing regulator for discrete-time dynamic systems with unknown but bounded state-delays. The presence of nonlinear constraints on the amplitude of controls such as saturation further complicates this problem and requires the development of new approaches and methods. Objective. The purpose of this study is to develop a procedure for calculating the control gain matrix of state feedback that ensures the asymptotic stability of the analyzed system, as well as a procedure for calculating the maximum permissible value of the state-delay under which the stability of the closed-loop system can be ensured for a given set of admissible initial conditions. Method. The paper uses the method of descriptor transformation of the model of a closed-loop system and extends the invariant ellipsoids method to systems with unknown but bounded state-delays. The application of the Lyapunov-Krasovskii functional and the technique of linear matrix inequalities made it possible to reduce the problem of calculating the control gain matrix to the problem of semi-definite programming, which can be solved numerically. An iterative algorithm for solving the bilinear matrix inequality is proposed for calculating the maximum permissible value of the time-delay. Results. The results of numerical modeling confirm the effectiveness of the proposed approach in the problems of stabilizing discrete-time systems under the conditions of state-delays and nonlinear constraints on controls, which allows to recommend the proposed method for practical use for the problem of stability analysis and synthesis of stabilizing regulator, as well as for calculating the maximum permissible value of time-delay. Conclusions. An approach is proposed that allows extending the invariant ellipsoids method to discrete-time dynamic systems with unknown but bounded state-delays for solving the problem of system stabilization using static state feedback based on the application of the Lyapunov-Krasovskii functional. The results of numerical modeling confirm the effectiveness of the proposed approach in the presence of the saturation type nonlinear constraints on the control signals. [ABSTRACT FROM AUTHOR]

The article discusses an automatic control system that implements the principle of predictive control in a closed loop. Objects of a technological type quite often have a large inertia in the control channels associated not only with a pure delay, but, to a greater extent, with the accumulation of matter and energy, the so-called capacitive delay. Full or partial compensation of this inertia can significantly improve the quality of regulation for such objects. Therefore, it is proposed to introduce prediction based on a cubic spline into the closed loop of the control system. The performed analysis shows that individual sections of the trajectory of the controlled variable can be represented by continuous, multiply differentiable functions (for example, a cubic spline). For the model of cubic splines, relations are found that allow estimating their parameters and calculating the predicted value of the variable in real time. Structural and optimal parametric synthesis of alternative variants of automatic control systems with a typical PID controller, which implement the principle of control by forecast and the principle of control by state, has been carried out. A comparative analysis of optimal systems, carried out in the time and frequency domains, showed a significant advantage of the control system that implements the principle of predictive control based on a cubic spline. In comparison with a system that implements the principle of state control, a control system with predictive cubic spline provides a decrease in the integral and direct quality indicators of transient processes by up to 40%. Checking for the roughness of automatic control systems showed that an automatic control system with predicting an adjustable variable along a cubic spline has a greater margin of stability than a control system that implements the principle of state control. [ABSTRACT FROM AUTHOR]