Search

Questions related to control of the performance of a measuring techniques are considered. Analytic and computer models of false failure and undetected failure (error functions) in tolerance control of the parameters of the components of a measuring technique are developed. A geometric interpretation of error functions in the form of two-dimensional surfaces that depend on the tolerance of a controllable parameter and the measurement error is presented. The newly developed models may be used both for theoretical as well as statistical laws of distribution of a measurable quantity and a measurement error. The results obtained will be useful in the development of metrological support of a measuring technique, performance of verifications of measurement instruments, metrological examination of technical documentation, and certification of measurement methods.

The diversity of measuring tasks gives rise to a number of algorithms for data processing. The algorithms are included in the software of modern measuring instruments and make their own contribution to the resulting measurement error. A method is proposed for synthesis of algorithms for estimation of the dynamic error, the use of which makes it possible to eliminate computational difficulties that are characteristic for the traditional approach. This method is based on the methodology of the combined maximum principle. Analysis of the results of mathematical modeling in the problem of estimating dynamic error of the algorithms for data processing of measuring instruments confirms the effectiveness of using the proposed algorithm for estimating dynamic error of the software of measuring systems.

Technical measurements representing a measurement of higher achieved accuracy, are characterized by the fact that we have to deal with unpredictable errors caused by interference. At the same time, there remains the problem of increasing the reliability of received data and results of measurements while continuously increasing demands on transmission rate information and the reliability of the received data. The article presents a new approach to the transmission of information when using non-traditional representations received and transmitted messages.

The problem of control synthesis under uncertainty associated with observation noise is considered. In accordance with the separation method, it is assumed that the problems of state assessment and control synthesis are solved sequentially. An uncertain mathematical model of a controlled dynamic system is given up to an accuracy of the production rules, which, unlike the known solutions, are proposed to be found using the maximum condition of the generalized power function. This allows us to find the commuting hypersurface up to an accuracy of a function called the synthesizing function. For its construction the analysis of dynamic measures on the commuting hypersurface and approximation of the obtained solution taking into account the phase space decomposition are carried out. The uncertainty of the boundaries of the phase space decomposition into areas where there is a change of modes of the dynamic system operation is considered on the basis of fuzzy inference. The constructivity of the developed method is confirmed by mathematical modeling of the solution of the problem for control synthesis of nonlinear dynamic system.

The problems to optimize the dynamic systems are complex and traditionally can be reduced to a two-point boundary value problem. Therefore, the problem to obtain the control laws close to optimal law and suitable for practical implementation is always relevant. Performed mathematical simulation shows that the controls synthesis based on the analysis of the variation of the extended functional on the set of functions of bounded variation leads to a constructive result.

A method for intellectualization of measurement procedures based on step-by-step optimization procedure is developed. The procedure reduces the stochastic estimation problem to a number of deterministic problems. The goal of the first stage of optimization is the synthesis of adaptive mathematical model of the measuring process based on the combined maximum principle. In contrast to the known methods this ensures the constructiveness of its use at the next stages of optimization. The goal of the subsequent stages of optimization is to solve the estimation problem based on the regularization method of A.N. Tikhonov. The equations for the iterative measurement procedure are obtained. The structure of their state transition vector functions differs from the known equations structure. This procedure belongs to the category of intelligent measurement procedures, since it makes a targeted choice of the parameter estimation closest to the true value the under the conditions of structural uncertainty of the model of the studied object and of parametric uncertainty of the observation model.

An urgent issue is to obtain optimal control laws for nonlinear dynamic objects. The article presents a numerical algorithm to control stabilization of the angular velocities of spacecraft with various axial moments of inertia based on optimization with respect to the Krasovsky functional using the predictive model method. The simulation results are presented. The character of the curves of angular velocity dynamics and control moments obtained in the simulation corresponds to the curves of angular velocity dynamics and control moments obtained in the simulation of control using the analytical law for an axisymmetric spacecraft, which confirms the adequacy of the synthesized algorithm.

The intelligent measuring system must provide for changing the parameters of the operation algorithm in order to increase the measurements accuracy. In addition it must have a self-monitoring function realized, for example, on the basis of intelligent sensors. One of the common problem of intelligent systems and intelligent sensors is the automatic error correction as a result of external factors (GOST R 8.673-2009). The quality of its solution determines the efficiency of the functioning of the modern radar systems. This is due to the growth of maneuvering capabilities of aerodynamic targets, which caused an increase in the uncertainty of their movement due to the inadequate efficiency of the kinematic models used for solving the problems of the current estimation of the motion parameters. As a result, the problem of intellectualization of the measuring system arises on the basis of the development of adaptive tracking filters.

A new method of solution of the inverse measurement problem is developed based on the use of the apparatus of nonsmooth analysis of the minimum condition of the residual functional together with the technique of expanding the state space. A decrease in computational costs is confirmed by the results of mathematical modeling of the solution of the inverse problem under the conditions of parametric uncertainty for a temperature transducer.

The problem of optimization of dynamic systems is considered. It is shown that, unlike the well-known methods of optimal control, through use of the maximum principle it becomes possible to synthesize an efficient control law that substantially reduces computational complexity, as is demonstrated by the results of numerical simulation.