scientific article; zbMATH DE number 1273484
automation of designcontrol accuracycontrollability and optimal solutionsdisturbancesfrequency domaingeneralized moments of spectral densitiesinput excitationmaximal values of derivativesnonadaptive automatic control systemsquality of systems of automatic controlreliabilityrobustnessspectral densitysynthesis
Reliability and life testing (62N05) Research exposition (monographs, survey articles) pertaining to systems and control theory (93-02) Algebraic methods (93B25) Sensitivity (robustness) (93B35) Synthesis problems (93B50) Frequency-response methods in control theory (93C80) Estimation and detection in stochastic control theory (93E10) Filtering in stochastic control theory (93E11)
In this book the methods of analysis and synthesis of nonadaptive automatic control systems and other dynamic systems capable of providing an acceptable or the best guaranteed level of functioning accuracy under an allowable uncertainty on excitation properties and system characteristics are stated. The control accuracy is estimated by an upper bound of the total root-mean-square or the maximal error. When specifying the classes of input excitations, the restrictions of possible values of derivatives, generalized moments of spectral densities, width of a spectrum and other numerical characteristics easily controlled in practice are used. The book does not cover all known methods of ensuring control accuracy, which are rather diverse and hardly can be united in a uniform theory except through the concept of robustness. The methods chosen deal in general with the frequency domain and the time domain based on mathematical results dealing with the general problem of moments, the theory of disturbances accumulation and numerical methods. The main advantages of such a choice are their relative simplicity and their high reliability for the representation of the initial data, which will be appreciated by a practitioner. The frequency domain approach is considered, too. The book can be read by a wide range of persons: designers, students, control scientists. The contents are: Dynamic systems. Kinds of external excitations. Indexes of successfully functioning systems. Concept of robustness. Comparison of low sensitive, adaptive and robust systems. Root-mean-square, maximal and practically maximal values of control error. Representation of excitation as the sum of additive components. Obtainment and analysis of a priori information about input excitations and their derivatives. Relationship between dispersions and maximal values of derivatives of various orders. Estimation of spectrum width using dispersions of derivatives. Experimental and theoretical estimations of correlation functions, spectral densities and dispersions of derivatives of casual processes. Optimal filtering under complete a priori information. Minimax robust filtering with strip model and \(\varepsilon\)-pollution model of spectral indefiniteness. Upper and lower bounds of control error dispersion analysis under the numerical characteristics of excitations. Use of canonical representations of a sequence of spectral density moments. Conditions justifying the use of equivalent harmonic excitations. Analysis of maximal control error, time domain and frequency domain methods. Synthesis of control systems with given accuracy of functioning. Principles of construction of forbidden areas in the frequency domain. Control systems optimization by the criterion of the best guaranteed accuracy. Estimation of potential accuracy loss at the expense of incompleteness of a priori information. Synthesis of open-loop and closed-loop robust digital systems. Robust prediction, smoothing and differentiating filters.
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4237333)
