G. Guardabassi - Academia.edu (original) (raw)

Papers by G. Guardabassi

IFAC Proceedings Volumes, 1979

The aim of this paper is to discuss th..e problem of improĩ ng the stability of low-frequency osc... more The aim of this paper is to discuss th..e problem of improĩ ng the stability of low-frequency oscillations in longitudinal ne! works by making the internal generator e.m.f. of a single machine dependent on the current speed through a suitable transfer function. In particular, the question. of establishing to what an extent the network stability can be improved by a dynamic rather than a purely alg'tlbraic compensator is addressed and discussed in depth.

IFAC Proceedings Volumes, 1985

ASstract. This paper deals with the problem of qualifying the class of linear dynamical tystems w... more ASstract. This paper deals with the problem of qualifying the class of linear dynamical tystems which can b e r e presented as time-varying ARMA models. It is proved that any rstep uni fo rmly observable system admits an ARMA representation of order r and. conversely. an ARMA model of order r can be represented by means of an r-step observable system. ~inimality of ARMA models is then discussed. It is shown that the ARMA representation of certain systems requires more parameters than the correspondent minimal state-space realizati on.

IEEE Transactions on Circuits and Systems, 1974

ABSTRACT Further results on minimal essential sets in directed graphs are presented and an indire... more ABSTRACT Further results on minimal essential sets in directed graphs are presented and an indirect algorithm based on the analysis of the complementary graph is derived.

IEEE Transactions on Circuit Theory, 1971

Consider the network and digraph shown in Fig. Z(a) and (b), respectively. The composite graph is... more Consider the network and digraph shown in Fig. Z(a) and (b), respectively. The composite graph is shown in Fig. 2(c). Step 1: Graphs G,, and G, are shown, respectively, in Fig. 2(d) and (e). Step 2: The k-tree terms for Fig. 2(e) are listed in Section II and are given there in (l)-(14). Step 3: All possible directed trees .which can be used to form nonduplicated edges of G, are shown in Fig. 3(eHh). Note that the null tree (symbol +) is included. Directed tree admittance products and vertex sets are shown in Table I. Step 4: For each row of Table I perform the product of column 2 and the terms indicated in column 5. Step 5: Summing these products, the determinant D of the node admittance matrix of N becomes D = Y~YIYL + Y~YL) + Y~(YIY. + Y~YL) + YO.(YIY~ + YIYL) + (Yl + Ye + YL). [YdY" + Yoc) + YtlYocl + YIYCYL '-gm[Yr(Yl + Ye) + Y1Ye + YlYO,l + Srn[YU(Yl + Yd + YIYL + Y1Yo.l + !Jrn[YdYl + Ye) + YlYe + YILY,l. Cofactors of elements of the determinant D are computed in the same manner by computing directed k-tree terms. Hence, the algorithm permits computation of any network function for which a node admittance matrix can be written.

1985 24th IEEE Conference on Decision and Control, 1985

In this paper, the periodic Riccati equation is considered. It is proved that a unique periodic p... more In this paper, the periodic Riccati equation is considered. It is proved that a unique periodic positive semidefinite solution exists and the associated closed-loop system is asymptotically stable if and only if the underlying periodic system is stabilizable and detectable. The proof is based on the decomposition of the Riccati equation obtained by considering the system decomposed in the Kalman

IEEE Transactions on Automatic Control, 1972

ABSTRACT A necessary and sufficient condition for state terminal insensitivity in time-invariant ... more ABSTRACT A necessary and sufficient condition for state terminal insensitivity in time-invariant linear systems is given for the particular case of a single input and a single parameter. It leads to a sufficient condition for the construction of an algebraic state feedback which allows the system to be terminal insensitive.

IEEE Transactions on Automatic Control, 1971

Discussed in this short paper is a controllability question pertaining to a class of sensitivity ... more Discussed in this short paper is a controllability question pertaining to a class of sensitivity problems. More specifically, the possibility is considered of transferring a time-invariant linear system from each initial state to any terminal state in a finite time and in such a way that the terminal condition is insensitive to small parameter variations. Partial controllability, suitably defined, leads to a necessary and s d c i e n t condition for the existence of the solution to such a problem. Terminal output insensitivity is also discussed and a similar result is proved when the output transformation isnot affected by some parameter variations. ommended by D. G. Luenberger, past' ChaiAan of the IEEE %-.4C Linear Sys-Manuscript received September 11 1970-revised April 26 19il. Paper recterns Committee. This work was supported by the Consiglio Nazionale delle Ricerche. Elettrotecnica ed Elettronica. Politecnico di Milano, Milsno. Italy. The authors are with the Laboratorio di Controlli Xutomatici, Istituto di

Operations Research, 1970

This paper deals with two problems of asymptotic aggregation of an ergodic Markov chain. The firs... more This paper deals with two problems of asymptotic aggregation of an ergodic Markov chain. The first is concerned with the determination of an aggregated Markov chain, the asymptotic probabilities of which are just equal to conditional asymptotic probabilities defined on the original Markov chain. The second one relates to the determination of an aggregated chain, the asymptotic probabilities of which are equal to the asymptotic probabilities of the elements of a given state partition. Both problems are approached from a topological point of view and solutions are given in terms of signal-flow graph techniques. Comments about the main computational features of the method are also included.

IFAC Proceedings Volumes, 1975

Systems & Control Letters, 1997

ABSTRACT The design of feedback-linearization and poles-placement controllers for discrete-time n... more ABSTRACT The design of feedback-linearization and poles-placement controllers for discrete-time non-linear plants, using Input/Output/State measurements only, is typically addressed via indirect design. In this paper we propose the use of a new technique, based on a Virtual Input Direct Design (VID2) approach. The main feature of such a technique is to reduce the control design problem into a standard non-linear mapping approximation problem, without calling for the preliminary construction of an appropriate model of the plant. As compared with the existing methods, the new one requires less computational effort, while taking full advantage of the non-linear approximation software tools already available. In this paper, the new method is described, a simple theoretical analysis is given, and some numerical examples are presented.

Journal of Combinatorial Theory, Series B, 1972

The main problem is that of counting partial subgraphs (or patterns, for short) in a given graph ... more The main problem is that of counting partial subgraphs (or patterns, for short) in a given graph by inspection of the complementary graph (indirect method). As an illustration of the general approach described, specific formulas are derived for Hamiltonian circuits and complete subgraphs of any order.

Nonlinear Control Systems Design 1989, 1990

Roughly, a model is said to be nicely nonlinear if it is readily linearizable by output feedback.... more Roughly, a model is said to be nicely nonlinear if it is readily linearizable by output feedback. This paper is primarily concerned with “control-oriented nicely nonlinear modelling”; namely, with the problem of finding, within a specific class of nicely nonlinear models, the one that best fits with significant aspects of the (external) behavior of a given plant, relative to a particular control problem defined on it. The approach proposed here is based on the least square fitting, over a given frequency range, of the plant “nominal” describing function. As an application, a simulation study concerning a binary distillation column is presented. Possible extensions and some current investigations are briefly discussed in the concluding remarks.

Summary for each entry ors of A and each entry Brs of /B C'j . The LQG approach to an optimal... more Summary for each entry ors of A and each entry Brs of /B C'j . The LQG approach to an optimal cyclostationary control problem is presen5ed. The periodic solu- The vector w = /w1 wpl is a cyclostationary Wiener tions of a periodic Siccati differential equation, random, process of period T i.e. a Wiener random for which a sufficient condition is given, are then process the incremental covariance Wdt of which is shown to pleg a cer-tral role in the problen above. ?.-periodic. The classes of admissible control and A "

Information on control systems is presented concerning the following topics: sensitivity analysis... more Information on control systems is presented concerning the following topics: sensitivity analysis; sensitivity methods; techniques of sensitivity reduction; optimization under uncertainty; adoptive systems; min-max strategies and differential games; and large scale systems. A separate abstract is included for one paper concerning the control of nuclear reactors. (DCC)

Control Science and Technology for Development, 1986

Journal of the Franklin Institute, 1971

The problem of counting spanning trees in linear graphs is dealt with. A quite general indirect a... more The problem of counting spanning trees in linear graphs is dealt with. A quite general indirect approach, based on the analysis of the complementary graph, is followed and a new counting formula, which applies to any kind of linear non-oriented graph, is given. The paper ends with some comments and concluding remarks.

This paper presents a survey of the early development of automation technology and control scienc... more This paper presents a survey of the early development of automation technology and control science in Italy. The main focus is on the crucial period from 1950 through 1970. Some attention, though, has been paid also to more ancient significant events, from the Middle Ages to the first half of the XXth century. Equally included are some hints to events that occurred in the decades following 1970. However, these more recent scientific achievements are better understood within the frame of the main international stream rather than in any local perspective. As such, they have been somewhat overlooked in the present survey.

Directions in Large-Scale Systems, 1976

Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304), 1999

... As for v;,(Q), we have: 0 , M#O 3 A(M)=m anya~[-b, b] , M=O 3 A(M)= 1 M#=O , A#=1; thus, v&am... more ... As for v;,(Q), we have: 0 , M#O 3 A(M)=m anya~[-b, b] , M=O 3 A(M)= 1 M#=O , A#=1; thus, v&Q) = A' = 1; ie, according to v;, our system S is "maximally nonlinear" (its best linear approximation, in the sense of tiu, is the null operator). ...