Some results on simultaneous stabilization with multiple domains of stability (Q579201)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Some results on simultaneous stabilization with multiple domains of stability
scientific article

    Statements

    Some results on simultaneous stabilization with multiple domains of stability (English)
    0 references
    0 references
    1987
    0 references
    This paper considers the problem of simultaneous stabilization in the case where each plant has a different domain of stability associated with it. Specifically, it is supposed that there are given plants \(P_ 0,P_ 1,...,P_ r\) together with domains of stability \(D_ 0,D_ 1,...,D_ r\), and the objective is to find a common controller C such that the poles of the closed-loop transfer matrix \[ H(P_ i,C)=\left[ \begin{matrix} (I+PC)^{-1}\\ C(I+PC)^{-1}\end{matrix} \begin{matrix} -P(I+CP)^{-1}\\ (I+CP)^{-1}\end{matrix} \right] \] are all inside the corresponding domain of stability \(D_ i\) for \(i=0,...,r\). This is a more realistic formulation of the problem of preserving system stability in the event of sensor or actuator failures. Let \(M(S_ i)\) denote the set of matrices with elements in \(S_ i\) where \(S_ i\) denotes the set of proper rational functions whose poles all lie in \(D_ i\). Each plant \(P_ i\) has a doubly coprime factorization (dcf) such that \[ (2.1)\quad P_ i=N_ iD_ i^{-1}=\tilde D_ i^{-1}\tilde N_ i \] \[ (2.2)\quad \left[ \begin{matrix} Y_ i\\ -\tilde N_ i\end{matrix} \begin{matrix} X_ i\\ \tilde D_ i\end{matrix} \right]\left[ \begin{matrix} D_ i\\ N_ i\end{matrix} \begin{matrix} -\tilde X_ i\\ \tilde Y_ i\end{matrix} \right]=I \] where all eight matrices \(N_ i\), \(D_ i\), \(\tilde N_ i\), \(\tilde D_ i\), \(X_ i\), \(Y_ i\), \(\tilde X_ i\), \(\tilde Y_ i\) belong to the set \(M(S_ i)\). Suppose that one of the domains is a subset of the rest. Then, the following result is obtained. Theorem 2.1. Given the plants \(P_ 0,...,P_ r\), choose dcfs over \(S_ i\) for the plant \(P_ i\) for each i, as in (2.1)-(2.2), and define \[ \left[ \begin{matrix} A_ i\\ B_ i\end{matrix} \right]=\left[ \begin{matrix} Y_ 0\\ -\tilde N_ 0\end{matrix} \begin{matrix} X_ 0\\ \tilde D_ 0\end{matrix} \right]\left[ \begin{matrix} D_ i\\ N_ i\end{matrix} \right]. \] Then, there exists a controller C such that \(H(P_ i,C)\in M(S_ i)\) for \(i=0,...,r\) if and only if there exists an \(R_ 0\in M(S_ 0)\) such that \(A_ i+R_ 0B_ i\in {\mathcal U}(S_ i)\) for \(i=1,...,r\), where \(U(S_ i)\) denotes the set of unimodular matrices over \(S_ i.\) Section 3 considers the case of two plants \(P_ 0\) and \(P_ 1\) with \(D_ 0\) a subset of \(D_ 1\). The following result is established: the two plants can be simultaneously stabilized if and only if they can be simultaneously stabilized with respect to the larger of the two regions.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    robust control
    0 references
    frequency domain method
    0 references
    system integrity
    0 references
    simultaneous stabilization
    0 references
    domains of stability
    0 references
    0 references