Geometry of quadratic differential systems in the neighborhood of infinity (Q2484460): Difference between revisions

The authors are concerned with the class of real planar quadratic systems (\textbf{QS}) \[ \frac{dx}{dt}=p(x,y),\qquad\frac{dy}{dt}=q(x,y), \] where \(p\) and \(q\) are polynomials in \(x\) and \(y\) with real coefficients and \(\max(\deg(p), \deg(q))=2\). The main goal is to obtain a global geometric scheme of singularities at infinity in invariant form for the whole class \textbf{QS}. The set of \textbf{QS} depends on 12 parameters -- coefficients of the polynomials \(p\) and \(q\), but due to the action of the group of affine transformations and of change of scale of the time, the orbit space of \textbf{QS} is five-dimensional. Nevertheless, it is expected that this class contains more than 2000 topologically distinct phase portraits. As the interest is in global behavior of all solutions in the whole plane and at infinity for the whole family of such systems, the problem is really hard. To achieve the goal, the authors use global algebro-geometric concept (e.g., the notion of index divisor and of zero-cycle on the line at infinity) and state (and prove) two classification theorems. The first one (Theorem~5.1) classifies the \textbf{QS} according their multiplicity divisors at infinity. They give necessary and sufficient conditions for each class in terms of algebraic invariants and comitants with respect to the group action. The second one (Theorem~7.1) -- the topological classification theorem employs geometric, affine integer-valued invariants, and algebraic invariants and comitants. In this theorem, they give a complete dictionary connecting the integer-valued geometric invariants (encoding the geometry of the system at infinity) with the algebraic invariants and comitants (enabling usage of the power of computer algebra). A first version of this article appeared by the authors [Geometry of quadratic differential systems in the neighbourhood of the line at infinity, CMR Report no.~2701, Université de Montréal (2001)]. An analogous work for the finite singularities is ``presently in progress''.