On the maximum number of limit cycles for a piecewise smooth differential system (Q2208967): Difference between revisions

The article considers the system \[\dot{x}=-y,\quad \dot{y}=x+\varepsilon \left(1+\cos ^m\theta \right)Q_k(x,y),\] where \(m\in\mathrm N\cup \{0\}\), \(\theta =\arctan y/x\), \(Q_1 \left( {x,y} \right),Q_2 \left( {x,y} \right)\) polynomials of degree \(n\), given in the upper half-plane (\(k=1\)) and lower half-plane (\(k=2\)). The first-order averaging method is used to calculate the maximum number \(L\) of limit cycles bifurcating from periodic solutions of a linear center \(\ddot {x}+x=0\) under discontinuous perturbations and for \(\varepsilon >0\) sufficiently small. It is proved that \(L=n\) for any values of \(m\). It follows from the proof of the main result of the article that the estimate \(L=n\) can be achieved by setting \(m=0\), \(Q_2 (x,y)=0\), \(Q_1 (x,y)=Q(y)\), where \(Q(y)\) is a polynomial of degree \(n\). Therefore, it is not clear how the presence of the factor \(1+\cos ^m\theta \) in the perturbation of the linear center is justified. This multiplier does not affect the result in any way. In [\textit{T. Chen} and \textit{J. Llibre}, Appl. Math. Lett. 88, 111--117 (2019; Zbl 1416.34021)], which the authors refer to, this factor plays an essential role.