scientific article; zbMATH DE number 3858913
[For the entire collection see Zbl 0509.00039.] This paper deals with the generalized Fisher system: (1) \(\dot p=P(Gp- e\theta(p))\), where \(p=(p_ 1,...,p_ n)\), \(P=diag (p_ i)\), \(G=(g_{ij})\)- real matrix, \(\theta(p)=p^ TGp\), \(e=(1,...,1)\). We study the global behaviour of its solutions on the simplex \(\sigma =\{p:\sum^{n}_{i=1}p_ i=1\), \(p\geq 0\}\) which is the invariant set of the system (1). For the solution of this problem two types of energetical Lyapunov functions are introduced. Theorem I. If there exists an isolated equilibrium point \(\hat p\in Int \sigma\) and a diagonal matrix \(D=diag(d_ i)\), such that \(x^ TGD^{- 1}x<0\) if \(\sum^{n}_{i=1}x_ i=0\), \(x\neq 0\), then the function \[ E(p)=\sum^{n}_{i=1}d_ i\hat p_ i\ln(p_ i/\hat p_ i)-<d,\hat p>\ln<d,p>, \] where \(<\cdot,\cdot>\) an inner product, is increasing on the solutions of the system (1), i.e. Ė(p(t))\(>0\) if p(t)\(\in Int \sigma \backslash \{\hat p\}.\) Theorem 2. Suppose that \(G=D_ 1BD_ 2\), where \(B=B^ T\), \(D_ 1\) and \(D_ 2\)- diagonal matrices, such that the matrix \(D\equiv D_ 1^{- 1}D_ 2\equiv diag(d_ i)\) has all \(d_ i>0\), \(i=1,...,n\). Then the function \({\mathcal E}(p)= p^ TD_ 2BD_ 2/<d,p>^ 2\), where \(d=(d_ 1,...,d_ n)\), increases on all solutions of the system (1) on those \(\sigma\), which differ from the equilibrium points.
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 Q3328345)