Page 177 - Complementarity and Variational Inequalities in Electronics
P. 177

168  Complementarity and Variational Inequalities in Electronics


                           Proof. Let T> 0. We define the mapping V :[t 0 ;+∞) → R by the formula
                                                              ∗
                                                  ∗
                                                 V (t) = V(x(t;t 0 ,x 0 )).
                           The function x(·) ≡ x(·;t 0 ,x 0 ) is absolutely continuous on [t 0 ,t 0 +T ], and thus
                            ∗
                           V is a.e. strongly differentiable on [t 0 ,t 0 + T ].Wehave
                                      dV  ∗              dx
                                          (t) =
∇V (x(t)),  (t) , a.e. t ∈[t 0 ,t 0 + T ].
                                       dt                dt
                           We know (by hypothesis) that

                                               (∀t ≥ t 0 ) : x(t) ∈ D(∂ϕ) ∩
                           and

                             dx                                           n
                            
  (t)+F(x(t)),v−x(t) +ϕ(v)−ϕ(x(t)) ≥ 0, ∀v ∈ R , a.e. t ≥ t 0 . (5.50)
                             dt
                           Setting v = x(t) −@ V(x(t)) in (5.50), we obtain
                                    dx
                                   
   (t),∇V(x(t)) ≤−ϕ(x(t)) + ϕ(x(t) −@ V (x(t)))
                                     dt
                                                     − F(x(t)),∇V(x(t)) , a.e. t ≥ t 0 ,

                           and we may use (5.49) to get
                                             dx
                                            
   (t),∇V(x(t)) ≤ 0, a.e. t ≥ t 0 .     (5.51)
                                             dt
                           Thus
                                             dV  ∗
                                                 (t) ≤ 0, a.e. t ∈[t 0 ,t 0 + T ].
                                              dt
                                                             dx                 n
                                            0
                                                          n
                                                                   ∞
                           We know that x ∈ C ([t 0 ,t 0 + T ];R ),  ∈ L (t 0 ,t 0 + T ;R ), and V ∈
                                                             dt
                               n
                            1
                                                                    n
                                                  ∗
                           C (R ;R). It follows that V ∈ W  1,1 (t 0 ,t 0 + T ;R ), and applying, for exam-
                                                     ∗
                           ple, Lemma 2, we obtain that V is decreasing on [t 0 ,t 0 + T ]. Since T was
                                    ∗
                                                                      ∗
                           arbitrary, V is decreasing on [t 0 ,+∞). Moreover, V is bounded from below
                           on [t 0 ,+∞) since γ(x 0 ) ⊂   and V is continuous on the compact set  . There-
                           fore
                                                 lim V(x(τ;t 0 ,x 0 )) = k
                                                τ→+∞
                           for some k ∈ R.
                              Let y ∈  (x 0 ). There exists {τ i }⊂[t 0 ,+∞) such that τ i →+∞ and
                                                   x(τ i ;t 0 ,x 0 ) → y.
   172   173   174   175   176   177   178   179   180   181   182