Page 143 - Complementarity and Variational Inequalities in Electronics
P. 143

134  Complementarity and Variational Inequalities in Electronics


                                          4
                           x 4 = 0 since x ∈ R .From Corollary 6 it follows that for each u ∈ R, problem
                                          +
                           VI(M, ,Fu) has at least one solution.
                           Example 65. Suppose that diodes D 1 and D 4 are ideal and for diodes D 2
                           and D 3 , consider the practical model with electrical superpotential


                                                            V 1 x  if  x ≥ 0
                                          (∀x ∈ R) : ϕ(x) =
                                                            V 2 x  if  x< 0
                           with V 2 < 0 <V 1 .Here

                                        4                    ∗
                                  (∀x ∈ R ) :  (x) =   R +  (x 1 ) + ϕ (−x 2 ) +   R + (x 3 ) + ϕ(x 4 )
                               ∗
                           and ϕ ≡   [V 2 ,V 1 ] . Thus D( ) = R + ×[V 2 ,V 1 ]× R + × R, D( ) ∞ = R + ×
                           {0}× R + × R, and D(  ∞ ) = R + ×{0}× R + × R. We check that

                                                           T
                                         D( ) ∞ ∩ ker{M + M }∩ K(M, ) ={0}.
                                                             T
                           Indeed, here x 2 = 0, and x ∈ ker{M + M } yields x 1 =−x 3 and thus x 1 =
                           x 3 = 0 since x 1 ≥ 0 and x 2 ≥ 0. Using then x ∈ K(M, ), we also get x 4 = 0.
                           From Corollary 6 it follows that for each u ∈ R, problem VI(M, ,Fu) has at
                           least one solution.

                              Let us now consider the second variational inequality of the model described
                                                                          3
                           in Section 3.4 of Chapter 3, that is, VI( , ,q) with q ∈ R ,   =   3 , and
                                                                                  R
                                                                                   +
                                        ⎛                                        ⎞
                                          R 1 K   R 1 (1 − α N )    R 1 (1 − α I )
                                        ⎝ R 1 K
                                                                                 ⎟
                                 
 =  1 ⎜       R 1 (1 − α N ) + R 2  R 1 (1 − α I ) − α I R 2 ⎠ .
                                     K
                                          R 1 K   R 1 (1 − α N )    R 1 (1 − α I )
                           The matrix 
 is of class P 0 . Indeed,
                                                           1
                                      1 (
) = R 1 > 0,  2 (
) =  (R 1 (1 − α N ) + R 2 )> 0,
                                                           K
                                              1
                                       3 (
) =  (R 1 (1 − α I )) > 0,  12 (
) = R 1 R 2 > 0,
                                             K
                           and

                                        13 (
) = 0,  23 (
) = R 1 R 2 > 0,  123 (
) = 0.
                           We have
                                           T
                                          z 
z = z 1 (
z) 1 + z 2 (
z) 2 + z 3 (
z) 3 ,
   138   139   140   141   142   143   144   145   146   147   148