Jump to content

Category:Equilibrium constraints: Difference between revisions

From mintOC
m Initial setup of IMA paper text
 
m Correction
Line 4: Line 4:
<math>
<math>
\begin{array}{llcl}
\begin{array}{llcl}
  \displaystyle \min_{y_1, y_2, y_3} & \Phi(y_1, y_2, y_3)  \\[1.5ex]
  \displaystyle \min_{y_1, y_2, y_3} & & & \Phi(y_1, y_2, y_3)  \\[1.5ex]
  \mbox{s.t.} & 0 & = & F ( y_1, y_2, y_3), \\
  \mbox{s.t.} & 0 & = & F ( y_1, y_2, y_3), \\
  & 0 & \le & C ( y_1, y_2, y_3),  \\
  & 0 & \le & C ( y_1, y_2, y_3),  \\
  & 0 & \le & y_1 \perp y_2 \ge 0,
  & 0 & \le & (\mu - y_2)^T \; \phi( y_1, y_2), \;  y_2 \in Y(y_1), \; \forall \mu \in Y(y_1),
\end{array}  
\end{array}  
</math>
</math>
Line 13: Line 13:


where <math>Y(y_1)</math> is the feasible region for the variational inequality and given function <math>\phi(\cdot)</math>. Variational inequalities arise in many domains and are generally referred to as equilibrium constraints. The variables <math>y_1</math> and <math>y_2</math> may be controls or states.
where <math>Y(y_1)</math> is the feasible region for the variational inequality and given function <math>\phi(\cdot)</math>. Variational inequalities arise in many domains and are generally referred to as equilibrium constraints. The variables <math>y_1</math> and <math>y_2</math> may be controls or states.
[[:Category:Complementarity constraints | Complementarity constraints]] are a special case.


[[Category:Objective characterization]]
[[Category:Objective characterization]]

Revision as of 12:05, 20 November 2010

This category contains mathematical programs with equilibrium constraints (MPECs). An MPEC is an optimization problem constrained by a variational inequality, which takes for generic variables / functions y1,y2 the following general form:

miny1,y2,y3Φ(y1,y2,y3)s.t.0=F(y1,y2,y3),0≤C(y1,y2,y3),0≤(μ−y2)Tϕ(y1,y2),y2∈Y(y1),∀μ∈Y(y1),

where Y(y1) is the feasible region for the variational inequality and given function ϕ(⋅). Variational inequalities arise in many domains and are generally referred to as equilibrium constraints. The variables y1 and y2 may be controls or states.

Complementarity constraints are a special case.

This category currently contains no pages or media.