1: \begin{abstract}

2:   We present a method for sampling microscopic configurations

3:   of a physical system distributed according to a

4:   canonical (Boltzmann-Gibbs) measure, with a constraint holding in average.

5:   Assuming that the constraint can be controlled by the volume and/or

6:   the temperature of the system, and considering

7:   the control parameter as a dynamical variable, a sampling

8:   strategy based on a nonlinear stochastic process is

9:   proposed. Convergence results for this dynamics are proved using

10:   entropy estimates.

11:   As an application, we consider the computation of points along the

12:   Hugoniot curve, which are equilibrium states obtained after

13:   equilibration of a material heated and compressed by a shock wave.

14: \end{abstract}

15: