Parameter estimation of unknown air exchange rates and effective mixing volumes from tracer gas measurement for complex multi-zone indoor air models.

The steepest descent and simulated annealing optimization techniques are used to simultaneously estimate the effective mixing volumes and air exchange rates of a large partition less building exhibiting heterogeneous spatial air flow conditions. The optimization is conducted using varying quantities and qualities of simulated tracer gas measurements. A simulated three-compartment system is numerically investigated to assess the performance of the parameter estimation methods. When simulated tracer gases are released in each zone, both techniques estimate actual parameter values within 10-35 percent. When tracers are released in selected zones, reasonable estimates were obtained for those zones in which a simulated gas was released, but significant errors are evident for the non-release zones